EP3097048A1

EP3097048A1 - Method for obtaining by creeping at least one structure that approximates a desired structure

Info

Publication number: EP3097048A1
Application number: EP15700594.3A
Authority: EP
Inventors: Vincent REBOUD; Stéphan Landis; Etienne Rognin
Original assignee: Commissariat a lEnergie Atomique et aux Energies Alternatives CEA
Current assignee: Commissariat a lEnergie Atomique et aux Energies Alternatives CEA
Priority date: 2014-01-20
Filing date: 2015-01-19
Publication date: 2016-11-30
Also published as: US10739583B2; US20160334620A1; JP2017509161A; FR3016700A1; WO2015107202A1; JP6800753B2

Abstract

The invention relates to a method for determining at least one creeping parameter that makes it possible to obtain a structure that approximates a desired structure (S1) by creeping of an initial structure (S2) that is different from the desired structure (S1), the initial structure (S2) being made up of at least one pattern formed in a heat-deformable layer arranged on a substrate, characterised in that the heat-deformable layer forms a residual layer which surrounds each pattern, from which each pattern extends such that each pattern has only one interface with the surrounding medium, and in that the method includes at least the following steps: a step of predicting (1104) the variation over time of the geometry of the initial structure (200) (S2) subjected to creeping, such as to obtain a plurality of geometries of predicted structures (S3) each associated with creeping parameters including at least one creeping time and one creeping temperature; a step of calculating (1105) correlation values of the geometry of each predicted structure (S3) relative to the desired structure (S1); a step of identifying (1106, 1108) creeping parameters that make it possible to obtain the predicted structure (S3) while offering the highest correlation value.

Description

Process for obtaining by creep at least one structure approximating a desired structure

TECHNICAL FIELD OF THE INVENTION

The present invention relates to a process for obtaining three-dimensional patterns of micrometric or nanometric size having complex profiles. It relates more particularly to the simultaneous realization, at the level of a wafer (slice or plate in French), matrices of patterns with complex profiles. It finds advantageous but not limitative application the formation of optical lenses of micrometric or nanometric size and in particular aspherical optical lenses.

STATE OF THE ART

To obtain patterns having complex profiles, a first known solution is to approximate any complex profile by a profile formed of a multitude of steps, that is to say formed by a stack of discrete volumes. The techniques usually used require either several successive stages of lithography or long sequential lithography techniques, complex and expensive. Moreover, in practice, the number of height levels that can be achieved remains limited, typically a few tens of levels for the most complex forms.

It is also known since the eighties, to make flow, that is to say to partially melt, patterns previously obtained by conventional lithography techniques (ultraviolet lithography or e-beam electron beam ) to achieve three-dimensional patterns in a single wafer scale creep operation.

For example, optical micro-lens matrices have been obtained in a single creep operation of a resin layer in which UV-patterned units of simple shape, typically rectangular section pads arranged on a substrate, had previously been formed by UV lithography. . This type of process was for example proposed in 1988 by ZD Popovic and his co-authors in an article published in Applied Optics, No. 27, pages 1281-1284, under the title "Technique for monolithic manufacture of microlens arrays" (1988). ).

This type of method, however, does not allow to obtain after creep any arbitrary complex shape. For example, while it is known to obtain matrices of micro lenses of hemispherical shape, obtaining aspherical micro lens matrices whose optical properties are much better is not simply achievable by this technique. For example, spherical lenses, unlike aspherical lenses, induce optical aberrations because the rays passing through the center of the lens do not converge exactly at the same point as those passing by the edges, which causes a blur at large openings and an enlargement of the focus spot.

To improve the approximation of the desired ideal shape, it is known to use, inter alia, laser exposure techniques or laser ablation. In this case, the process is always very long since shaping shapes must be done individually at each pattern.

Furthermore, it has been proposed to expose, by several stages of UV or e-beam lithography, shapes initially having more than two levels in order to obtain, after creep, a final shape which better approximates the desired final shape. However, it is currently not possible to predict, from the desired final form, the initial shape that must be flirted due to the complexity of the mechanisms involved during creep. Similarly, starting from a given initial shape, it is also not possible to accurately predict the final shape that will be obtained after creep of this initial shape. It is therefore necessary to proceed to many tests which represents a cost often prohibitive mainly due to the cost of making masks and molds.

There is therefore a need to provide a solution to obtain complex three-dimensional shapes closer to the desired ideal shapes while maintaining a limited cost.

Other objects, features and advantages of the present invention will become apparent from the following description and accompanying drawings. It is understood that other benefits may be incorporated.

SUMMARY OF THE INVENTION

According to one embodiment, the invention relates to a method for determining at least one creep parameter for obtaining a structure approximating a desired structure by creep of an initial structure different from the desired structure. The initial structure is made of at least one pattern formed in a thermo-deformable layer disposed on a substrate. The thermo-deformable layer forms a residual layer surrounding each pattern and from which extends each pattern so that each pattern has an interface only with the surrounding medium. The method comprises at least the following steps implemented by at least one microprocessor:

a step of predicting the evolution over time of the geometry of the initial structure subjected to creep, so as to obtain a plurality of predicted structure geometries each associated with creep parameters comprising at least one creep time and a creep temperature;

a step of calculating correlation values of the geometry of each predicted structure with respect to the desired structure;

a step of identifying the creep parameters making it possible to obtain the predicted structure offering the highest correlation value.

Thus, the invention makes it possible to precisely determine the parameters that will make it possible to obtain shapes with arbitrary and possibly complex profiles by creep, thanks in particular to the presence of a residual layer at the base of each of the patterns of the initial structure. and deterministically controlling the profile of the initial patterns and / or the thermal annealing conditions to obtain a very good approximation of the desired structure.

The identified creep parameters are intended to be applied to a creep stage of the initial structure. Thus, the invention makes it possible to respond to each of the following problems:

starting from a given initial structure, the invention makes it possible to determine the optimum creep conditions in order to obtain the final structure closest to the desired structure;

by imposing creep conditions (typically temperature and creep time), the invention makes it possible to determine what is the optimal initial structure for obtaining, after a creep under these imposed conditions of temperature and time, the structure closest to the desired structure;

the invention also makes it possible to define both the initial structure and the optimum creep conditions so as to obtain the structure closest to the desired structure.

Typically, if in practice one can not exceed a certain number of different height levels, in particular because the lithography equipment available does not allow it, the invention makes it possible to define the shape of the initial structure and the temperature of the creep. The invention thus makes it possible to considerably reduce the costs of obtaining complex structures, for example structures with three-dimensional patterns.

For example, in the context of the invention, in a perfectly surprising manner, it has been found that in order to obtain domes, the most desirable initial structure is often a pyramid which is then fired and not an initial structure. made of stacked cubes approximating a dome.

The invention makes it possible to obtain various profiles and in particular profiles having curves or profiles whose tangents evolve continuously while traversing the profile. It makes it possible to obtain angles, albeit a little rounded due to creep, but does not make it possible to obtain truly sharp edges, that is to say profiles whose tangents do not evolve in a continuous manner.

Optionally, this method according to the invention may furthermore have the following optional steps and characteristics:

Advantageously, the residual layer completely covers the substrate.

More generally, the residual layer surrounds each pattern so that all the lines or edges taken on the contour of each pattern are only in contact with an environment in which the initial structure is located and are therefore not in contact with the surface. substrate on which the thermo-deformable layer rests. Thus, the initial structure has no line or edge taken on the contour of the layer thermo-deformable and which is in contact with both the ambient environment and the substrate on which the thermo-deformable layer rests.

Preferably, the initial structure has only double points also called double lines or double interfaces. A double point is a point in the envelope of a structure (also referred to as a pattern) that is in contact with a single medium. Thus, all the points of the envelope of the layer forming the initial structure are in contact with only one material: the surrounding air. This surface is in no way in contact with another material such as the substrate on which the initial structure rests.

According to another embodiment, the initial structure has triple points or triple lines, in contact with both an ambient medium and the substrate on which the layer of thermo-deformable material rests, the triple points being distant from each pattern. a distance at least equal to Dmini, with

Dmini = 2 Max (hr, hd, À) and preferably Dmini = 5 Max (hr, hd,, λ) with hr = average thickness of the residual layer between the pattern closest to the triple point and the triple point; with hd = height of the pattern closest to the triple point. The height of a pattern is measured between the highest point of the pattern and the base of the pattern. The basis of the pattern is at the free surface of the residual layer. The height thus corresponds, for a residual layer of constant thickness disposed on a substrate whose upper surface is flat, to the difference between the height of the pattern measured from the upper face of the substrate and the residual thickness measured from this same upper face of the substrate and; with λ = the distance that separates the two motifs that are closest to the triple point.

Thus, whether or not the thermo-deformable layer covers the entire substrate, none of the patterns has an interface with the substrate or with a layer underlying it.

Advantageously, apart from the patterns, the residual layer has a substantially constant thickness. That is to say a thickness that does not vary by more than 10% relative to the average thickness of the residual layer located under the patterns. This non-limiting embodiment facilitates the implementation of the invention.

Preferably, the thermo-deformable layer forms a residual layer extending substantially in a plane, each pattern extending from this plane so that each pattern has only an interface with the surrounding medium. The patterns are hollow or projecting patterns.

Advantageously, the steps of prediction, calculation of correlation values and identification of the creep parameters are repeated with a plurality of initial structures whose geometries are different from each other.

Advantageously, one identifies, among the plurality of initial structures, the initial structure making it possible to obtain the highest correlation value.

Advantageously, the steps of prediction, calculation of correlation values and identification of creep parameters are repeated with a plurality of initial structures whose geometries are different from each other only if the highest correlation value for a structure considered is less than a predetermined correlation threshold.

According to an advantageous embodiment, a maximum creep temperature is imposed.

According to one embodiment, a maximum creep time is imposed.

Advantageously, the step of predicting the evolution over time of the geometry of the initial structure subjected to creep depends on the thickness of the residual layer.

Advantageously, at least two creep modes are defined: one starting from a thick residual layer where the capillarity phenomena are predominant, the other starting from a thin layer where the nonlinear lubrication phenomena dominate.

Advantageously, the initial structure is formed at least in part by superimposed cubes or blocks.

According to an advantageous embodiment, the initial structure has a triangular section in a section perpendicular to the plane of the substrate.

According to an advantageous embodiment, the mold of triangular section patterns is obtained by preferential chemical etching according to the crystallographic planes of a semiconductor material comprising silicon.

According to an advantageous embodiment, the final structure comprises one or more aspherical lenses or one or more Fresnel lenses.

According to an advantageous embodiment, the predicted structure (S3) offering the highest correlation value forms a topography making it possible to reinforce the light emission in the LED devices.

Advantageously, the steps of prediction, calculation of correlation values and identification of the creep parameters of the method are performed by at least one microprocessor. According to an advantageous embodiment, the predicted structure (S3) offering the highest correlation value is a structure for a tool for manufacturing a microelectronic device or a structure of a microelectronic device.

By microelectronic device is meant any type of device made with microelectronics means. These devices include, in addition, devices for purely electronic purposes, micromechanical or electromechanical devices (MEMS, NEMS, etc.) as well as optical or optoelectronic devices (MOEMS, etc.).

Thus, it appears that the method according to the invention based on a simulation from ^a very specific structure, that is to say without "triple point" near the grounds, performs technical functions specific to engineering modern by predicting, in a concrete manner, creep conditions to obtain a shape very close to that desired.

The ^invention enables to orient the design process of the initial shape and the manufacturing process (including settings of ^"PC and creep time) with an accuracy such that one can estimate the chances of a successful creep operation even before carrying out this operation concretely.In addition, the method according to the invention makes it possible to determine the optimal initial shape that will be suitable for flowing to obtain the desired structure.

With existing techniques, identifying optimal creep conditions and identifying the initial structure to be predicted or choosing, in a set of creep parameters and initial structures, those that offer the best chance of success, would be costly and would take a lot of time.

In addition, since ^it determines the optimal initial form and the conditions of creep, the present invention is therefore a concrete and practical tool for expert creep. The method according to the invention is implemented by a computer comprising at least one microprocessor.

In another aspect, the present invention relates to a computer program product or to a computer-readable non-transitory medium, comprising instructions, which when performed by at least one processor, execute the steps of the method according to the invention mentioned above. These steps are at least the steps of prediction, calculation of correlation values and identification of creep parameters. According to another embodiment, the invention relates to a method for obtaining at least one structure approximating a desired structure from at least one initial structure, different from the desired structure, the initial structure being made of at least one pattern formed in a thermo-deformable layer disposed on a substrate, characterized in that the thermo-deformable layer forms a residual layer surrounding and from which each pattern extends so that each pattern has only one interface with the surrounding medium and in that the method comprises at least the following steps:

Preferably, said identified creep parameters are provided to creep equipment. and in that it comprises a creep step of the initial structure, the creep step being performed by applying the creep parameters to obtain the predicted structure with the highest correlation value.

The creep equipment performs the creep step. Preferably, during the creep step, said equipment applies said identified creep parameters. Creep equipment includes:

a device for bringing the thermo-deformable layer to a given temperature;

a device for controlling the temperature of the thermo-deformable layer;

a device for controlling the duration during which the temperature of the thermo-deformable layer is subjected to a given temperature; a computer device for reading instructions in order to apply parameters, in particular temperature and time, during a creep step.

Optionally, this method according to the invention may further have the following optional steps and features.

Advantageously, the steps of prediction, calculation of correlation values and identification of the creep parameters are repeated with a plurality of initial structures whose geometries are different from each other. Advantageously, one identifies, among the plurality of initial structures, the initial structure making it possible to obtain the highest correlation value.

Advantageously, the creep step is carried out by taking said identified initial structure as well as the creep parameters making it possible to reach the highest correlation value for this identified initial structure.

According to an advantageous embodiment, the initial structure is obtained by printing the resin layer from a mold.

According to an advantageous embodiment, the initial structure is obtained by grayscale photolithography including the presence of a residual layer.

According to another embodiment, the invention relates to a method for simultaneous obtaining by creep, at a wafer, three-dimensional pattern matrices with possibly arbitrary arbitrary profiles. The method comprises the following steps:

an initial profile of the patterns formed in a thermo-deformable resin is determined by simulation, the said initial profile comprising a residual layer of non-zero controlled thickness;

simulation of creep conditions including at least an optimum temperature and creep time is determined by simulation;

the two preceding steps are repeated until, by successive approximations, a predetermined correlation value is obtained between the profile obtained and the one desired, by simultaneously or successively correcting either the initial profile and the residual thickness, or the creep conditions, or both ;

the simulation results are accumulated and the profile and / or creep conditions are determined to obtain the best correlation value.

In another aspect, the present invention relates to a computer program product or to a computer-readable non-transitory medium, comprising instructions, which when performed by at least one processor, execute the steps of the process mentioned above.

According to another embodiment, the invention relates to a method for manufacturing a nanoscale printing mold carrying a plurality of structures intended to penetrate a deformable material in order to print in said material said plurality of structures, the method of characterized in that the structures are obtained by performing the steps of the method according to the invention.

BRIEF DESCRIPTION OF THE FIGURES

The objects, objects, as well as the features and advantages of the invention will become more apparent from the detailed description of an embodiment thereof which is illustrated by the following accompanying drawings in which:

FIGURE 1 illustrates the prior art and standard formation of microlenses by creep. FIG. 2 illustrates the process of the invention in which a residual layer is provided during creep.

FIGURE 3 discusses the characteristics of aspherical lenses.

FIGURES 4a and 4b illustrate a simulation of the effect of the annealing step after printing in the resin of a simple structure of rectangular section leaving in place, respectively, a thick layer and a residual thin layer of resin. FIG. 5 illustrates steps of an exemplary method according to the invention which makes it possible to determine the creep conditions suitable for obtaining a structure identical or close to a desired structure.

FIGURES 6a and 6b illustrate two types of initial profiles, one rectangular and the other pyramidal, used to form a network of aspherical microlenses.

FIGURES 6c and 6d illustrate simulation results for defining the conditions for forming an aspherical microlens array from the two types of initial profiles illustrated in FIGS. 6a and 6b.

FIGURES 7a and 7c briefly describe an example of a method suitable for forming pyramidal-shaped initial patterns in a polymer resin.

FIGURES 8a to 8e show results of annealing experiments carried out from pyramidal patterns resting on a thin residual layer.

FIGS. 9a to 9e illustrate an implementation of the method of the invention for the production of microlenses of the Fresnel type.

FIG. 10 illustrates steps of an exemplary method according to one embodiment of the invention and which make it possible to obtain complex forms by creep. FIG. 11 describes the details of the simulation steps used as a function of the structure of the pattern to be fluted. The drawings are given by way of examples and are not limiting of the invention. They constitute schematic representations of principle intended to facilitate the understanding of the invention and are not necessarily at the scale of practical applications. In particular, the relative dimensions of the different patterns and layers are not representative of reality.

DETAILED DESCRIPTION OF THE INVENTION

In the context of the present invention it is specified that the term "over", "overcomes", "covers" or "underlying" or their equivalent do not necessarily mean "in contact with". For example, the deposition of a layer on a substrate does not necessarily mean that the layer and the substrate are in direct contact with one another, but that means that the layer at least partially covers the substrate while being directly to its contact either by being separated from it by another layer or another element.

In the context of the present invention, a three-dimensional pattern designates a pattern having in a given layer, for example a resin, at least two height levels above an upper face of the layer when the pattern is in protrusion or at least two depth levels below the upper face of the layer when the pattern is recessed. The three-dimensional pattern may have a curved profile.

Before describing in detail an embodiment of the present invention, a solution according to the state of the art will be commented to facilitate the understanding of the invention.

The document cited previously in the section on the state of the art,

"Technique for monolithic manufacture of microlens arrays" (ZD Popovic et al.) Aims to provide a solution to reproducibly form and in a limited series of operations spherical microlens matrices. The authors describe a method for creeping a polymer resin, of the type of photoresist used in photolithography, which makes it possible to form the microlenses. As shown in FIG. 1, a cylinder or a resin line 1 10, most often of substantially rectangular section, is generally created by photolithography on a substrate 120, and is heated during an annealing operation 130 until that the section acquires a substantially spherical shape 140 thus conferring on the lens its optical properties. The microlenses obtained by this method, however, can only be thin in order not to degrade the optical characteristics of the microlens. Long focal lengths are therefore obtained by this method.

The annealing is performed on a structure having triple lines, or so-called triple points, 150 where three elements interact: the air, the polymer resin and the substrate material. The envelope of the structure 1 10, 140 has in fact at its base a contour (rectangular for the linear structure 1 10 or circular for the cylindrical structure 1 10 and for the spherical structure 140) formed of a multitude of points which are in contact with two materials in addition to the material of the structure itself. Thus, each point of this contour is in contact with the surrounding air and the substrate. The free surface of the thermosetting layer forming the pattern 1 10, 140 is in contact with the substrate 120 at the base of the pattern.

In the development of the present invention, it has been discovered that it is largely the presence of triple points that makes it extremely difficult to predict the final shape of the microlens after creep. It has also been found that the final shape strongly depends on the underlying substrate and parameters such as the surface tensions and the wetting angles defined at the triple points. Consequently, in practice only a very limited number of forms can be obtained by this method and it is always necessary to carry out numerous preliminary experiments to approach a possible form which may, however, be substantially different from the ideally desired form. for a given application. For example, with the above method it is known to perform by creep microlenses of substantially spherical shape.

As illustrated in FIG. 2, the invention provides a solution to these problems by describing a method in which the creep of a material (usually referred to as "reflow" in English) is always carried out while maintaining a residual layer 201. Thus, there is no triple point 150 near the pattern or patterns. The initial structure 200 is continuous. The triple points are replaced by points here double 215 formed by the single interface thermo-deformable layer / surrounding environment. The surrounding medium, referenced 217 here is typically air. Thus, a double point 215 is defined as a point on the envelope of the structure 200 which is in contact with a single medium (air) 217 in addition to the material of the thermo-deformable layer. Thus, the free surface of the layer forming the initial structure 200 is not in contact with another medium such as the substrate on which the initial structure 200 rests and this even at the base of the initial structure 200. Thus, the initial structure 200 has no line or edge taken on the contour of the thermo-deformable layer and which is in contact with both the environment 217 and the substrate 120 on which the thermo-deformable layer rests.

In the following description of the invention, it will be noted that such a profile is referred to as a "continuous profile". In the context of the invention, a network of three-dimensional patterns 210 formed in the same material, without there being discontinuity of the material between the reliefs of the three-dimensional patterns formed before creep, is a continuous profile. A continuous profile according to the invention is therefore most generally characterized, as mentioned above, by the presence of a continuous residual layer 201. The patterns 210, here protruding reliefs formed before creep can be generally of any shape. They may be, for example, of triangular shape as illustrated in Figure 2 or comprise one or more discrete levels 212 as shown in dashed lines. In general, their shape is chosen to obtain, after creep, the form 240 best suited for the application in question.

Conversely, a shape is considered non-continuous if the layer in which the pattern is formed, typically the resin, is interrupted. Such would be the case with a network of pads 1 10 arranged on a substrate 120 without continuity of material.

Unexpectedly, it has been found that the presence of the residual layer 201 makes it possible to predict with reliability and reproducibility the final shape 240 obtained after annealing 130. In this case, the evolution of the shape of the initial patterns 230, which occurs during the annealing operation 130, is the same regardless of the underlying substrate while it strongly influences the dynamics of the triple points in the case of a creep without residual layer. Moreover, the choice of the thickness of the residual layer gives an additional degree of freedom to control the final profile 240 of the microlenses as will be seen below.

The invention relates to plates, usually designated in English wafers, a substrate is completely covered with a layer in which the patterns are formed. It nevertheless extends to wafers whose substrate is partially covered with such a layer. In general, this layer extends in all the reasons which must be subjected to a controlled creep (reasons which will have an operational function after creep). More specifically, this layer extends under all the patterns that must be creep controlled and up to a minimum distance separating these patterns from any triple points born from an interruption of the layer.

Preferably, when the layer according to the invention defines a triple point, the pattern closest to this triple point must be distant from this triple point by a distance greater than or equal to the distance Dmini, with Dmini defined by the equation next :

Dmini = 2 Max (hr, hd, À) and preferably Dmini = 5 Max (hr, hd,, λ) with hr = average thickness of the residual layer between the pattern closest to the triple point and the triple point; with hd = height of the pattern closest to the triple point. The height of a pattern is measured between the highest point of the pattern and the base of the pattern. The base of the pattern 210, 240 is at the free surface 216 of the residual layer between two patterns. The height thus corresponds, for a residual layer of constant thickness disposed on a substrate 120 whose upper face 216 is flat, the difference between the height of the pattern measured from the upper face of the substrate and the residual thickness measured at from this same upper face of the substrate and; with λ = the distance measured between the two patterns that are closest to the triple point.

The thicknesses and heights are taken in a direction perpendicular to the main faces of the substrate on which the layer in which the patterns are formed rests. In the figures, the direction of measurement of the thicknesses and heights is vertical.

Thus, subject to respecting this condition of distance between the triple points and the patterns that one wishes to deform by creep, the invention extends to the wafers whose layer defining the patterns form one or more separate zones and partially covering the underlying substrate.

Naturally, the invention is not limited to stacks of layers in which the layer in which the patterns to be flowing are formed is in direct contact with the substrate. It also extends to the stacks of layers in which one or more layers or other elements are arranged between the substrate and the layer in which the patterns to be fluted are formed.

The method of the invention makes it possible, for example, to form so-called aspherical microlenses whose profile and its mathematical expression 320 are illustrated in FIG. 3. The aspherical lenses are conventionally defined using two main parameters: their radius of curvature "R" 312 and their degree of conicity "k" according to equation 320. Diagram 310 of FIG. 3 illustrates the difference between a spherical profile 314 and an aspherical profile 316. In equation 320 the constants of aspherical deformation higher than two have been neglected. When k = 0 the lens is spherical, with k greater than -1 it is elliptical, with k equal to -1 it is parabolic and finally when k is less than -1 the lens is hyperbolic. The lenses qualified as aspherical are all those where k is not zero.

By carefully controlling the profile of the aspherical lenses they may not be affected by the problems of convergence and blurring large openings that are observed with spherical lenses. The use of aspheric microlenses adapted to the chosen application results in a spectacular increase in optical performance, a single aspherical lens is then sufficient to obtain a very good image at all focal lengths. They can drastically reduce the number of lenses in optical systems, while significantly improving performance.

The invention provides a simple solution to the manufacture of aspherical microlens arrays by allowing the optimal profile of the microlenses can be obtained directly after a creep operation and without subsequent shaping of each lens. The method of the invention is not limited to this application example and is likely to be generally suitable for the formation of complex 3D shapes for all kinds of applications. In particular, the creep method of the invention will advantageously be used for the manufacture of printing molds of micrometric or nanometric size patterns.

FIGS. 4a and 4b illustrate a simulation of the effect of the annealing step after printing in the resin of a simple structure of rectangular section, 410 and 440 respectively, each comprising only two different printing levels 412 and 442 of 414 and 444. FIGS. 4a and 4b make it possible to compare the effect of a thick residual layer 430 and that of a thin layer 460 on the final shapes obtained after creep; ie 420 and 450. Note that a thick residual layer 430 produces a rounded pattern background 422 while a thin residual layer 460 produces a flat pattern background 452. The residual thickness is therefore, as already mentioned above, an important factor for controlling the final shape of the profiles 420 and 450. The diagrams of Figures 4a and 4b use on the abscissa and on the ordinate dimensionless normalized values which are, respectively, those of the width and the height simulated patterns.

It will be noted here that the modes of simulation of the final form which are implemented by the invention make use of models of behavior of the resin during creep which are of very different natures depending on whether the residual layer is thin or thick . If the layer is thick as in the example of FIG. 4a, the capillarity phenomena are then predominant. If the layer is thin as in the example of Figure 4b the behavior of the resin is then governed by lubrication type models. The method for determining which of the models is to be used, among the model of the phenomena of capillarity and the model of the lubrication, will be detailed later with reference to FIG. FIG. 5 illustrates steps of an exemplary method according to the invention for determining the creep conditions suitable for obtaining a structure identical or close to a desired structure. A step 101 consists in defining a desired shape, also designated desired structure S1, which one wishes to obtain ideally. This form can be freely chosen. For example, one can choose a form of aspherical microlens.

Step 1102 includes selecting an initial form S2 from which the creep operation will be performed. This step also comprises the representation numerically or mathematically of this form S2. In a particularly advantageous manner, this form has a residual layer as defined above, that is to say a layer which extends under the patterns to remove any triple point or push back possible triples at a sufficient distance from each of the patterns whose evolution must be controlled by creep.

The choice of the initial form S2 depends on many parameters and in particular the techniques and practical constraints for the realization. This form will differ for example depending on the lithography technique used. Moreover, for the same lithography technique this form will depend on the equipment used. For example, if the realization involves nano-printing, this form will depend on the number of levels of the mold.

Creep parameters such as creep temperature (step 103) and time can also be defined. This step makes it possible, for example, to define a maximum temperature which makes it possible not to damage other layers or components of the stack of layers to which the layer to be flowed belongs.

During the step 1 104, a simulation of the evolution in time of the initial form S2 during creep is carried out. At each moment, under the effect of creep, the simulation predicts a shape S3. At the end of the simulation, we have n forms S3 ,, each corresponding to a moment of creep. For example, the prediction of the form S3 at time t = 0, that is to say when the creep has not yet begun, provides a shape S3 ₀ identical to the initial form S2 since this initial form n has not changed yet. The prediction of the form S3 at time t = ∞, that is to say at the end of a time long enough for the form S3 to no longer evolve, provides a shape S3 _∞ plane, all the patterns of the initial shape S2 having been fully smoothed. In order to make these predictions of the S3i forms, the equation mentioned above is advantageously used. A computing unit equipped with a microprocessor makes these predictions and uses the equations adapted to the initial form S2.

Step 1 illustrates the calculation of a correlation factor between the forms S3i predicted by simulation and the desired form S1. This is typically the ratio of covariances and nonzero product of standard deviations. These calculations may be performed at the end of the simulation as illustrated on the graph or be carried out in parallel with the simulation of step 1 104, ie as and when the predicted forms S3i are determined.

Step 1 106, performed at the end of simulation or in parallel with step 1, comprises the identification of the predicted form S3i which makes it possible to obtain the best correlation factor.

According to a first embodiment, step 1 106 leads directly to step 1 108 at which the creep parameters are determined which make it possible to obtain this form S3i for which the correlation is the best. The creep parameters are taken from: creep temperature, creep time.

The method comprising the preceding steps thus makes it possible to determine the optimal creep conditions from a given initial form S2. Thus, by imposing an initial structure, for example to meet constraints related to available lithography equipment, the invention makes it possible to identify the optimum temperature and the optimal creep time for this initial structure.

At the end of this method of determining the optimum creep conditions, it is then possible to carry out the initial structure S2 and subject it to creep under these conditions. This step 110, which follows the simulation process, may or may not be integrated into the invention.

According to a second embodiment, the simulation method comprises additional and optional steps that make it possible to optimize the initial structure by performing successive iterations.

At the end of step 1, it is determined whether the highest correlation factor obtained from the initial form S2 is greater than or equal to a predetermined correlation threshold (step 1 107).

If this correlation factor is greater than the threshold, then step 1 108 is performed. The initial shape S2 and the optimal creep parameters are retained. The method of predicting and determining the parameters then ends and the creep operation 1109 can be performed. On the other hand, if the correlation factor is lower than the threshold, then an additional step 11 of modifying the initial form S2 is performed. Then, steps 1 103 to 1 107 are performed again based on the modified initial form S2. These iterations are repeated until a correlation factor higher than the threshold is obtained.

This embodiment then makes it possible to determine both the initial structure and at the same time the creep parameters which make it possible to obtain a final shape S3 that is identical or similar to the desired shape S1.

It will be noted that this method can be performed by modifying the initial form S2 (step 1 1 10) without using a correlation threshold. Indeed, it will be possible to make as many iterations as the initial number of shapes that one wishes to test.

It will also be noted that this method can also be reiterated by modifying only the creep parameters of step 1 103 and in particular the creep temperature and the creep time. The invention thus makes it possible to determine the ideal creep temperature and creep time for a given initial shape S2.

Thus, by imposing a maximum creep time, for example in order to meet the requirements of industrial productivity, the invention makes it possible to identify the optimum creep temperature for a given structure.

Similarly, by imposing a maximum creep temperature, for example to preserve the integrity of underlying layers or components surrounding the patterns, the invention makes it possible to identify the optimal creep time for a given structure.

It will also be noted that this method can also be reiterated by modifying both the creep parameters of step 1 103 and the initial form S2 (step 1 1 10 and 1 102). The invention thus makes it possible to determine the optimum initial shape S2 as well as the ideal creep temperature and creep time for this initial optimum shape S2.

Preferably, at least steps 1 104 to 1 108 are performed by a microprocessor.

FIGS. 6a to 6d illustrate simulation results that can for example be obtained by following the steps of FIG. 5. In this illustrative example, these simulations are intended to define the conditions for forming a network of aspherical microlenses starting from two initial profiles, one rectangular 510 and the other pyramid 520 as shown respectively in Figures 6a and 6b. The geometrical parameters that will influence the final shape after creep are, as indicated, the thickness of the residual layer hr, the height of the patterns hd and their width. hb. The other parameters that condition the final shape of the lenses are of course the temperature and creep time.

FIGS. 6c and 6d show, using diagrams 530 and 540, the concomitant evolution, during the creep phase, of the taper parameters (R and k) of the aspherical lenses respectively obtained starting from a profile. rectangular 510 and a pyramidal profile 520 for various values of the geometric parameters: hr, hd and hb.

It will be noted that for FIGS. 6d and 6c the repetition period of each of the patterns is 15 μηη, the value of hb is 14 μηη and that of hr is 500 nm.

The six curves in FIGS. 6c and 6d correspond to six different hd values which are: 3.5 μηη, 8 μηη, 12.5 μηη, 17 μηη, 21.5 μηη and 26 μηη, as indicated directly thereon.

Increasing creep times are indicated by the orientation of the arrows. The following conclusions can be drawn from the observation of these curves:

- The evolution curves of the parameters R and k present mainly the same type of behavior in the two diagrams. Referring for example to the diagram 540, it is found, in a first step 541, with the increase of the annealing time, a simultaneous increase of the two parameters R and k. The positioning of the different curves depends essentially on the initial profile chosen and the associated geometrical parameters.

In a second step 542, when we continue to increase the annealing time, while the parameter R continues to grow much more slowly, there is a very rapid decrease in the parameter k.

It can also be seen that the correlation or deviation between the desired shape, that is to say that corresponding to equation 320 of FIG. 3, and that obtained by simulation, that is to say that represented on FIG. the diagrams, is always very close to 1 except in the area 531 of the diagram 530, corresponding to the choice of a rectangular initial profile 510. For this diagram is found for weak annealing times a poorer correlation. This observation can be done on diagrams 530 and 540 using the ordinate scales, 535 and 545, located to the right of these and which indicate, in grayscale on the curves, the degree of correlation in the form a parameter (1 -R ² ) measuring the difference with the square of the correlation coefficient (R). The lowest values of deviations correspond to the darkest tones.

It will be noted on the diagram 540 that only hyperbolic lenses (k less than -1) can be produced starting from an initial pyramidal profile. In this case the taper parameters can be adjusted from the initial shape of the pyramid (angle) and the annealing time.

As mentioned above, the correlation at the beginning of annealing (zone 531 of the diagram 530) is lower in the case where one starts with rectangular shapes. However, it is possible in this case to make elliptical microlenses 532, that is to say with a parameter k of value greater than -1.

FIGS. 7a to 7c briefly describe an example of a method suitable for forming in a polymer resin pyramids of the type of those of FIG. 6b or of those of FIG. 8a.

A method well known to those skilled in the art consists of producing, using standard photolithography and etching processes developed by the microelectronics industry, a printing mold 610. The mold is preferably made of crystalline silicon. having a so-called crystalline orientation (1 1 1) which is the corresponding Miller index. To obtain the mold, a hard mask 620 is previously and conventionally created on the surface of a silicon substrate. This hard mask 620 will be used for etching 630 of the patterns to be printed in the silicon. The conditions of the etching are adapted so that it is preferably done along the crystalline plane (1 1 1). An etching with inclined flanks 640 is thus obtained as shown in FIG. 7a. The etching angle is then substantially 54.7 degrees which corresponds to the crystalline orientation (1 1 1) of the substrate.

Those skilled in the art also know how to obtain different etching angles by acting on the conditions of implementation of etching methods, dry or wet, commonly used by the microelectronics industry.

Figure 7b shows the mold 610 after etching, removal of the hard mask and reversal. It is then ready to be used to print pyramids in a polymer resin to manufacture, for example, a network of aspherical microlenses as previously described.

FIG. 7c shows a structure of pyramidal patterns 210 before annealing as already shown in FIG. 2. It is obtained in this embodiment by printing the mold 610 in a polymer resin previously deposited on a substrate 120. The printing leaves in place a residual layer 201 so as to obtain a continuous profile consisting of a single material as discussed above.

Preferably, a release layer consisting of a monolayer of molecules containing fluorinated atoms is deposited on the surface of the mold for facilitate its withdrawal after printing. The substrate to be structured 120 is for example covered with a heat-curable thermoplastic polymer film or by exposure to UV radiation. A thermoplastic is for example polymethyl methacrylate (PMMA) or polystyrene (PS). A UV-curable polymer is typically a photosensitive resin such as, for example, the so-called SU-8 resin which is widely used in lithography. The mold and the substrate are heated to a temperature above the glass transition temperature (Tg) of the polymer used. The heating temperature is typically selected in a temperature range of 10 to 50 ° C above the glass transition temperature. The mold is then pressed into the polymer film until the cavities are completely filled. The pressure exerted ranges from a few bars to forty bars. Then, the mold and the substrate are cooled to a temperature below the glass transition temperature and then separated.

Figures 8a to 8e show results of annealing experiments carried out from pyramidal patterns resting on a thin residual layer forming a structure similar to that of Figure 7c. The pyramids were printed in a SU-8 type resin using a silicone mold as described in FIGS. 7. The samples were then heated at 65 ° C for different periods of time and then cooled rapidly to at room temperature. FIGS. 8a to 8d are images obtained using a scanning electron microscope (SEM) for increasing annealing times:

- Figure 8a shows the initial pyramidal pattern after printing in the resin. The residual layer (not shown in its entire thickness) in this case has a thickness of 94 nm (nanometer = 10 ^"9 meters).

FIGS. 8b to 8c show the evolution of the shape of the pyramidal pattern, respectively: after an annealing of a duration of 1 minute, after 2.5 minutes and after 10 minutes. By increasing the annealing time one would obtain a flat shape, without motive.

The taper parameters (R and k) were calculated for the different shapes observed. As shown in FIG. 8e, we find the same type of evolution of the pairs of conicity parameters R, k as a function of the annealing time 710 (in minutes) than that obtained by simulation in FIG. 6d.

The invention thus makes it possible to simulate obtaining the aspherical microlens arrays having the desired conicity. It follows that on the basis of the results of these simulations it is possible to determine the optimal parameters which will allow us to effectively obtain the desired conicity at the end of a real creep operation. It is therefore known to produce with the method of the invention networks of aspherical microlenses whose conicity can be precisely controlled.

FIGS. 9a to 9e illustrate an implementation of the method of the invention for the production of microlenses of the so-called Fresnel type.

FIG. 9a illustrates a large lens 810 (relative to the dimensions of the microlens arrays to which the invention is directed), the thickness of which may have to be significant in order to obtain the desired optical properties. It has been known since the beginning of the 19th century (A. Fresnel) that a convex plane lens can be cut into concentric annular sections 820 which together provide the same optical properties as a single thick lens.

This type of lens can be advantageously made with the method of the invention. As shown in FIG. 9b, concentric rings 830 are initially formed, for example in a polymer resin disposed on a substrate 120, the profiles of which are at first approximated by steps of different levels, in greater or lesser numbers. These initial patterns are therefore formed of a stack of discrete volumes. At this stage, therefore, the initial patterns intended to be polished are formed, at the end of which they will acquire their final form. We always note the presence of the residual layer 201.

Various methods make it possible to produce this type of initial profile in stair steps preserving a residual layer. For example, it can be obtained by printing the corresponding mold in the resin in a manner similar to that described with reference to FIGS. 7a to 7c. It is known to produce standard printing molds having several discrete levels of printing. The number of discrete levels that can be obtained in practice is, however, very limited and their number directly impacts the manufacturing cost of the mold.

Another way to obtain the discrete levels of the initial patterns is to use a particular type of lithography known as "gray-scale lithography", that is, grayscale lithography. In this technique, it is possible to insolter differently the contiguous zones of a photosensitive resin in such a way that, after development, the latter has level differences which are a function of the locally received insolation.

The techniques briefly described above are well known to those skilled in the art.

Figure 9c illustrates the shape obtained after creep. The creep conditions are predictively determined. To do this, one of the embodiments of the methods illustrated in FIG. 5 can be applied. FIG. 9d is a photograph carried out using a scanning electron microscope (SEM) which shows initial patterns in stair steps, produced for example by grayscale lithography, which make it possible, after annealing, to obtain , a Fresnel lens as shown in Figure 9c.

FIG. 9e illustrates the evolution of an initial staircase profile 840 after different creep times 850. A smoothing of the initial shape is obtained which makes it possible to approximate with a good precision the ideally desired shape by carrying out simulations such as the allows the presence of a residual layer and by acting as previously on the various geometric and physical parameters available: in particular, shape and dimensions of the initial profile, thickness of the residual layer, temperature and duration of the annealing.

According to another embodiment, the 3D forms produced using the method of the invention are used for the production of a printing mold which will itself be used by a production method of a particular device. In this case, as already discussed in Figures 7, the realization of the mold may include the steps necessary for the deposition of a release layer. The production of a mold may also require stabilization of the resin, or the thermoplastic material used, and a treatment ensuring its longevity of use as a metallization of the exposed surfaces during printing.

FIG. 10 illustrates the steps of a method of the invention that make it possible to obtain possibly complex forms by creep. As shown in FIG. 11, obtaining the final shape 1010 of the patterns that it is desired to obtain firstly passes through the definition 1010 thereof as a function of the application in question. For example, in the case of a microlens array, it is known very precisely to define the shape of the microlenses required for the application in question as a function of the desired focal length and aperture.

The initial form 1020 is then selected from which the creep operation will be performed. This choice depends on many parameters including the definition of the means with which the initial shape can be obtained, for example, using standard photolithography operations or by using the so-called grayscale photolithography or starting from a mold printing. The choice will be done in an iterative manner by predicting the final form by simulation 1030 using a library of possible initial shapes 1025 already tested and possibly using a convergence algorithm 1035 to obtain by simulation a shape, otherwise absolutely identical, at least very close to the desired final shape. The appreciation of the proximity between form The desired shape and shape can be defined rigorously, for example, by using a correlation coefficient between the two forms and continuing the above iterations until the goal is reached or exceeded.

The 1025 form library can be enriched by all previous experiments. It will be recalled here that the simulation step is made possible by the presence of the residual layer.

When the final form obtained is satisfactory, all the simulation data including, inter alia, the temperature and creep time as well as the residual thickness which have made it possible to obtain the desired final simulated form are extracted. These data will allow the effective implementation of the manufacturing method according to the invention.

Moreover, depending on the choice of the initial form that has been performed for the simulation, the means for obtaining this initial shape are set up. This step may include the manufacture of lithography masks or a printing mold. It will not be necessary if the means for forming the initial patterns does not require a mask or mold, which would be the case if these initial patterns are obtained by electron beam lithography for example.

The actual formation of the patterns to be flowed can then be effected. This embodiment step includes controlling the thickness of the residual layer.

The creep operation itself is then carried out at a temperature and for a time defined by the simulation.

The results obtained are then validated 1090 with, possibly, corrective actions applied at the iterative simulation loop 1020, 1025, 1030 and 1035.

Figure 11 describes the simulation steps according to the structure of the pattern to be fluted. As shown in step 1210, the choice of the creep model to be applied depends on the geometry of the pattern, in particular by comparing the values of the average residual thickness (hr), the average height of the pattern (hd) and the horizontal extension (hb) of the pattern.

It is known that the movement of a fluid assimilable to a continuous, homogeneous, incompressible medium, devoid of a chemical reaction within it, and having a temperature identical at every point, is described by the following two equations:

Conservation of =

the mass: 0. (El) (E1)

Equation of

Navier-Stokes: (E2)

where v represents the velocity field in the fluid, p the density of the fluid, t the time, p the pressure field, T the tensor of the viscous stresses and f the sum of the remote action forces. To these equations, we must add the boundary conditions of the fluid domain which are in our case:

Zero speed (non-slip) to _n

the fluid-substrate interface: (0

Laplace pressure (voltage of "- n = y X κ (E4) <PA \ surface) at the fluid-air interface: ^t '* _. * ^J

where pi and pe are the inner and outer pressures, γ a physical parameter called surface tension, and κ the local curvature of the interface.

In the field of microfluidics, it is known that, due to the reduction of length scales, the Navier-Stokes equation is reduced to the Stokes equation:

V _/ > = V-T + f (E5) The cases we are dealing with here have no action forces at a distance (no electrostatic force, for example), so the Stokes equation is reduced to:

The last necessary equation, called closure equation, or law of behavior, links the tensor of the viscous stresses to the kinematics of the flow, even to the history of the flow:

VT = / (v, J ^' v, ...) (E7) In the simple case of an incompressible Newtonian fluid, the law of behavior is written:

VT = riV ² v (E8) where η is a physical parameter called Newtonian viscosity.

The simulation of a creep, that is to say the simulation of the topographic evolution of the patterns, amounts to solving the movement of the free interface (fluid-air interface). This movement is determined by the flow of the fluid, and its simulation requires the resolution of the system bringing together the equations (E1), (E3), (E4), (E6) and (E7), which we will call complete system.

We will now see that, subject to certain geometric conditions, the resolution of the flow can be approximated by simplified models. Consider an elementary pattern which one wishes to simulate the creep, as represented on the diagram 1210. This motive is characterized as it was seen by "hr" its average residual thickness, "hd" its average height, "hb" its extension in a plane parallel to the face of the substrate receiving the horizontal layer and the width "L" of the zone in which it is inscribed. We denote "L" the width of the zone considered, that is to say the width of the area comprising the pattern and area where there is only the residual thickness. The width "L" is illustrated in FIG.

The choice is made first at step 1220 where the average height of the pattern (hd) and the horizontal extension (hb) of the pattern are compared.

If the ratio hd / hb is not low, that is to say if it is not less than 1 then we use the first simulation model 1230 where we proceed to the resolution of the equation Stokes and the complete system. Indeed, if the form ratio hd / hb is medium or large, that is to say in practice greater than 1, then no approximate model exists and the resolution of the complete system is necessary. A finite element or finite volume calculation code can be used, for example, using commercially available software such as: COMSOL, FLUENT and OPENFOAM. The calculation time on a personal computer is of the order of a few minutes to several hours depending on the size and complexity of the pattern.

On the other hand, if the aspect ratio is small (in practice less than 1), a linearization of the boundary condition E4 can be made:

V ² where h is the Laplacian of the local thickness. Two models can then be used which correspond to steps 1250 and 1260 of FIG.

If the magnitude of the pattern is of the same order or larger than the residual thickness, i.e., it is not of low amplitude 1240, then the lubrication theory 1250 can be applied. Typically, if hd / hr <0.5 then, step 1250 is applied. This theory is widely used in the field of thin films [cf.

A. Oron, S. H. Davis, and S. G. Bankoff. Reviews of Modem Physics 69, 931 (1997); R.

V. Craster and O. K. Matar. Reviews of Modem Physics 81, 1 131 (2009)]. Its principle is as follows: a first-order asymptotic expansion of the Stokes equations (E6) and the mass balance (E1) can be done considering the form ratio

(hd / hb) as an infinitely small. The main result is that the pressure gradient is horizontal, so the flow is mainly parallel to the substrate.

Precisely, we find that the local variation of thickness h is given by the Reynolds equation:

It should be noted here that the fluid is considered to be a fluid

Newtonian. This equation, although strongly nonlinear, contains in it the boundary conditions and all the kinematics of the flow. It is only about the thickness of the film, that is to say on its topography, and we are therefore saving the resolution of the velocity field inside the fluid.

The Reynolds equation can be solved by a finite volume method [cf. Y. Ha, Y.-J. Kim and T. G. Myers. Journal of Computational Physics 227, 7246-7263 (2008)]. The calculation time on a personal computer is of the order of a few seconds to several minutes depending on the size and complexity of the pattern.

If now the amplitude of the pattern is much smaller than the residual thickness, that is to say in practice less than half, another theory can be advantageously employed: the theory of capillary waves 1260. The wave theory capillaires is a physical model for describing the evolution of a free liquid interface subjected to a small deformation. On a human scale this can be likened to wrinkles caused on the surface of a lake by the wind or a stone's throw. This theory can adapt to the creep of a nanometric or micrometric pattern. If the deformations of the interface are small, then the pressure at the interface can be approximated by that at the level of the average thickness (denoted Hm) p {h) ~ p {Hm) (E1 1)

Repeating Figure 11, we have Hm = hr + (hd * hb) / L.

The calculation consists in decomposing the topography of the free surface into plane waves (capillary waves of wave vector k), and studying the dynamics of the flow in the frequency domain (of frequency ω). The study in the frequency domain is not essential for a Newtonian fluid, but it allows to take into account viscoelastic fluids whose viscosity depends on the frequency (this viscosity, noted η (ω), is called in general complex viscosity ). This process makes it possible to transform the partial differential equations (E1) and (E6) into algebraic equations. The details of the calculations are not reported here [cf. E. Rognin, S. Landis, and L. Davoust. Physical Review E 84, 041805 (201 1)]. As is often the case in an undulatory theory, the result takes the form of a dispersion relation, ie a necessary condition which links the frequency of the wave ω to its wave vector k via the different physical and geometrical parameters. . This dispersion relation is expressed by:

where i is the imaginary unit, k is the norm of the wave vector, and f is a dimensionless function of the wave vector normalized by the mean thickness Hm:

The difficulty of the problem now lies in solving the dispersion equation, that is, expressing ω as the explicit function of k. This is done simply in the case of a Newtonian fluid, that is to say when the viscosity does not depend on the frequency. Indeed, we obtain that each capillary wave, that is to say each mode the topography, decreases exponentially like:

whereTi (k, t) is the amplitude of the wave vector mode k at time t.

Simulating the topographic evolution of the film is therefore, in this case, to decompose the topography into plane waves by a Fourier transform algorithm, using software such as MATLAB or OCTAVE, and to apply to each mode the exponential multiplying coefficient of Equation (E14). The computing time on a personal computer is less than the second.

In view of the foregoing description, it is clear that the invention makes it possible to provide a reliable, simple and inexpensive solution for each of the following problems:

determination of the optimum creep conditions to obtain the final structure closest to the desired structure starting from a given initial structure;

determination of the optimal initial structure to obtain, after a creep under imposed conditions of temperature and time, the structure closest to the desired geometrical structure;

determination of both the initial structure and the optimal creep conditions to obtain the structure closest to the desired structure.

The invention thus makes it possible to accurately predict the evolution over time of a form subjected to creep. It therefore makes it possible to considerably reduce the number of experiments that were necessary with the existing solutions, in particular to obtain complex structures.

The invention therefore makes it possible to significantly reduce the cost of obtaining these structures. It also appears that the method according to the invention offers many additional advantages among which:

- It allows to make 3D molds whose shape would be difficult to achieve otherwise. The form obtained by annealing may be used as a polymeric mold or etched in a substrate to form the mold itself.

it gives access to shapes in a simple manner by reducing the number of conventional lithography steps necessary to obtain them;

it gives access to a large number of geometries from the same impression or from the same lithography;

it allows the production of molds on a wafer scale, typically 8 to 12 inches;

allows the production of molds for injection from, for example, electroplating of the annealed mold.

Moreover, the method of the invention provides the field of producing aspherical microlens array solutions or significantly improves the following points:

- reduction of the focusing spot of microlenses and reduces their spherical aberrations;

- increase in the numerical aperture of microlenses;

for a fixed lens diameter, the focusing distance can be chosen as close as possible to the aspherical microlens, which is not the case with hemispherical or spherical microlenses;

access to a large number of microlens geometries from an impression;

- increase in microlens spatial density and fill rate;

- potential application to all applications that require high focus and large numerical apertures.

The invention is not limited to the previously described embodiments and extends to all the embodiments covered by the claims.

In particular, the invention is not limited to the production of microlenses and extends to the realization of all types of devices for electronic, micromechanical, electromechanical (MEMS, NEMS ...), optical or optoelectronic (MOEMS. ..)

Claims

1. Method for determining at least one creep parameter for obtaining a structure approximating a desired structure (S1) by creep of an initial structure (S2) (200) different from the desired structure (S1), the structure initial (S2) (200) being made of at least one pattern formed in a thermo-deformable layer disposed on a substrate (120), characterized in that the thermo-deformable layer forms a residual layer (201) surrounding each pattern ( 210, 212) and from which each pattern (210, 212) extends so that each pattern (210, 212) has an interface only with the surrounding medium (217) and in that the method comprises less the following computer-implemented steps:

a step of prediction (1 104) of the evolution in time of the geometry of the initial structure (200) (S2) subjected to a creep, so as to obtain a plurality of predicted structure geometries (S3) associated each with creep parameters comprising at least one creep time and one creep temperature;

a step of calculating (1 105) correlation values of the geometry of each predicted structure (S3) with respect to the desired structure (S1);

a step of identifying (1 106, 1 108) the creep parameters making it possible to obtain the predicted structure (S3) offering the highest correlation value, the identified creep parameters being intended to be applied to a step of creep of the initial structure (200) (S2).

2. Method according to the preceding claim, wherein the residual layer completely covers the substrate (120).

A method according to any one of the preceding claims, wherein the residual layer (201) surrounds each pattern (210, 212) so that all lines or edges taken on the contour of each pattern (210, 212) only in contact with an environment (217).

4. The method of claim 1, wherein the initial structure (S2) (200) has so-called triple points, in contact with both an environment (217) and the substrate (120) on which rests the layer of thermo-deformable material, the triple points being distant from each pattern (210, 212) by a distance of at least Dmini, with Dmini = 2 Max (hr, ηά, λ) and preferably Dmini = 5 Max (hr, hd,, λ) with hr = average thickness of the residual layer between the pattern closest to the triple point and the triple point; with hd = height of the pattern closest to the triple point, and; with λ = the distance that separates the two patterns that are closest to the triple point.

5. Method according to any one of the preceding claims, wherein apart from the patterns, the residual layer (210) has a thickness that does not vary by more than 10% relative to the average thickness of the thermo-deformable layer. outside the patterns (210, 212).

6. Method according to any one of the preceding claims, in which the steps of prediction (1 104), calculation (1 105) of correlation values and identification (1 106, 1 108) of the creep parameters are repeated. with a plurality of initial structures (200) (S2) whose geometries are different from each other.

7. Method according to the preceding claim, wherein one of the plurality of initial structures (200) (S2), the initial structure (200) (S2) for obtaining the highest correlation value.

8. Method according to any one of the two preceding claims, in which the steps of prediction (1 104), calculation (1 105) of correlation values and identification (1 106, 1 108) of the parameters of creep with a plurality of initial structures (200) (S2) whose geometries are different from each other only if the highest correlation value for an initial structure (200) (S2) considered is less than a predetermined correlation threshold .

9. Process according to any one of the preceding claims, in which a maximum creep temperature is imposed.

10. Process according to any one of the preceding claims, in which a maximum duration of creep is imposed.

1 1. A method according to any one of the preceding claims, wherein the step of predicting (1104) the time course of the geometry of the initial creep (200) structure (S2) is dependent on the thickness of the residual layer.

A method as claimed in any one of the preceding claims, wherein the initial structure (S2) (200) is formed at least in part by superimposed cubes or blocks.

13. A method according to any one of claims 1 to 1 1, wherein the initial structure (S2) (200) has a triangular section in a section perpendicular to the plane in which extends mainly the substrate (210).

The method of any of the preceding claims, wherein the predicted structure structure (S3) providing the highest correlation value comprises one or more aspherical lenses or one or more Fresnel lenses.

A method according to any one of the preceding claims, wherein the predicted structure (S3) offering the highest correlation value is a structure for a tool for manufacturing a microelectronic device or a structure of a microelectronic device. .

16. A computer program product comprising instructions, which when performed by at least one processor, performs at least the steps of predicting (1 104), calculating (1 105) correlation and identification values ( 1 106, 1 108) creep parameters of the method according to any one of the preceding claims.

17. A method for obtaining at least one structure approximating a desired structure (S1) from at least one initial structure (S2) (200), different from the desired structure (S1), the initial structure (S2) (200) being made of at least one pattern formed in a thermo-deformable layer disposed on a substrate (120), characterized in that the thermo-deformable layer forms a residual layer (201) surrounding and from which extends each pattern (210, 212) so that each pattern (210, 212) has an interface only with the surrounding medium (217) and in that the method comprises at least the following steps:

a step of prediction (1 104) of the evolution in time of the geometry of the initial structure (200) (S2) subjected to a creep, so as to obtain a plurality of predicted structure geometries (S3) associated each with Has creep parameters comprising at least one creep time and one creep temperature;

a step of identifying (1 106, 1 108) the creep parameters making it possible to obtain the predicted structure (S3) offering the highest correlation value, said identified creep parameters being supplied to a creep equipment;

and in that it comprises a creep step of the initial structure (200) (S2), the creep step being performed by the creep equipment by applying said creep parameters to obtain the predicted structure (S3 ) offering the highest correlation value.

18. The method according to the preceding claim, wherein the steps of predicting (1 104), computing (1 105) correlation values and identifying (1 106, 1 108) creep parameters with a plurality of initial structures (200) (S2) whose geometries are different from one another and in which one of the plurality of initial structures (200) (S2) is identified, the initial structure (200) (S2) for obtaining the highest correlation value.

19. Method according to the preceding claim, wherein the creep step is performed by taking said identified initial structure and the creep parameters to achieve the highest correlation value for this identified initial structure.

20. The method according to any of the three preceding claims, wherein the initial structure (S2) (200) is obtained by printing the heat-deformable layer from a mold.

21. A method according to any one of claims 17 to 19, wherein the initial structure (S2) (200) is obtained by grayscale photolithography.

22. A method of manufacturing a nanoscale printing mold carrying a plurality of structures each approximating a desired structure (S1) and intended to penetrate into a deformable material in order to print in said material said plurality of structures, the manufacturing method being characterized in that each structure is obtained from at least one initial structure (S2) (200), different from the desired structure (S1), the initial structure (S2) (200) being made of at least one pattern formed in a thermo-deformable layer disposed on a substrate (120), characterized in that the thermo-deformable layer forms a residual layer (201) surrounding and from which each pattern (210, 212) extends so that each pattern (210, 212) has an interface only with the surrounding medium (217) and that the process comprises at least the steps of prediction, calculation, identification and creep of the method according to any one of claims 17 to 19.