EP1415200A1

EP1415200A1 - Method for printing a near-field photoinduced stable structure and optical fibre tip therefor

Info

Publication number: EP1415200A1
Application number: EP02772467A
Authority: EP
Inventors: Nathalie Landraud; Jacques Peretti; Frédéric Chaput; Georges Lampel; Jean-Pierre Boilot; Khalid Résidence Mondétour LAHLIL; Viatcheslav Safarov
Original assignee: Essilor International Compagnie Generale dOptique SA
Current assignee: EssilorLuxottica SA
Priority date: 2001-07-26
Filing date: 2002-07-26
Publication date: 2004-05-06
Also published as: FR2827967B1; JP4812251B2; JP2004536465A; WO2003010604A1; US20080199126A1; FR2827967A1; US20040214115A1; US7352384B2

Abstract

The invention concerns a method for printing a near-field photoinduced stable structure, and an optical fibre tip therefor. The method comprises emitting light waves with appropriate polarisation on a photosensitive material layer to induce therein a topographical modification through an opening of not more than 100 nm delimited by an opaque zone, said layer being located at not more than 100 nm from the opening. The invention is applicable to ophthalmic optics.

Description

Method for printing a stable photoinduced structure in the near field, and fiber optic tip for its implementation.

The present invention relates to a method of printing a stable structure in relief in a material comprising a photosensitive compound, by irradiation of said material from one or more localized light sources of dimension smaller than the wavelength of said irradiation. .

The growing development of computer processing powers and the generalization of information exchanges makes it necessary to have the means to increase the density of storage on information media.

In general, the information on the usual supports is printed in the form of repeating patterns obtained by modification of index or by local topographic modification. One of the major problems is to increase the resolution of the printing.

It is already known to make topographic surface modifications, for example in hybrid sol / gel materials containing azobenzene groups.

In the thesis “near field optical microscopy. Optical response at the submicron scale ”of January 22, 1999 by Philippe Bertrand, accessible at the university library of the Ecole Polytechnique, 91128 Palaiseau, France, an experiment is described in which a topographic structure was photoinduced in a soil layer / gel, using as light source an optical fiber whose tip was obtained by sequential hot stretching and not attacked in acid.

The inscribed structure is made up of letters themselves made up of studs 15 nm high and 40 nm by 40 nm sideways.

A similar result is described in the article entitled “photochromic sol-gel films containing dithienylethene and azobenzene derivatives: from the design of optical components to optical data recording. »Mol.Cryst. Liq. Cryst, 2000, vol 344, pp 77-82.

In general, these two documents underline that it seems excluded that this topographical modification comes from a simple thermal effect, as we had seen in other documents of the prior art ("Near-field magnetooptics and high density data storage ”. Near-field magneto-optics and high-density data storage. E.Betzig et al; Apρl.Phys.Letters, vol. 61 (2) pp. 142-144 (1992), in the measure where the polarization of the light plays a role on the deformation of the surface of the layer, and not only the intensity of the light source.

The first two documents above report preliminary results and the process which they describe succinctly has the drawback of lacking reproducibility.

Indeed, for a readable inscription, the method must make it possible, on the one hand, to perfectly master the geometry of the patterns inscribed, on the other hand to avoid any alteration of the structure of the layer at the periphery of the patterns, and precisely in the area separating two patterns.

Furthermore, it would be desirable to obtain patterns having a substantially identical relief.

Too much fluctuation in the altitude of the patterns could indeed compromise the reading of the information. Finally, it would be desirable to control and in particular to increase the relief, that is to say the height of the patterns, without altering the resolution of the printed network, while keeping the stability of the printing over time.

The object of the invention is therefore to provide an improved local printing process, in particular solving the above problems. The method according to the invention is a method of local printing of a stable structure in relief comprising a step of emitting light waves of appropriate polarization on a layer of a material comprising a photosensitive compound capable of inducing a modification topographic of said layer under the effect of said light waves, and is characterized in that said light waves of appropriate polarization are emitted on said layer through an opening of maximum dimension less than or equal to 100 nm, peripherally delimited by a zone opaque preventing propagation of light waves outside said opening, said layer having a surface located at a distance less than or equal to 100 nm from said opening. The aforementioned opaque zone limits light leaks. It allows to define a truly localized opening for light emission and to control its dimensions. The electromagnetic field is thus confined to the vicinity of the surface to be printed and the dimensions of the printed pattern and the resolution of a network made up of such patterns are perfectly controlled. Given the dimensions of the abovementioned opening and its positioning relative to the layer of photosensitive material, the method makes it possible to print the structure by light irradiation in the near field.

As is known, by near field light irradiation is meant irradiation by the components of the electromagnetic field, the amplitude of which decreases exponentially with the distance from the light source generating the field.

It is also quite surprising to note that the opening thus locally delimited makes it possible to keep the polarization of the light emitted through it.

Preferably, said opening has a dimension less than or equal to 60 nm, and better still, a dimension between 40 and 60 nm.

According to a preferred embodiment of the invention, said light waves of appropriate polarization emitted through said opening are transmitted by an optical fiber comprising a terminal portion having the general shape of a point at the end of which is located said opening .

More precisely, the end of the tip has the shape of a truncated cone, delimited by a generally planar face and a peripheral surface and the opening constitutes the generally planar face of the truncated cone, while the opaque zone is the peripheral surface of the cone.

Preferably, the terminal portion of the optical fiber has the property of retaining the polarization characteristics of the propagating light waves.

For this, the terminal portion of the optical fiber was obtained by sequential hot stretching.

The rest of the description refers to the accompanying drawings. FIGS. 1 to 6 represent ends of optical fiber observed by scanning electron microscope (TV1EB) at different stages of the method according to the invention. FIG. 7 schematically represents the device used for implementing the printing method according to the invention.

Figure 8 shows a top view of a network obtained according to the invention. FIG. 9 represents a sectional view of a relief structure printed on the surface of a layer comprising a photosensitive material. The preparation of optical fibers which can be used for implementing the method will now be described in more detail. The general principle of thinning an optical fiber by hot drawing is derived from that developed for biological micropipettes. It consists in locally heating, by absorption of infrared radiation from a CO ₂ laser, an optical fiber maintained in tension. Above a certain temperature, the heated area of the fiber stretches and thins. This results, under certain conditions, in the breaking of the fiber and the formation of a point.

The sequential hot stretching makes it possible to control the morphology of the point in situ, during its formation. For this purpose, the hot drawing of the fiber is carried out in a sequence of heating-cooling stages which are controlled by measuring the elongation and the thinning of the fiber.

Sequential stretching consists in heating the optical fiber by a series

(typically a hundred) laser pulses of variable duration, controlled by electronic system. Pulse number n, of duration τ _n , produces a stretch

Δl _n . Between two pulses, a delay is provided to allow the cooling of the fiber. The procedure is broken down into four phases:

- Phase 1 consists in defining the initial conditions of the stretch.

- Phase 2 achieves a thinning of the fiber in constant stages and ensures a conical morphology of the tip.

- Phase 3 precisely prepares the diameter of the fiber for the formation of the tip.

- Phase 4 consists of only a single laser pulse, the duration of which is such that it ensures the breaking of the fiber and the formation of the tip.

During phase 1, the duration of the laser pulses is incremented by Δτi as long as the elongation caused by a pulse does not exceed a certain threshold Δlτ. When after the pulse number i, a stretch Δli, greater than the threshold Δlτ, is measured, phase 1 ends and the initial conditions for stretching are defined. The duration Tj of this pulse is chosen as the reference time τ _re to define the time scale of the whole process; it is the duration of impulse necessary to stretch, of the length Δl _Is the fiber of typical initial diameter Di = 125 μm. The stretch threshold is an adjustable parameter whose typical value chosen is Δ ^ = 3μm. Under the operating conditions described above, for a CO ₂ laser with a power of 14 W, delivering a beam with a diameter of 2 mm, and for the indicated value of Δ _. (+ 3 μm), the value of the reference pulse duration is typically τ _ref = 50 ms. The value of the increment Δτ _T of the duration of the pulses is chosen to be small compared to the value of τ _ref : Δτi = 1 ms. Phase 2 corresponds to a slimming regime during which it is desired to keep the stretch produced by each pulse constant. The first parameter that we define is the setpoint, that is to say the value Δlc of the stretching that we want to obtain. It is then, for the impulse number n, to choose the value of τ _n , which will produce the stretching Δl _n = Δl _c . It is clear that, as the fiber becomes thinner, the duration of the pulse necessary to obtain a fixed value of the stretch decreases. Thus, τ _n will necessarily be less than i.

The decrease in the pulse duration is iteratively controlled. To determine the duration of the nth laser pulse, we first define the duration τ _n ° whose deviation from τ _n- ι is identical to that between τ _n _ι and τ _n _ ₂ . o τ _n -2 - Vι ^~ r τ _n

(1)

Then, the effective duration τ _n of the nth pulse is determined by adding a correction to τ _n which takes into account, on the one hand, the deviation from the setpoint measured during the pulse n-1 and, on the other hand, of the slope with which the stretch (or the deviation from the setpoint) varies between the impulses n-2 and n-1 τ „= τ _n ° + C _r δv _n = τ _π ° + C _r [ (Δl _c - Δl _n _ - (Δl _n-1 - Δl _n . ₂ )] / Δl _c (2)

It is this correction, the value of which depends on the measurement of the effect produced by the preceding pulses, which gives the process a retroactive character. The value of the correction depends on the coefficient C _r . This coefficient plays the role of a feedback gain applied to the error signal δv _n . Its value is chosen to keep the error signal as low as possible, that is to say to keep Δl _n as close as possible to Δl _c . The setpoint Δl _c and the feedback coefficient C _r are also adjustable parameters. Under the operating conditions described above, their respective values are typically Δl _c = 3 μm and C _r = 1.3 ms. Phase 2 ends when the value of the pulse duration calculated according to equation 2 becomes less than ατ _ref .

The coefficient α is an adjustable parameter whose typical value is 0.45. At this stage, the thinning of the fiber has made it possible to reach a diameter D ₂ in steps δD ₂ (the typical values of D ₂ and δD ₂ under the operating conditions described above are respectively: 20 μm and -1 , 5 μm). It is then necessary to continue the process in finer steps so as to prepare with the most high possible precision the diameter of the fiber which will allow the formation of the tip end with the desired geometric characteristics.

Phase 3 is then defined during which the duration of the pulses decreases in steps. During this phase, it is no longer only the measurement of the elongation of the fiber which is used to determine the value of the bearings but also the diameter D ₃ measured in situ using a microscope. As long as the thinning of the fiber (i.e. the variation δD _n of the diameter of the fiber) is less than the threshold 3D ₃ = 0.5 μm and that the stretching δl _n of the fiber is less than threshold δl _s = 1 μm, the procedure continues without changing the duration of the pulse. It should be noted that the choice of the threshold value δD ₃ = 0.5 μm is directly linked to the ultimate resolution of the microscope used to measure the diameter of the fiber.

Given the instrumental constraints, this resolution can hardly be improved, and can in no case be much less than 0.5 μm. The threshold value of the elongation (1 μm) is arbitrarily determined as being a typical value corresponding to the threshold thinning. The transition from one level to another therefore occurs when one of the measured values, the thinning or stretching of the fiber, exceeds the threshold value. When the fiber reaches a diameter D ₃ , phase 3 ends. D ₃ is an adjustable parameter. Under the operating conditions described above, its typical value is D ₃ = 15 μm.

The feedback principle changes again for phase 4. The duration of the last pulse is not fixed according to the result of the previous pulses. The stretching speed V _e of the fiber is continuously measured. As long as the laser does not heat the fiber, V _e is zero. The electronic control of the laser supply triggers the pulse which is only interrupted when V _e exceeds a certain threshold V _es . The stretch speed threshold is an adjustable parameter whose typical value chosen, according to the operating conditions described above, to form the desired spikes, is V _es = 0.002 ms-1. The tip obtained following this last pulse, visualized by SEM in Figure 1 and magnified by a factor of 50,000, has the shape of a truncated cone whose end has a diameter D ₄ between 0.5 μm and 1 .mu.m. The corresponding “apex” angle values θ are between 40 ° and 60 °.

D4 and θ are shown in FIG. 2, on which the fiber is the same as that of FIG. 1. The precise values for the fiber in Figures 1 and 2 are θ = 48 ° and D4 = 700 nm.

By this thinning procedure, the fiber separates into two identical points. One of these two twin tips is used as a control tip, the other will be used to manufacture the optical fiber according to the invention, or nano-probe.

After sequential hot stretching, the terminal portion of the optical fiber undergoes an isotropic chemical attack, preferably with an acid, in particular with hydrofluoric acid. The chemical attack step can be seen as the last step of the retroactive process: we measure the progress of the process and, depending on the result of this measurement, we adjust the chemical treatment that we undergo to the best. This last step has the particularity of not modifying the general shape of the tip, but it makes it possible to precisely adjust the size of its end. At the end of the stretching, the size and shape of the preformed tip, which characterize the progress of the process, are measured on SEM photographs of the control tip.

The complementary chemical attack of a preformed tip by sequential hot thinning is based on three principles: - The chemical attack is isotropic and therefore retains the shape of a truncated cone of the tip: the diameter of the upper face of the truncated cone ( the diameter of the opening of the probe) can be reduced from the value D4 to the value D chosen without the angle at the top of the tip, θ, being changed;

- The chemical attack speed can be adjusted independently of the other process parameters by choosing the concentration of the hydrofluoric acid solution;

- The chemical attack time required to obtain the desired final diameter depends only on the chemical attack speed and the values of D4 and θ (measured on the control tip). If e is defined as the thickness to be dissolved and the linear attack speed or attack speed in the direction normal to the surface V = de / dt, the attack time t necessary to obtain the value D final for opening is [cos (θ / 2) (D4-D) / 2V [l-sin (θ / 2)].

FIG. 3 represents the end of a point observed by SEM (magnification 50,000) before chemical attack and FIG. 4 represents the same point after attack in 12% HF at 25 ° C. We see that the chemical attack is isotropic and retains the truncated cone shape of the preformed tips. In addition, the complementary attack does not measurably affect, on this scale, the roughness of the surface. It even seems that certain irregularities are "erased". In the next step, the peripheral surface of the cone is made opaque.

This step is preferably carried out by depositing at least one metal layer.

All deposition techniques are possible as long as the deposition technique used does not affect the geometry of the end of the preformed tip. Preferably, the deposit is made under vacuum.

The nature of the metals used and the corresponding thickness deposited are variable and must be such that there is no passage for the light, and that any light leakage out of the opening formed by the generally planar face of the truncated cone is prevented. In general, the total thickness of the metallic layer (s) varies from

50 to 200 nm.

Preferably, a first layer based on chromium is deposited then a layer based on aluminum.

In this case, the chromium-based layer has a thickness of 2 to 20 nm and the aluminiiim-based layer a thickness of 50 to 180 nm.

Typically, evaporation is carried out under secondary vacuum from metallic filaments heated by the Joule effect.

In the evaporation chamber, the end of the tip is located at a suitable distance, for example 20 cm, above the metal filaments. So that the upper face of the truncated cone formed by the end of the point does not receive metal, i.e. so that the opening of the probe is preserved, the point is generally slightly inclined upwards, for example at an angle of 15 ° to the horizontal.

In order to obtain a homogeneous deposit, the point follows during evaporation, two rotational movements:

- a rotation around its axis;

- an azimuthal rotation around the vertical, that is to say along the axis carried by the mean direction of the evaporation flow.

The optical fiber obtained in the end is an optical fiber usable as a near field probe comprising a terminal portion having the general shape of a point whose end has the shape of a truncated cone delimited by an opaque peripheral surface and a generally planar face with a maximum dimension of 100 nm, defining an opening for the passage of light waves, said terminal portion retaining the polarization properties of propagating light waves. Figure 5 shows an inclined view of such a fiber, thinned, chemically etched and metallized as indicated above, magnified 20,000 times and Figure 6 shows a front view of the same fiber, magnified 103,920 times.

Figure 5 shows that the metallization treatment did not affect the overall geometry of the tip end. In Figure 6, the opening of the probe is observable.

Its diameter is of the order of 50 nm. We can clearly see, on the periphery of the opening, the deposit made which is in the form of compact islands of average size of 100 nm. An increase in the metal evaporation rates mentioned above can make it possible to obtain a more regular deposit.

The metallic layer obtained eliminates any light leakage from the opening.

Such a tip is ready for use in the printing process according to the invention. The stable relief structures are printed in layers of a material comprising a photosensitive compound and generally having a thickness of less than 50 nm.

A preferred photosensitive compound is a compound which undergoes isomerization under the effect of appropriate light waves. Preferably, this photosensitive compound is a compound undergoing cis / trans or trans / cis isomerization under the effect of light waves of appropriate polarization, in particular a photochromic compound comprising an azobenzene group.

The photosensitive compound is preferably attached by a covalent bond with the material of the layer in which the stable structure is printed in relief.

Preferably, the covalent bond is made via a chain with flexibilizing properties, and more particularly, the chain with fiexibilizing properties comprises at least one urethane group. The material particularly recommended for the layer in which the structure is printed comprises a polysiloxane matrix. The polysiloxane matrix is preferably obtained by the sol / gel route from a composition of hydrolysates of silane precursors.

Silane precursors are well known in the state of the art and comprise one or more hydrolysable functions such as Si-O-Alkyl, Si-Ci. A recommended class is constituted by alkoxysilanes, in particular epoxyalkoxysilanes such as γ-glycidoxypropyl trimethoxysilane, epoxycyclohexyl trimethoxysilane, as well as by the following alkoxysilanes:

3-aminopropyltriethoxysilane, 3-aminopropyltrimethoxysilane, 3- aminopropylmethyldiethoxysilane, methyltrimethoxysilane, ethyltriethoxysilane, vinyltriethoxysilane, phenyltriethoxysilane, dimethyldiethoxysilane, tetraox.

The alkoxysilanes which are particularly recommended are the tetraalkoxysilanes, and in particular the tetraethoxysilane (TEOS). These alkoxysilanes play the role of tetrafunctional crosslinking agents and allow the increase in the rigidity of the structure.

Since the photochromic molecules of the azobenzene family are sparingly soluble in sol / gel type matrices, a molecule is preferably used whose azobenzene group has been grafted onto an organic chain terminated by a silicon alkoxide. The polymerizable composition therefore preferably comprises a hydrolyzate of at least one precursor of formula (I):

A-Z-B in which

A represents an azobenzene group or a substituted derivative thereof, Z is a divalent hydrocarbon chain interrupted by at least one urethane group,

B is a group -If R _(3. _N) (X) _n , n varies from 2 to 3, in which X represents a hydrolysable group such

R and R, 'identical or different, denote an alkyl group of 1 to 4 carbon atoms, and a hydrolyzate of at least one precursor of formula R _(4. _a ) Si (X) _a in which a varies from 2 to 4 and X has the same meaning as above. Preferably, R 'denotes C ₂ H ₅ and a = 4 and n = 3.

The preferred azobenzene precursor has the formula:

and will be designated subsequently by Si-DRl. In order to inhibit the formation of aggregates and give freedom of movement to the photochromic compound, the composition can also comprise a hydrolyzate of a precursor of formula (II) A'-Z'-B 'in which:

A 'is a cyclic hydrocarbon group optionally comprising a heteroatom intracy clique

Z ′ is a divalent hydrocarbon chain comprising at least one urethane group,

B 'is a group of formula -Si R _(3-n) X _n , in which X, R and n have the same meaning as in formula (I) above. The preferred precursor preventing the formation of aggregates is a trialkoxysilane comprising a carbazole unit:

He will then be designated by Si-K.

The hydrolysis process for hydrolyzable silanes is conventional and known to those skilled in the art.

Reference may be made, for example, to patents FR 2,702,486, FR 2,704,581 or US 4,211,823.

One can, for example, mix the starting silanes and effect the hydrolysis of the mixture. Catalysts such as mineral or organic acids can be used.

It is also preferred to use, for hydrolysis, an amount of water at least stoichiometric, that is to say a molar amount of water corresponding to the number of moles of function capable of generating silanol groups. Organic co-solvents such as THF (tetrahydrofuran) can be used.

According to a recommended mode,

A. A complete hydrolysis is carried out of a mixture of silicon alkoxides, previously dissolved in an organic solvent for the alkoxides, with an acidic aqueous solution of pH equal to 1;

B. In order to better control the evolution of the soil during the condensation stage, and before the samples are formed, the acidity of the soil is reduced by partially neutralizing the acid necessary for the hydrolysis by adding a base; C. The soil is then deposited on a substrate by centrifugation technique. The film can then be dried at room temperature or subjected to a heat treatment, for example by heating from 50 to 120 ° C, for 5 to 30 minutes.

At the end of the deposition, a thin film is obtained ready to be used.

The impression of a stable structure is carried out by positioning, by means of an appropriate device, the opening of a fiber tip obtained according to the method described above, in the immediate vicinity of the surface of the layer comprising the photosensitive material, at a distance less than 100 nm, and preferably at a distance less than 50 nm, and better still less than 10 nm.

A light beam is then injected into the fiber tip and irradiated for a time necessary to obtain the local topographic modification. The local topographic modification results directly from the irradiation step and does not require an additional revelation step, as can be seen in certain documents of the prior art.

Typically, the general irradiation conditions are: output power, at the aperture at the tip end: 0.1 to lnW (nano att), the irradiation time varying from 1 to 30 seconds. One of the advantages of the method according to the invention is that it makes it possible to carry out a precise control of the height of the studs inscribed as a function of the irradiation time and / or of the power of the light waves emitted on the layer of material. comprising the photosensitive compound.

The invention also relates to a method of printing a stable figure represented by raised studs in a layer of a material comprising a photosensitive compound capable of inducing a topographic modification of said layer under the effect of light waves. emitted through at least one opening as defined above, which is characterized in that a) a first pad is printed according to the method described above, b) a controlled lateral movement of said opening is carried out, c) another pad is printed by operating as in step a) above, d) repeating operations b) and c) so as to cover with pads a surface corresponding to said figure.

Generally, the pads obtained have a height of at least 3 nm, preferably from 10 to 50 nm, and a lateral dimension substantially identical to the dimension of the opening and two consecutive pads are distant from less than 200 nm, preferably less than 100 nm, and better still, less than 50 nm.

To cover large areas more quickly, it is advantageous to use a mask, consisting of an opaque surface, generally parallel to the surface to be printed, and pierced with one or more openings having the characteristics described above. The invention has many advantages:

In particular, in addition to those already mentioned, the invention makes it possible to obtain networks of studs whose altitude fluctuation, in the registration area, remains less than their average height, which could not have been obtained in the networks. of the prior art.

The invention finds applications in the most diverse fields such as the storage of information on discs, fine markings, biotechnology, molecular electronics. A particularly interesting application is in the field of ophthalmic lenses.

It is indeed possible to produce microstructure networks, of no less than 200 nm, to produce anti-reflective coatings.

Such a network can be printed directly on the surface of the ophthalmic glass, or on the surface of a mold part and transferred during the molding of the glass.

Another possible application is the printing of distinctive marks, such as a logo, invisible to the naked eye, and making it possible to identify the origin of the glass.

Another possible application consists in carrying out discrete markings for the identification of stored glasses or during their manufacturing process. The following examples illustrate the present invention. EXAMPLE 1:

1) Manufacture of an Optical Fiber According to the Invention An optical fiber is manufactured comprising a terminal portion having the general shape of a point, the end of which has the shape of a truncated cone from a single mode optical fiber marketed by 3M and having the following characteristics:

- diameter of the fiber, including its protective plastic cover 250 micrometers, - diameter of the stripped fiber: 125 micrometers,

- core diameter: 3.7 micrometers, - profile of optical properties between core and cladding: index jump type,

- cut-off wavelength: 460 nm.

This fiber is subjected to a sequential hot stretching following the general process described above with the following parameters

- phase I

- stretching threshold Δlχ = 3 micrometers

- CO ₂ laser power: 14 watts

- diameter of the laser beam: about 2 mm - value of the reference pulse duration τ _ref : 50 ms

- value of the increment Δτi = 1 ms

- Phase II

- Δlc - 3 micrometers

- Cr = 1.3 ms - α = 0.45

- Phase III

- D3 = 15 micrometers

- Phase IV

- The stretching speed threshold is Ves = 0.002 ms ^"1 The sequential stretching device which has been used is described more particularly in the thesis of Philippe Bertrand of January 22, 1999, cited above.

Briefly, the optical fiber is held by pinching between two rails, which slide horizontally on ball cages. The ends of the rails are connected to each other by a wire which is guided by pulleys and on which hangs a free weight, or any means making it possible to ensure a fixed tension.

The device includes a laser, the power supply of which is adjusted by a TTL control, as well as a device for measuring the elongation of the fiber.

A microscope equipped with a camera makes it possible to visualize the fiber during drawing.

The characteristics of the preformed tip of the fiber obtained after sequential hot stretching are as follows: Diameter of the opening: 477 nm

Cone angle: 46 ° The tip is then cleaned for 3 minutes in a sulfochromic mixture placed in an ultrasonic tank, then rinsed for three minutes in distilled water (still in an ultrasonic tank).

It is then subjected to a chemical attack with hydrofluoric acid HF 12% by mass for a period of 2 min 30 seconds. Then it is rinsed in distilled water and dried with isopropanol.

A metallization step is then carried out consisting in depositing layers of chromium and aluminum by evaporation under secondary vacuum from metallic filaments heated by Joule effect on the periphery of the fiber tip.

The first layer is a 10 nm thick bonding layer of chromium deposited over the entire circumference of the tip. The chromium is evaporated with a speed of 0.23 nm / s. Still by evaporation, an aluminum layer 100 nm thick is then deposited at an evaporation rate of 1.82 nm / s.

The tip is kept rotating during the deposition of the layers

- azimuthal rotation speed: 1 revolution / sec

- axial rotation speed: 5 rpm

The optical fiber is then ready for use. The size of the opening of the fiber tip is of the order of 55 nm. 2}

Manufacture of photosensitive samples The sample is a thin layer deposited on a glass substrate. The layer is obtained by the sol / gel route. 2.1 - Preparation of the soil The starting reagents are the following:

- 0.207 g of Si-DRI, ie 3.68 x 10 ⁻⁴ mole;

- 0.307 g of Si-K, ie 7.37 x 10 ^' Vol;

- 1.2 cm ³ of THF, or tetrahydrofuran, organic solvent of the alkoxides used; - 0.041 cm ³ of TEOS, ie 1.84 x 10 ^"4 mole;

- 0.093 cm ³ , ie 5.15 x 10 ⁻³ mole, of aqueous solution at pH = 1, adjusted by addition of hydrochloric acid.

The Si-DRI compound, then the Si-K compound, which is in powder form, are weighed in a pill box. The solvent (THF) is then added to the open air, and the solution is homogenized by magnetic stirring. Then, the TEOS and the acidified water are added, and the mixture is replaced with magnetic stirring, protected from light for two hours.

At the end of the hydrolysis, the medium is neutralized by adding 0.0987 cm ³ of pyridine, ie 1.22 x 10 ^" mole

A In order to reduce the viscosity of the soil formed, 0.5 cm of solvent (THF) is incorporated.

Filtration of the soil is then carried out (pore diameter of the filter: 0.45 μm), then a surfactant at 2.5% by mass is added.

The soil is then ready to be deposited (mother solution of concentration denoted C ₀ ).

2.2 - Layers deposition:

The deposits are made by centrifugation on glass substrates (microscope slides previously cleaned), for a period of 20 s.

During the rotation, the condensation process continues, the solvent evaporates, and in the end a thin layer remains, the thickness of which depends on the speed of rotation and the concentration of the soil used. The thickness typically varies between 10 nm and approximately 1 μm; it is determined using a profilometer.

Adjustment of the thickness of the layer: A deposit of the mother solution with a speed of rotation of 3000 revolutions / min makes it possible to obtain a sample of approximately 1 μm in thickness.

In order to obtain layers of lower thicknesses, the mother solution is diluted and the speed of centrifugation is increased.

The table below gives the correspondence between the dilution of the mother solution and the thickness of the sample obtained for a rotation speed fixed at 6000 revolutions / min.

The thickness measurements are carried out after placing the freshly deposited sample for ten minutes in an oven at 100 ° C. to complete the condensation process and to evaporate the residual solvent.

The sample corresponding to E is used for the further implementation of the method of the invention.

3 - Inscription of stable structures in relief Device used The device used is shown in Figure 7.

An Argon-Krypton laser (1) is used with a wavelength of 568.2 nm (yellow Krypton line) generating a laser beam (3).

At the output of the laser (1) an optical density (2) or neutral filter G4, 2 mm thick, is inserted. The laser beam (3) crosses the optical density (2) then is reflected by a mirror (4), crosses a quarter-wave plate (5) which transforms the vertical rectilinear polarization coming from the laser (l) into a circular polarization, itself sent to a polarizing filter (6). The filter (6) is typically a Polaroid model HN22 polarizer.

At the output of the polarizer (6), the beam contacts a mirror (7) which directs it towards a focusing lens (8) (f = 10 mm). The beam will then be injected into an optical fiber (10), prepared during step 1 above and having one end (11) having a thinned tip and peripherally metallized:

At the lens outlet, the beam passes through a connector (9) corresponding to the unmodified end of the optical fiber (10).

The optical fiber (10) passes through a microscope head (12) containing two piezoelectric tubes, not shown, which make it possible to ensure the displacement, in 3 directions of space, of the end (11) of the optical fiber tip. thinned and peripherally metallized (10). Such a piezoelectric system is described precisely in the thesis of

Philippe Bertrand above.

The light beam from the opening, not shown, at the end (11) of the fiber tip then passes through the sample (13) prepared during step 2) above. Then the light beam is collected by a microscope objective (14) typically a Zeiss objective, “Epiplan” LD20x / 0.40 and detected by a photomultiplier (16), typically Hamamatsu R943 Ga-As 10 stages, equipped with a polarization analyzer (15). Registration of a network The sample is placed as shown in figure 7 at a distance of

5 nm from the fiber tip (11).

Beforehand, a first sweep is carried out to check the cleanliness of the surface of the layer on which the network will be written.

For this, the aperture at the end of the tip is kept at a distance of 5 nm and a scan is carried out in a plane.

A 2μm surface intended to be printed is then checked. Then, by acting on the piezoelectric system, the tip is positioned above a determined point on the swept surface (original position).

The tip is then moved in two directions x, y perpendicular, in a plane parallel to the surface of the layer, and maintaining the opening at a distance of 5 nm from the surface of the layer.

A first line of pads is printed by injecting, for two seconds for each pad, a light beam giving a peak output power, at the opening, of 170 pW. Between each light pulse, the point is moved along the x axis by a distance of 116 nm. After the printing of the first line, the point is again placed in its original position and one then carries out a displacement along the axis y of 136 mn, then one prints a new line of pads.

The operation is repeated until the network is obtained over the entire surface to be printed.

On the final network, the studs inscribed have an average height of 5 nm and a half-height width of 55 nm.

The registered network is shown in Figure 8.

80 plots are inscribed in a 1 μm zone, without slaving of the light power and it is noted that the altitude fluctuation is 4 nm for an average height of the plots of 5 nm.

The central stud has been deliberately forgotten so as to show that the inscription of all of the studs does not disturb the non-illuminated central zone. EXAMPLE 2:

A network of studs is printed according to the method described in example 1, with a displacement of 200 min from the tip, along the x axis, between each print, and increasing the irradiation time (from 5 to 35 seconds in 5 second steps) for a fixed average output power. FIG. 9 represents the sectional view of the network obtained, along one of the lines of the network of studs.

It can be seen that the height of the pads is directly proportional to the irradiation time.

Claims

1. A method of local printing of a stable structure in relief comprising a step of emitting light waves of appropriate polarization on a layer of a material comprising a photosensitive compound capable of inducing a topographic modification of said layer under the effect of said light waves, characterized in that said appropriate polarization light waves are emitted on said layer through an opening of maximum dimension less than or equal to 100 nm, peripherally delimited by an opaque zone preventing propagation of the light waves by outside said opening, said layer having a surface located at a distance less than or equal to 100 nm from said opening.

2. Method according to claim 1, characterized in that said opening has a dimension less than or equal to 60 nm.

3. Method according to claim 1 or 2 characterized in that said light waves of appropriate polarization emitted through said opening are transmitted by an optical fiber comprising a terminal portion having the general shape of a point at the end of which is located said opening.

4. Method according to claim 3, characterized in that the end of the tip has the shape of a truncated cone, delimited by a generally planar face and a peripheral surface.

5. Method according to claim 4, characterized in that said opening constitutes the generally planar face of the truncated cone.

6. Method according to claim 5, characterized in that the opaque zone is the peripheral surface of the cone.

7. Method according to claim 6, characterized in that the peripheral surface has been made opaque by depositing at least one metal layer.

8. Method according to claim 7, characterized in that the deposition is carried out under vacuum.

9. Method according to claim 7 or 8, characterized in that the total thickness of the (or) layer (s) metal (s) varies from 50 to 200 nm.

10. Method according to any one of claims 7 to 9 characterized in that one deposits a first layer based on chromium and then a layer based on aluminum.

11. Method according to claim 10, characterized in that the chromium-based layer has a thickness of 2 to 20 nm and the aluminum-based layer a thickness of 50 to 180 nm.

12. Method according to any one of claims 3 to 11, characterized in that the terminal portion of the optical fiber has the property of retaining the polarization characteristics of the propagating light waves.

13. Method according to claim 12, characterized in that the terminal portion of the optical fiber was obtained by sequential hot stretching.

14. Method according to claim 13, characterized in that, after the sequential hot stretching, the terminal portion of the optical fiber has undergone an isotropic chemical attack.

15. Method according to any one of the preceding claims, characterized in that said layer comprising the photosensitive material has a thickness of less than 50 nm.

16. Method according to any one of the preceding claims, characterized in that the surface of said layer is located at a distance less than or equal to 50 nm from said opening.

17. Method according to any one of the preceding claims, characterized in that the photosensitive compound is a compound undergoing cis / trans or trans / cis isomerization under the effect of light waves of appropriate polarization.

18. The method of claim 17, characterized in that the photosensitive compound is attached by a covalent bond with the material of the layer in which the stable structure is printed in relief.

19. The method of claim 18, characterized in that the covalent bond is made via a chain with flexibilizing properties.

20. The method of claim 19, characterized in that the chain with flexibilizing properties comprises at least one urethane group.

21. Method according to any one of claims 17 to 20, characterized in that the photosensitive compound is a compound comprising an azobenzene group.

22. Method according to any one of the preceding claims, characterized in that the material of the layer in which the structure is printed comprises a polysiloxane matrix.

23. The method of claim 22, characterized in that the polysiloxane matrix is obtained by sol / gel route from a composition of hydrolysates of silane precursors.

24. Method according to claim 23, characterized in that the polysiloxane matrix is obtained by polymerization of a composition comprising a hydrolyzate of at least one precursor of formula:

A-Z-B in which:

B is a group -SI R _(3. _N) (X) _n , n varies from 2 to 3 in which X represents a hydrolyzable group such

GOLD' ; OCR '; Cl

O

R and R ′, identical or different, denote an alkyl group of 1 to 4 carbon atoms, and a hydrolyzate of at least one precursor of formula R ₍₄ _ _a) Si (X) _a in which a varies from 2 to 4 and X has the same meaning as above.

25. The method of claim 24, wherein R 'denotes C ₂ H ₅ , a =

4 etn = 3.

26. Method according to any one of claims 23 to 25, characterized in that the composition comprises a hydrolyzate of a precursor of formula:

A'-Z'-B 'in which: A' is a cyclic hydrocarbon group optionally comprising an intracyclic heteroatom,

Z ′ is a divalent hydrocarbon chain comprising at least one urethane group,

B 'is a group of formula -SiR ^ ₃ . _a) X _n , in which X, R and n have the same meaning as in claim 24.

27. Method for printing a stable figure represented by raised studs in a layer of a material comprising a photosensitive compound capable of inducing a topographic modification of said layer under the effect of light waves emitted through '' at least one opening as defined in claim 1, characterized in that: 1) a first pad is printed by operating according to the method of claim 1 or 2,

2) a controlled lateral displacement of said opening is carried out,

3) another pad is printed by operating as in step 1 above, 4) operations 2 and 3 are reproduced so as to cover pads with a surface corresponding to said figure.

28. The method of claim 27, characterized in that the dimension of each of the pads is substantially identical to the dimension of the opening.

29. Method according to claim 27, characterized in that the distance between two pads is less than 200 nm, preferably less than 100 nm, and better still less than 50 nm.

30. Method according to any one of claims 27 to 29, characterized in that the studs have a height of at least 3 nm, preferably between 10 and 50 nm.

31. Method according to any one of claims 27 to 30, characterized in that the height of the pads is controlled by the irradiation time and / or the intensity of the light waves emitted on the material.

32. A method according to any one of claims 27 to 31 characterized in that the light waves are emitted through an opaque mask in which said openings are arranged and step 2) is eliminated.

33. A method of manufacturing an optical fiber usable as a near field probe, according to which:

1) a sequential hot stretching of the optical fiber is carried out in order to cause it to thin until it breaks and a fiber is obtained, the terminal portion of which has the general shape of a point, the tip of which end has the shape of a truncated cone delimited by a generally planar face and a peripheral surface,

2) an isotropic chemical attack on the fiber tip is carried out, characterized in that the peripheral surface of the truncated cone is made opaque, the generally flat face constituting an opening, for the passage of light waves, of maximum dimension of 100 nm.

34. Method according to claim 33, characterized in that the isotropic chemical attack on the fiber tip is carried out with an acid, preferably with hydrofluoric acid HF.

35. Method according to claim 33 or 34, characterized in that the peripheral surface of the truncated cone is made opaque by depositing at least one metallic layer.

36. Method according to claim 35, characterized in that the deposition of the metal layer (s) is carried out under vacuum, in the presence of a metallization source.

37. Method according to claim 34 or 35, characterized in that, during the deposition of at least one metallic layer, the optical fiber is oriented towards the metallization source so as to avoid the deposition of metal on the generally flat face of the truncated cone.

38. Method according to any one of claims 36 or 37, characterized in that the terminal portion of the optical fiber is subjected to a rotation around its axis and to a rotation around the axis carried by the mean direction of the flow evaporation.

39. Optical fiber usable as a near field probe comprising a terminal portion having the general shape of a point, the end of which has the shape of a truncated cone delimited by a peripheral surface and a generally planar face defining an opening for the passage light waves, said terminal portion retaining the polarization properties of propagating light waves, characterized in that the generally planar face has a maximum dimension of 100 min and said peripheral surface of the truncated cone is opaque.

40. Optical fiber according to claim 39, characterized in that the truncated cone has an apex angle varying from 30 to 70 °, preferably from 40 to 70 °.