Method and apparatus for inspecting a surface of an object comprising dissimilar materials

文档序号:958250 发布日期:2020-10-30 浏览:4次 中文

阅读说明:本技术 用于检查包括非相似材料的对象的表面的方法和装置 (Method and apparatus for inspecting a surface of an object comprising dissimilar materials ) 是由 珍-弗朗索瓦·布朗热 斯特凡·戈德尼 于 2019-01-25 设计创作,主要内容包括:本发明涉及一种用于测量待测量对象(300)的表面(400)的轮廓的方法(100),所述表面尤其包括由至少两种不同材料制成的区域,待测对象(300)形成多个基本上相同的对象的部分,多个对象还包括具有至少一个参考表面的至少一个参考对象(304、306),所述方法(100)包括以下步骤:根据第一参考表面的第一轮廓信号和第二参考表面的第二轮廓信号,确定(102)校正函数,所述第二参考表面被金属涂覆;获取(110)待测对象的表面的轮廓信号;以及将校正函数应用(116)于待测对象(300)的表面(400)的轮廓信号,以获得经校正的轮廓信号;所述轮廓信号是从干涉测量(104、112)获得的。本发明还涉及一种用于使用这种方法来测量对象的表面的轮廓的装置。(The invention relates to a method (100) for measuring a contour of a surface (400) of an object (300) to be measured, the surface comprising in particular an area made of at least two different materials, the object (300) to be measured forming part of a plurality of substantially identical objects, the plurality of objects further comprising at least one reference object (304, 306) having at least one reference surface, the method (100) comprising the steps of: determining (102) a correction function from a first profile signal of a first reference surface and a second profile signal of a second reference surface, the second reference surface being metal coated; acquiring (110) a contour signal of the surface of the object to be measured; and applying (116) a correction function to the profile signal of the surface (400) of the object (300) to be measured to obtain a corrected profile signal; the profile signal is obtained from an interferometric measurement (104, 112). The invention also relates to an apparatus for measuring the profile of a surface of an object using such a method.)

1. A method (100) for measuring a contour of a surface (400) of an object to be measured (300), the surface comprising in particular regions or structures made of at least two different materials, the object to be measured (300) forming part of a plurality of substantially identical objects, the plurality of objects further comprising at least one reference object (304, 306) having at least one reference surface, the method (100) comprising the steps of:

-determining (102) a correction function from a first profile signal from a first reference surface and a second profile signal from a second reference surface, the second reference surface being metallized;

-acquiring (110) a contour signal from a surface of the object to be measured; and

-applying (116) the correction function to a contour signal from a surface (400) of the object (300) to be measured to obtain a corrected contour signal;

wherein the profile signal is obtained from an interferometric measurement (104, 112).

2. The method (100) according to claim 1, wherein the steps of acquiring (110) contour signals and applying (116) the correction function are performed for a plurality of objects to be measured originating from the same production series.

3. The method (100) according to the preceding claim, wherein the step of determining (102) a correction function is performed using at least one reference object (304, 306) originating from the same production series as the object to be measured.

4. The method (100) according to one of claims 1 to 3, wherein the second reference surface corresponds to a metallized first reference surface of the same reference object.

5. The method (100) according to one of claims 1 to 3, wherein the first reference surface and the second reference surface are corresponding surfaces of two different reference objects (304, 306).

6. The method (100) according to one of the preceding claims, comprising a metallization step comprising depositing a conformal metal layer on a reference surface of a reference object.

7. The method (100) according to any one of the preceding claims, wherein the step (102) of determining the correction function is performed at several locations, called positions of interest, of the first and second reference surfaces, said several locations being located at the same position on the first and second reference surfaces, respectively.

8. The method (100) according to the preceding claim, wherein:

-the step (110) of acquiring contour signals is performed at several locations, called measurement locations, of the surface of the object to be measured, said several locations being located at the same location as the location of interest of the reference surface or within the same area of material as the location of interest; and

-performing the step (116) of applying the correction function on the measured positions of the surface (400) of the object (300) by taking into account the respective positions of interest.

9. The method (100) according to any one of the preceding claims, wherein the step of determining a correction function (102) comprises making a difference between the second profile signal and the first profile signal.

10. The method (100) according to any one of the preceding claims, wherein the step of applying the correction function (116) comprises summing a correction profile signal and the profile signal.

11. The method (100) according to one of the preceding claims, further comprising at least one of the following steps:

-a step of geometrically aligning a second contour signal from the second reference surface with respect to a first contour signal from the first reference surface;

-a step of geometrically aligning the profile signal from the surface of the object to be measured with respect to the first or second profile signal of the reference surface.

12. Method according to any of the preceding claims, characterized in that the method is carried out in order to measure the profile of the surface of an object to be measured, which object to be measured comprises a substrate, such as a wafer, or comprises components, such as chips produced on a substrate.

13. The method according to claim 12, characterized in that it comprises the steps of:

-determining (102) a correction function from at least one reference substrate,

-acquiring profile signals from another substrate under test.

14. An apparatus (600) for measuring a contour of a surface (400) of an object to be measured (300), the surface (400) comprising areas made of at least two different materials, the object to be measured (300) forming part of a plurality of substantially identical objects, the plurality of objects further comprising at least one reference object having at least one reference surface, the apparatus comprising:

an interferometric measuring device (610) arranged to acquire a first profile signal from a first reference surface, a second profile signal from a second metallized reference surface and a profile signal from a surface (400) of the object (300) to be measured, respectively,

-a processing module (612) configured to determine a correction function from the first and second profile signals and to apply the correction function to a profile signal from a surface of the object to be measured to obtain a corrected profile signal.

15. The device (600) according to the preceding claim, wherein the interferometric measuring device (610) comprises a full-field interferometric sensor.

16. The device (600) according to the preceding claim, wherein the interferometric sensor (610) comprises one of a phase-shifting interferometer and a vertical scanning interferometer.

Technical Field

The present invention relates to a method for measuring a contour or shape of an object surface, which can comprise a structure or pattern made of at least two different materials. The invention also relates to a device for measuring the surface of an object for carrying out such a method.

Background

The optical profilometry can determine the profile or topography of the face or surface of an object in order to determine the surface shape or imaging pattern, roughness, etc. present on the face and to obtain their height.

The method is based on measuring and studying interference signals obtained between reference and inspection optical radiations emitted from the same source, respectively sent to and reflected by the reference and inspected surfaces. By varying the optical paths of the reference or inspection radiation relative to each other, it is possible to determine from the analysis of the interference fringes the difference in the optical path length travelled by the reflected inspection optical radiation relative to the optical path length of the reference radiation and to infer therefrom the depth or height of the inspected surface at each measurement point, thereby detecting the relative heights of different patterns or structures, such as steps or grooves, present on the surface.

However, the waves reflected at the object surface experience a phase shift that depends on the physical properties of the materials, such as their complex refractive index and the thickness of the wafer stack (for transparent materials). Due to the above-mentioned topography, this phase shift upon reflection is added to the phase shift. When the pattern or structure present on the object surface is made of different materials, the phase shift upon reflection of each material is different. Therefore, the relative heights obtained between patterns made of different materials are inaccurate. For example, a step made of a certain material deposited on a substrate of another material may appear higher than it actually is, or conversely, the step may appear lower or even like a trench. Similarly, the two coplanar surfaces may appear to be highly different. Such a profile measurement cannot be used conveniently.

To overcome said drawback, it is known to calculate the theoretical reflection phase by exploiting a priori information items of the material or of the stack of layers of the material present. In this way, a theoretical reflectance model for correcting the measured values is constructed. However, this method is very cumbersome, and has limited utility, especially due to the uncertainty of the existing a priori information items. Furthermore, in order to correct the topography measurements in this way, it is also necessary to identify the different materials present in the micro-domain, which may present difficulties on multi-resolution or sub-resolution structures.

Disclosure of Invention

It is an object of the present invention to propose a method and a system for measuring the surface profile of an object which overcome these disadvantages.

It is another object of the invention to propose a method and system for measuring the surface profile of an object without the need to use a priori knowledge about the material properties and the geometry of the pattern.

Another object of the invention is to propose a method and a system for measuring the surface profile of an object, which make it possible to accurately measure the relative height of the patterns or structures present on the surface without requiring complex or expensive modifications of existing measuring means.

These objects are at least partly achieved by a method for measuring a contour of a surface of an object to be measured, the object surface especially comprising areas or structures made of at least two different materials, the object forming part of a plurality of substantially identical objects, the plurality of objects further comprising at least one reference object having at least one reference surface, the method comprising the steps of:

-determining a correction function from a first profile signal from a first reference surface and a second profile signal from a second reference surface, the second reference surface being metallized;

-acquiring a contour signal from a surface of an object to be measured; and

-applying a correction function to the profile signal from the surface of the object to be measured to obtain a corrected profile signal;

wherein the profile signal is obtained from an interferometric measurement.

The method according to the invention can advantageously be implemented to measure the profile of a surface of an object to be measured, said surface comprising areas or structures made of at least two different materials. Of course, the method can also be carried out to measure the profile of the surface of an object to be measured that is homogeneous or comprises only one material.

Within the framework of the invention, the profile of the surface corresponds to the relative height of a set of points on the surface distributed along one or two axes (X, Y) of a reference system or to the height relative to a reference point. The surface profile thus represents the shape of the surface and is intended in particular to determine the relative height of the structures present and/or the surface condition, for example a measure of roughness.

The contour signal can be a contour or correspond to a contour, or correspond to a variable representing a contour.

The correction function is a function applied to the contour signal, which transmits a corrected contour signal. The correction function can comprise, for example, a combination with a correction profile signal or a combination with a variable representing the correction profile.

Within the framework of the present invention, objects that are referred to as "substantially identical" are objects that are assumed to be identical, or theoretically or by design identical, but they may exhibit variations or differences, for example due to variability or uncertainty in their manufacturing process. These "substantially identical" objects can in particular constitute surfaces with regions having the same locations and the same material, but the contours of the surfaces differ at least in some locations. Thus, multiple objects can originate from one manufacturing series or batch, or from the same manufacturing process, among other things. For example, the object can be from a production series or batch of substrates (wafers, etc.) that include optical or electronic components.

The interferometry is performed by using optical interference or interferometric techniques known to those skilled in the art to function in the terahertz band. These interferometry measurements utilize interference between a reference beam or radiation and an inspection beam or radiation reflected by the surface to be inspected. These interferometric measurements make it possible to determine the altitude or height at different points of the surface. These combined elevations generate a profile signal representing the shape or (elevation) profile of the surface.

The second reference surface is metallized, i.e. it comprises a metal layer deposited prior to the interferometric measurement. The metal layer is uniform over the entire surface and has a thickness such that an incident measurement wave from the measurement system is totally reflected by the metallized surface without penetrating into the metal layer. Furthermore, the metal layer is fine enough to conform to the contour or shape of the covered surface, i.e. to accurately reproduce the contour or shape of the covered surface on its surface.

Preferably, the first reference surface is not metallized, or at least it is not modified, i.e. it has not been subjected to deposition of a metal layer prior to the interferometric measurement. However, of course, if the first surface forms part of the object to be measured, the metalized area can be included.

The method according to the invention therefore proposes: at least one reference object is selected from a plurality of objects forming part of said objects, e.g. of the same production series, for determining a correction function or contour, in particular at each location of interest. The correction function or profile is then applied to all profile measurements made on other objects to be measured, for example from this production series.

Thus, the method according to the invention enables a correction of the profile measurement in order to obtain a corrected profile of all objects, provided that these objects are sufficiently identical to each other and to the reference object, at least in terms of the complex reflectivity of the regions or structures present. The corrected contours represent the correct relative heights of different regions or structures on the object surface.

Then, by using the reference measurement, the effect of the phase shift upon reflection on the surface, which depends on the physical properties of the different types of materials present on the object surface, on the measured height of the region or pattern can be eliminated. Then, only the reference object is not available due to the metallization of its surface.

It should be noted that "wear" of metallized reference objects is not very inconvenient, since during the production of substrates (wafers, etc.) containing optical or electronic components, it is common to metallize some of these substrates in order to perform measurements for quality control by sampling. In this regard, the present invention allows the number of objects that are metallized and thus damaged to be minimized by performing non-destructive testing on all or a portion of the objects produced. Furthermore, the invention makes it possible to test objects that are actually produced and available, and thus not destroyed, in contrast to destructive testing methods by sampling.

The method according to the invention can therefore advantageously be used in a production process of test objects. Thus, the quality of the process can be monitored and process parameters can be adjusted in the event of deviations from the desired profile. As previously described, only a small number of objects (e.g., one per manufacturing lot or series) are discarded in order to construct the reference object.

Thus, no a priori knowledge about the nature or thickness of the surface material or about the geometry of the layer or pattern is required for the determining step. The method according to the invention can thus be implemented by means of the present measuring device in combination with a device for depositing a metal layer known to the person skilled in the art. No special devices need to be added.

Advantageously, the steps of acquiring contour signals and applying correction functions can be performed for a plurality of objects to be measured originating from the same production series.

Similarly, the step of determining the correction function can be performed using at least one reference object originating from the objects to be measured of the same production series.

Thus, the correction function can be applied to each of the profile measurements of all objects under test from, for example, a plurality of objects originating from the same production series. Thus, the variability of the manufacturing process from one object to another or over time can be known. In addition, only one object or sample in the series under examination becomes unusable after metallization of its surface.

Advantageously, the second reference surface can correspond to a metallized (or metallized) first reference surface of the same reference object.

In this case, a single reference object can be selected and the reference profile signal acquired before and after metallization of the reference surface.

Alternatively, the first reference surface and the second reference surface can be corresponding surfaces of two different reference objects.

In this embodiment, the correction function is determined by two reference objects, only one of which has a metallized surface.

It is also possible to generate the correction function by a plurality of measurements made respectively on a plurality of first reference surfaces and a plurality of second reference surfaces belonging to a plurality of identical or different reference objects. Thus, a correction function corresponding to an average of corrections determined over a plurality of reference objects can be obtained.

The method according to the invention can comprise a metallization step comprising depositing a conformal metal layer on the reference surface of the reference object.

The metal layer must be thick enough to make it opaque to light, but thin enough not to change the shape of the surface and thus to conform closely. In practice, it can consist of a metal layer of the order of a few tens of nanometers (for example 40nm of tantalum).

Said metallization step can be carried out by implementing techniques known to those skilled in the art, for example of the CVD (chemical vapour deposition) type or of the PVD (physical vapour deposition) type. Among the PVD-type techniques, vacuum deposition and cathode spraying (sputtering) may be mentioned in particular.

Advantageously, the step of determining the correction function can be performed at several locations, called positions of interest, of the first reference surface and the second reference surface, said several locations being located at the same position on the first reference surface and the second reference surface, respectively. In particular, the positions of interest can be selected such that each position of interest of the first reference surface corresponds to one and the same position of interest of the second reference surface.

The location of interest corresponds to a point of interest in the profile signal (or profiles) of the first reference surface and the second reference surface. Thus, the correction function can be determined from the values of the contour signal at these points of interest from the first reference surface and the second reference surface. This therefore constitutes a correction profile which makes it possible to correct systematic errors in the profile signal.

The step of acquiring the contour signal can be performed at several locations, called measurement locations, of the surface of the object to be measured, located in the same region as the location of interest of the reference surface, or in a region having the same characteristics as the characteristics of the location of interest (i.e. made of the same material, and preferably of the same geometry). The step of applying a correction function can be performed on said measured positions of the object surface by taking into account the respective positions of interest.

As previously mentioned, the measurement positions correspond to measurement points in the profile signal from the surface of the object to be measured. Therefore, at the measurement point corresponding to the measurement position, a correction function (or a correction profile) can be applied to the profile signal from the surface of the object to be measured.

According to an embodiment, the step of determining the correction function can comprise a difference (or subtraction) between the second profile signal and the first profile signal, the two profile signals being derived from a reference measurement made on a reference surface. Said step enables the generation of a correction profile signal.

Such a subtraction or subtraction step is not very complex to implement, requiring very little resources and very short processing times.

According to an embodiment, the step of applying a correction function can comprise correcting the profile signal and a summation of the profile signals, optionally algebraically.

Such a summing step is not very complex to implement, requiring few resources and very short processing times.

According to an embodiment, the method according to the invention can comprise the step of geometrically aligning the second contour signal from the second reference surface with respect to the first contour signal from the first reference surface. The steps can be performed during the determination of the correction function.

According to an embodiment, the method according to the invention can comprise the step of geometrically aligning the profile signal from the surface of the object to be measured with respect to the first or second profile signal of the reference surface. This step can be performed during the application of the correction function to ensure that the correction function is correctly "positioned" relative to the surface of the object to be measured.

These alignment steps can include applying a spatial transformation function, such as translation, rotation and/or amplification, to at least one of the contour signals to cause the contour signals to overlap optimally in space. Thus, errors of the acquisition means, such as any positioning errors of the object, can be corrected.

According to an embodiment, the method according to the invention can be implemented in order to measure the profile of the surface of an object to be measured comprising a substrate or an element produced on a substrate.

The substrate can be a substrate for producing integrated optical components, for example of the glass type. The substrate can also be a semiconductor substrate, for example made of silicon. In general terms, the substrate can include, but is not limited to, a wafer, a panel, a wafer carrier, a reconstituted wafer, a wafer on a frame, and the like.

The elements produced on the substrate can comprise, for example, chips, or electronic or optical circuits and/or structures such as patterns, gratings, trenches, vias or connecting elements.

According to an embodiment, the method according to the invention can comprise the following steps:

-determining a correction function from at least one reference substrate,

-acquiring a profile signal from another substrate under test.

According to another aspect of the invention, there is provided an apparatus for measuring the profile of a surface of an object, the surface comprising regions made of at least two different materials, the object forming part of a plurality of substantially identical objects originating from a same production series of the objects, the plurality of objects comprising at least one reference object having at least one reference surface, the apparatus comprising:

an interferometric measuring device arranged to acquire a first profile signal from the first reference surface, a second profile signal from the second metallized reference surface and a profile signal from the surface of the object to be measured, respectively.

A processing module configured for determining a correction function from the first and second profile signals and applying the correction function to the profile signal from the surface of the object to be measured to obtain a corrected profile signal.

The configuration of the processing module can be performed electronically and/or computationally, in particular with instructions executable by a processor or an electronic chip.

The processing module can be incorporated into the interferometric device or external to the measurement device and linked to the measurement device in a wired or wireless manner.

According to an advantageous embodiment, the interferometric measuring device can comprise a full-field interferometric sensor. According to non-limiting examples, the interferometric sensor can comprise a phase-shifting interferometer (PSI) or a low coherence Vertical Scanning Interferometer (VSI).

According to an embodiment, the apparatus of the invention can further comprise means for depositing a metal layer to be deposited on the second reference surface before acquiring the interference signal from the second reference surface.

Description of the figures and embodiments

Further advantages and features of the invention will become apparent from reading the detailed description of non-limiting embodiments and examples and from the accompanying drawings, in which:

FIG. 1 is a schematic view of a non-limiting embodiment of a measurement method according to the invention;

figure 2 shows an example of a surface to be measured of an object;

figures 3a and 3b show diagrams of a metallized wafer and a non-metallized wafer, respectively, each wafer comprising several chips;

4 a-4 c are illustrations of non-limiting examples of measuring a surface profile of an object, such as a wafer, using the present invention;

Figures 5a and 5b represent non-limiting examples of measurements obtained with the present invention; and-figure 6 is a schematic view of a non-limiting embodiment of a measuring system according to the present invention.

It should be understood that the embodiments to be described hereinafter are in no way limiting. In particular, if the selection of features described hereinafter is sufficient to confer technical advantages or to distinguish the invention from the prior art, it is particularly envisaged that the variants of the invention comprise only a selection of such features, separately from the other features described. Said selection comprises at least one feature, preferably functional, without structural details or with only partial structural details, provided that only said partial structural details are sufficient to confer technical advantages or to distinguish the invention from the prior art.

In particular, all the variants and all the embodiments described can be combined together, provided that such a combination is not objected from a technical point of view.

In the drawings, elements common to several figures retain the same reference numeral.

FIG. 1 is a schematic diagram of a non-limiting embodiment of a method for measuring a surface profile of an object according to the present invention.

An object whose one or more surfaces are to be inspected or measured to determine its contour forms part of a plurality of substantially identical objects. In the embodiment of the invention illustrated with reference to the drawings, the plurality of objects to be measured is constituted by a plurality of semiconductor substrates, for example in the form of wafers comprising electronic circuits, chips or other semiconductor components, without limitation. These objects (or wafers) can, for example, form part of the same production series, and are therefore assumed to be identical (or substantially identical) to provide or employ manufacturing variations.

The plurality of objects further includes at least one reference object having at least one reference surface. The reference object or objects can be, for example, one or more wafers originating from the same series as the object to be measured or from a reference series.

The method 100 includes a calibration step 102 performed using the first reference surface and the second reference surface.

The calibration phase 102 comprises a measurement step 104 of the interference signal at several measurement positions on two reference surfaces. The reference surface is a surface of a reference object.

The interferometry is performed by a measurement system comprising a full field optical interferometric sensor such as that shown in fig. 6. The interferometric sensor includes an imaging system that corresponds measurement locations on the surface to pixels of the sensor field of view, thereby delivering an interference signal for each pixel or measurement point of the sensor.

The second reference surface is metallized, i.e. it comprises a metal layer deposited before the measurement step 102. The metallization of the surface comprises in particular the deposition of a uniform and homogeneous metal layer over the entire surface. The thickness of the metal layer must be adjusted so that the incident measurement wave from the measurement system is totally reflected by the metallized surface and does not reach the surface of the object and is thus not affected by the material properties or material layer of the object. Furthermore, the metal layer must be thin enough not to alter the relief or contour of the object surface.

In practice, the metal layer can be produced, for example, by vacuum deposition techniques, for example, by Physical Vapor Deposition (PVD). The metal layer can consist of a tantalum layer of about 40 nm.

The method further includes calculating a profile signal for each of the reference surfaces

Figure BDA0002666334600000091

Step 106, the contour signal is hereinafter simply referred to as contour

Figure BDA0002666334600000092

These profiles are obtained from interference signals measured at measurement points (x, y) respectively corresponding to measurement positions on a reference surfaceFor this purpose, each interference signal is obtained by using a known phase extraction algorithm, for example a PSI (phase-shift interferometry) algorithmPhase. The above-mentionedThe phase depends on the phase shift ψ at the reflection of the measurement wave and the topographical contribution from the shape of the surface being measured. Then use

Figure BDA0002666334600000101

Determining the profile by phaseSaid profileMeasured heights corresponding to the surfaces at the respective measurement positions:

Figure BDA0002666334600000104

where h is the set of actual heights or altitudes of the surface, and λ is the measurement wavelength. The value psi of the phase shift depends on the material of the surface at the measurement location in question.

In the following, the profile of the first reference surface will be referred to as first profileThe profile of the second reference surface will be referred to as the second profile

Figure BDA0002666334600000106

The first and second contours correspond to a reference acquisition. The phase shift upon reflection on the second metallized reference surface is constant over the entire surface, # 2(x,y)=ψ2And a phase shift on reflection on the first reference surface ψ1(x, y) depends on the position on the surface corresponding to the measurement point (x, y) of the interference signal being measured.

During step 108, a first profile is determined from each measurement point (x, y), in this case referred to as a point of interestAnd a second profileTo determine a correction function C (x, y) or a correction profile C (x, y). In an implemented embodiment, the correction profile C (x, y), also called correction map, is obtained by determining the difference between the first profile and the second profile at each point of interest (x, y):

Figure BDA0002666334600000109

to this end, the step 108 of determining the correction function can also comprise a geometric alignment step prior to the subtraction, for aligning the first and second contours with respect to one another, for example in translation, rotation and/or magnification, so that the structures present on these two contours optimally overlap in the plane (x, y).

The phase shift term in reflection is evident from the previous formula:

Figure BDA00026663346000001010

and

Figure BDA00026663346000001011

wherein psii(x, y) is the phase shift upon reflection on each of the two reference surfaces.

The step 108 of determining the correction function C (x, y) is performed at several points of interest corresponding to several positions of interest of the first reference surface and the second reference surface, each position of interest of the first reference surface corresponding to the same position of interest of the second reference surface.

The calibration phase 104 ends in step 108.

The plurality of objects can include a single reference object, such as a wafer, having reference surfaces measured before and after metallization thereof to obtain the correction profile. In this case, the second reference surface corresponds to the first reference surface after metallization.

Alternatively, the plurality of objects can comprise two different reference objects, each of the reference objects having a reference surface, the two reference surfaces being corresponding surfaces (i.e. the same surfaces) of the two reference objects, one of the reference surfaces being metallized and the other being unmetallized. In this case, these objects can be two wafers originating from the same series, or two different objects on a reference wafer whose surface has been partially metallized.

The method 100 according to the proposed embodiment further comprises a step 110 of obtaining a surface profile of the object to be measured from a plurality of objects.

The step of acquiring a profile 110 comprises a step 112 of acquiring interference signals at a plurality of measurement positions from the surface of the object to be measured. These measured positions of the object surface must substantially correspond to the position of interest of the reference surface on which the calibration step 102 is performed, or more generally to the vicinity of the position of interest of the reference surface, wherein it can be assumed that the correction profile C (x, y) is known (for example because the materials present have similar characteristics).

In step 114, the measured profile is calculated from the interference signal

Figure BDA0002666334600000111

The calculation is preferably the same as explained above for the reference measurement. A measurement profile of the surface to be inspected is thus obtained.

The stage 110 of acquiring the profile to be measured ends in step 114.

Then the application of the correction function (or correction profile) C (x, y) to the measured profile is performed for all measurement points or positions

Figure BDA0002666334600000112

Step 116.

During said step 116, a correction profile is added to the measured profile to obtain a profile signal or a corrected profile

Figure BDA0002666334600000122

When it is assumed that the phase shift effect on reflection is equivalent between the non-metallised reference surface and the surface to be measured, the expression for the latter can be re-expressed in the form:

the corrected profileRepresenting the shape of the measured surface, wherein the phase shift ψ in reflection depending on the different material pairs on the measured surface has been eliminated1(x, y) contribution. When considering the relative height, i.e. the difference in height between different measuring points or positions on the measured surface, it should be noted that the relative corrected heightActually corresponds to the actual physical relative height Δ h:

independent of the material present at the measurement location on the measured surface.

The step 114 of applying a correction function or profile C (x, y) is performed for each measurement point or position of the surface of the object to be measured corresponding to the same position of interest on the reference surface. If there is no overlap between the measured and the point or location of interest, then of course their spatial relationship is taken into account.

In order to enable an optimal application of the correction profile C (x, y), the step 114 of applying a correction function or profile C (x, y) can also comprise a step of geometric alignment before applying said correction function, in order to align the measured profile, for example in translation, rotation and/or magnificationAligned with the first reference profile and/or the second reference profile so that these profiles optimally overlap in the plane (x, y).

Fig. 2 shows an example of a surface of an object to be measured, said surface comprising a (central) convex portion. Two positions of interest, below (a) and above (B) the raised portion, respectively, are also shown to determine the correction profile C (x, y). In fact, without applying the method according to the invention, the relative height of these two positions measured by the interferometric method will not correspond to the physical relative height.

Fig. 3a and 3b schematically show wafers 308, 310 comprising several chips 300, 302, respectively. The two wafers originate from the same production series or the same batch. Wafer 310 represents a wafer to be tested or measured.

The wafer 308 in fig. 3a is used as a reference wafer 308 for determining a correction profile, for example by performing measurements on a first profile of the wafer surface before metallization and a second profile of the wafer surface after metallization according to the calibration step 102 of the method 100. The correction profile can be calculated for all chips 300 of the reference wafer 308. Alternatively, it is possible to define the reference chip 304, for example at the center of the reference wafer 308, and to calculate the correction profile only for said reference chip 304.

The corrected profile can then be used to inspect the chips 302 of the wafer 310 to be tested, for example, according to step 110 of the method 100 in fig. 1. Thereby obtaining a corrected profile for each of the dies 302 of the wafer 310 under testIf only the correction profile for the reference chip 304 is determined, the same correction function is applied in order to obtain the correction profile for each of the chips 302 of the wafer 310 to be tested by spatially compensating the correction profile

Figure BDA0002666334600000132

Fig. 4a to 4c show examples of profile measurements performed according to a method of the present invention, such as the method 100 in fig. 1. In particular, the surface 400 shown in fig. 4a is the surface to be inspected of a chip or a part of a chip having a pattern made of different materials, said chip being referred to as test chip hereinafter.

Fig. 4a shows an example of uncorrected measurements performed on a test chip, such as the chip or portion of chip 302 of the non-metalized wafer in fig. 3b, for example, according to steps 112 and 114 of method 100.

Fig. 4b shows an example of a corrected measurement performed on the same test chip, for example according to steps 112, 114 and 116 of the method 100.

By comparison with the corrected measurements, fig. 4c shows an example of measurements performed on similar chips or parts of chips of a metallized wafer, such as chip 300 of the metallized wafer in fig. 3 a. The measurement can be performed, for example, according to steps 112 and 114 of method 100.

In fig. 4a to 4c, the first row shows the complete topography of the chip or of a part of the chip over the entire field of the interferometric sensor, respectively. The chips each present elements or blocks 402, 404, 406 of different heights made of different materials.

The second row in fig. 4a to 4c represents the profile according to the lines 408, 410, 412 inserted into the complete topography, respectively. The y-axis corresponds to the measured heights of the different blocks along the lines 408, 410, 412. Note that the relative height of the block is different between the uncorrected measurement of the non-metallized test chip (fig. 4a) and the measurement of the metallized chip (fig. 4c), while the relative height of the block is similar for the corrected measurement (fig. 4b) and the measurement on the metallized chip (fig. 4 c).

For example, for three measurements, the relative height between the fourth and fifth blocks can be considered. The relative height between block 414 and block 416 is about 60nm for an uncorrected test chip (fig. 4a), and about 40nm for a metalized chip (fig. 4c) at the same location. For the corrected measurements made on the test chip (fig. 4b), the same relative height of about 40nm between blocks 414 "and 416" was effectively found. It is also noted that the overall configuration of the block is in fact reproduced for the corrected measurement (fig. 4b) with respect to the measurement made on the metallized chip (fig. 4 c).

The third row in fig. 4a to 4c represents details of the framing of the first line of the measurement field for each of the measurements. This detail makes it possible in particular to understand the difference in appearance between the uncorrected and the corrected measurements carried out on the test chip (fig. 4a and 4b), and the similarity between the corrected measurements and the measurements carried out on the metallized chip (fig. 4a and 4 c).

Fig. 5a and 5b show by way of example the difference between the corrected profile of the test chip shown in fig. 4b and the profile of the metallized chip shown in fig. 4 c. This difference makes it possible to show the accuracy of the corrected topography with respect to the topography on the metallization layer, which is considered as "ground truth". Fig. 5a shows this difference over the entire measurement field, and fig. 5b shows this difference along the line 510 that is inserted in fig. 5 a.

In fig. 5a, the difference between the topography measured for the test chip and the topography measured for the metallization chip is shown in grayscale. Note that the magnitude of the difference does not exceed a few nanometers across the measurement field. This residue is higher at the edges of the block, which may be due to residual positioning errors between reference chips or between the correction profile and the test chip.

FIG. 6 is a schematic diagram of a non-limiting embodiment of a measurement system according to the present invention.

The system 600 shown in fig. 6 comprises a light source 602, for example based on a light emitting diode or a halogen source. The light source 602 produces a light beam 604 in the visible wavelength and/or near infrared range. The beam 604 is directed by a cube or beamsplitter 608 to the full field interferometer 606.

In the full field interferometer 606, the light beam 604 is split into a reference beam that illuminates the reference mirror and a measurement beam that illuminates the surface 400 of the object to be examined. The surface 400 to be inspected can be, for example, the surface shown in fig. 4 a. The light reflected respectively by the surface 400 and by the reference mirror is redirected towards a sensor array 610, for example of the CCD or CMOS type.

The system 600 comprises optics and lenses comprising an imaging objective arranged to image the surface 400 on the sensor array 610. Interference fringes caused by interference between the measurement beam and the reference beam are also visible when the optical path difference between the measurement beam and the reference beam is less than the coherence length of the light source 602.

Different types of full field interferometers 606 can be used within the scope of the invention, such as phase-shifting interferometers or vertical scanning interferometers. These interferometers are well known to those skilled in the art and will therefore not be described in detail here.

The system 600 further comprises means 614 for depositing a metal layer on the second reference surface, the metallization being performed before acquiring the interference signal from the second reference surface. The deposition apparatus 614 can be, for example, a physical vapor deposition apparatus (PVD). Deposition apparatus for metal layers are well known to those skilled in the art and will not be described in detail herein.

The system 600 further comprises an electronic/computing module 612, for example a processor or an electronic chip or a computer, linked to the sensor array 610 and configured to implement all the steps of the method according to the invention, as for example the steps 104 to 116 of the method 100 described above with reference to fig. 1.

Of course, the invention is not limited to the examples just described, and many modifications can be made to these examples without departing from the scope of the invention.

20页详细技术资料下载
上一篇:一种医用注射器针头装配设备
下一篇:提取多个移动物体的统计样本的系统及方法

网友询问留言

已有0条留言

还没有人留言评论。精彩留言会获得点赞!

精彩留言,会给你点赞!