WO1999062028A1

WO1999062028A1 - Method for making a portable electronic device comprising at least an integrated circuit chip

Info

Publication number: WO1999062028A1
Application number: PCT/FR1999/001232
Authority: WO
Inventors: Jean Christophe Fidalgo; Olivier Brunet
Original assignee: Gemplus
Priority date: 1998-05-27
Filing date: 1999-05-26
Publication date: 1999-12-02
Also published as: JP2002517047A; BR9910718A; EP1084481A1; FR2779255B1; CA2333431A1; CN1309796A; AU3832299A; FR2779255A1

Abstract

The invention concerns a method for making an electronic device, such as a chip card, comprising at least an integrated circuit chip (200) housed in a card body (100) cavity (120). The chip (200) is connected, via strip conductors (112), to interface elements (110). The cavity (120) has sloping walls. The strip conductors (112) and the interface elements (110) form a pattern which results from three-dimensional electrically conducting printing. The pattern extends from the card support surface (100) along the cavity (120) sloping walls up to the base thereof. Said method enables to increase production rate and reduce production cost.

Description

METHOD FOR MANUFACTURING A PORTABLE ELECTRONIC DEVICE COMPRISING AT LEAST ONE CIRCUIT CHIP

INTEGRATED

The present invention relates to the manufacture of a portable electronic device, comprising at least one integrated circuit chip which is embedded in a support and electrically connected to interface elements constituted by a connection terminal block and / or an antenna. These portable electronic devices constitute, for example, smart cards with and / or contactless, or even electronic labels. Contact and / or contactless smart cards are intended for carrying out various operations, such as, for example, banking operations, telephone communications, various identification operations, or teleticketing operations.

Contact cards include metallizations flush with the surface of the card, arranged at a precise location on the card body, defined by the usual standard ISO 7816. These metallizations are intended to come into contact with a read head of a reader for electrical data transmission.

Contactless cards, on the other hand, have an antenna that allows information to be exchanged with the outside world thanks to an electromagnetic coupling between the card's electronics and a receiving or reading device. This coupling is carried out in read mode or in read / write mode and the data transmission takes place by radio frequency or microwave.

There are also hybrid cards ("combicards") comprising both metallizations flush with the surface of the card and an antenna embedded in the card body. This type of card can therefore exchange data with the outside either in contact mode or in contactless mode. As they are currently produced, contactless cards are, like contact cards, thin portable objects whose dimensions are standardized. The usual ISO 7810 standard corresponds to a standard format card 85 mm long, 54 mm wide and 0.76 mm thick.

The majority of chip card manufacturing methods are based on the assembly of the integrated circuit chip in a subassembly called a micromodule which is then inserted using traditional methods.

A conventional method, illustrated in FIG. 1, consists in bonding an integrated circuit chip 20 by placing its active face with its contact pads 22 upwards, and by bonding the opposite face to a dielectric support sheet 28. The dielectric sheet 28 is itself arranged on a contact grid 24 of a metallic plate of nickel-plated and gilded copper. Connection wells 21 are formed in the dielectric sheet 28 and connection wires 26 connect the contact pads 22 of the chip 20 to the contact pads of the grid 24 via its connection wells 21. Finally, a encapsulation resin 30, based on epoxy, protects the chip 20 and the connection wires 26 welded. The module is then cut and then inserted into the cavity of a card body previously decorated.

This process has the disadvantage of being expensive, however. Indeed, since the micromodule is manufactured separately from the card holder, the steps for manufacturing a smart card are very numerous and contribute to increasing the manufacturing cost. In addition, the metal support plate itself remains a very expensive element because the connection, in particular by wires, requires metallization in nickel and gold.

The present invention therefore aims to eliminate the intermediate stages of manufacturing a micromodule so as to increase the yield and reduce the manufacturing cost.

Processes for manufacturing smart cards, without an intermediate step in producing a micromodule, have already been studied. A first solution, described in patent applications FR2671416, FR2671417 and FR 2671418, consists in inserting an integrated circuit chip directly into a card body. For this, the card holder is locally softened and the chip is pressed in the softened area. No cavity is therefore formed in the card body. A card obtained using this technology is shown diagrammatically in plan view in FIG. 2. The chip 20 is arranged so that its contact pads 22 are flush with the surface of the card 10. Screen printing operations then make it possible to print, on the same plane, contact pads 25 and conductive tracks 27 making it possible to connect the contact pads 25 to the contact pads 22 of the chip 20. A protective varnish is then applied to the chip 20. This first solution however has several disadvantages. First of all, since no cavity is made in the card holder, this method can only be adapted to very small chips. In addition, the operation of serigraphy of the contact pads 25 and of the interconnection tracks 27 is difficult to implement because the positioning of the tracks 27 on the contact pads 22 of the chip 20 requires very high indexing precision which must be checked by VAO ( Computer Aided Vision). This constraint affects the rate and efficiency of the manufacturing process.

The chip must also be perfectly positioned, so that its contact pads 22 are arranged parallel to the lateral edges of the card, in order to be able to produce the contact pads 25 parallel to the lateral edges of the card. However, the chip being placed in a locally softened area, it is not easy to position it correctly, and the chip cards whose contact pads are arranged slightly at an angle are intended for scrap.

This process is therefore too delicate to implement to be suitable for industrial production. In addition, the percentage of cards intended for scrap remains high, so that the manufacturing cost is very high.

Another solution having been considered uses the "Chrysalis" technology. This technology is based on the application of electrically conductive tracks by a MID type process ("Molded Interconnection Device" in Anglo-Saxon literature). Several processes associated with this technology have already been the subject of patent applications. Patent applications EP-A-0 753 827, EP-A-0 688 050, and EP-A- 0 688 051, in particular, describe methods of manufacturing and assembling an integrated circuit card. The card includes a housing for receiving the integrated circuit, and electrically conductive tracks are arranged against the bottom and the side walls of the housing and are connected to metallic contact pads formed on the surface of the card holder.

The application of the conductive tracks in the housing can be carried out in three different ways.

A first way is to carry out hot stamping. For this, a sheet comprising metallizations of copper, optionally covered with tin or nickel, and provided with a hot-activatable glue, is cut and then glued hot in the housing.

A second way consists in applying, by means of a pad, a lacquer containing a palladium catalyst, to the places intended to be metallized; heating the lacquer; then metallizing, by depositing copper and / or nickel, using an electrochemical process of autocatalysis.

A third way is to make a lithogravure from laser holograms. This lithography allows three-dimensional metallization deposits to be produced with very high precision and high resolution.

All of these metallization application methods are however complex to implement and therefore expensive. They often require the use of specific tools and several additional steps. However, the object of the invention being to omit stages in the manufacture of the micromodule, this is not to add others when applying the conductive tracks.

In addition, since the metallizations are made of copper and / or nickel, they remain expensive and contribute to increasing the cost price of the cards. The "Chrysalis" technology therefore uses processes that are too complex, costly and detrimental to manufacturing efficiency, to be suitable for large-scale industrial production. To overcome the aforementioned drawbacks, the invention provides a method of manufacturing a portable electronic device, such as a smart card, according to which the steps for manufacturing a micromodule are eliminated. For this, conductive tracks and interface elements are produced by printing, in three dimensions, a conductive substance. The chip is then connected to the interface elements, via the conductive tracks. The invention more particularly relates to a method of manufacturing an electronic device, such as a smart card, comprising at least one integrated circuit chip which is embedded in a card holder and which comprises connected contact pads , via conductive tracks, to interface elements constituted by a connection terminal block and / or an antenna, characterized in that it consists of:

- Produce, in the card holder, a cavity having inclined walls, - Produce a conductive ink impression, in three dimensions, to form a pattern comprising the interface elements and the conductive tracks, said pattern extending from the surface of the card holder, along the inclined walls of the cavity as far as the bottom of the latter,

- transfer and connect the chip to the bottom of the cavity,

- coat the chip in a protective resin. The shape of the inclined-wall cavity facilitates the deposition of conductive ink by a printing technique. In addition, the conductive ink has an advantageous cost compared to copper or nickel used in the case of deposition of metallizations. Furthermore, the interface elements being printed, they have a negligible thickness.

The method according to the invention also has the advantage of being rapid and inexpensive. This advantage is notably due to the fact that the interface elements as well as the conductive tracks are formed in a single step consisting of a simple technique of printing conductive ink.

Other features and advantages of the invention will appear on reading the description given by way of illustrative and nonlimiting example, and made with reference to the appended figures which represent:

FIG. 1, already described, a diagram in cross section which illustrates a traditional method of manufacturing a smart card with contacts,

- Figure 2, already described, a diagram in top view, of a smart card manufactured according to a known technology, - Figure 3, a diagram of a smart card with contacts obtained according to a method of the invention ;

FIGS. 4A and 4B, respectively a top view and a sectional view of a smart card with contacts according to the invention, in which the chip is transferred in a "flip chip" arrangement,

FIGS. 5A and 5B, respectively a top view and a sectional view of a smart card with contacts according to the invention, in which the chip is transferred according to another type of assembly, - Figures 5C and 5D, two top views of a smart card with contacts according to the invention, in which the chip is respectively transferred and then connected according to another type of assembly, - Figures 6A and 6B, respectively a view from above of two electronic labels during their manufacture,

- Figures 7A and 7B, a top view of two hybrid cards, produced according to the method of the invention.

Figure 3 shows schematically a smart card with contacts obtained according to an embodiment of a method according to the invention. The card body, referenced 100, is obtained according to a conventional manufacturing process, for example by injecting plastic material into a mold. This card support 100 comprises, at a location defined by the ISO standard, an interface element constituted, in the example of FIG. 3, by a terminal block of connections 110 provided with contact pads 111 flush with the surface. These contact pads 111 are positioned around a cavity 120 formed in the card body. This cavity is made either by milling or during the injection of the card, which is more economical. It is preferably circular and has inclined walls. However, it can just as well be rectangular, lozenge or octagonal etc. Conductive tracks 112, attached to the contact pads 111, also line the bottom and the walls of the cavity 120.

In fact, the contact pads 111 and the conductive tracks 112 form a single pattern obtained in a single step, by printing, in three dimensions, conductive ink. Thus, the contact pads 111 consist of conductive ink, deposited by printing on the surface 100 of the card, and are extended by conductive tracks 112, along the inclined walls of the cavity, as far as the bottom of the latter. this .

The inclined shape of the cavity 120 is important because it facilitates the printing of conductive ink. In the example shown diagrammatically in FIG. 3, the cavity has two planes: the first plane is horizontal and defined inside a first circle 121 forming the bottom of the cavity; the second plane, forming the walls of the cavity 120, is inclined and defined inside a second circle 122. The depth of the cavity must be sufficiently small to facilitate the printing of the pattern. Thus, it is preferably between 100 and 600 μm, for example of the order of 300 μm.

The cavity can however be of any other shape, for example rectangular, lozenge or octagonal.

An integrated circuit chip 200 is transferred to the bottom of the cavity 120 and connected, via the conductive tracks 112, to the contact pads 111.

The three-dimensional printing of conductive ink to form the interface element 110 and the conductive tracks 112 can be carried out using different techniques. In a first embodiment, the printing of conductive ink is obtained by a pad printing technique. For this, an ink pad allows to transfer the conductive ink, according to the desired pattern, on the surface of the card and in the cavity. Preferably, the tampon is made of deformable material, for example silicone material, in order to adapt to the shape of the cavity. In fact, the shape and the material of the stamp are defined as a function not only of the shape of the cavity but also of the desired resolution for the pattern to be printed.

This technique can be implemented either with a pad moving vertically towards the card, or with a rotating pad. In a second embodiment, the printing of conductive ink is obtained by an offset printing technique using a compressible roller and of low hardness of the blanket type for the transfer of the ink onto the card. Except for the constraints on the blanket type roller, the rest of the printing parameters are similar to conventional printing techniques, that is to say the use of an inkwell, of a polymer or metallic plate, comprising the pattern to be printed hollow or embossed, and an ink transfer roller.

In both pad printing and offset printing techniques, the depth of the cavity should not be too great compared to the flexibility of the roller or pad used. Typically, the depth of the cavity is between 100 μm and 600 μm.

A third embodiment, for printing conductive ink in three dimensions, consists in using an ink jet printing technique. Traditionally, the inkjet printing technique can be carried out in two different and well-known ways: either by a so-called drop-on-demand method, or by deviated continuous inkjet. This last inkjet printing technique consists in projecting drops charged with static electricity along a defined trajectory. During printing, the trajectory of these drops can be modified by applying a different polarization to deflection plates.

In order to be able to produce a good quality three-dimensional printing and correctly print the pattern comprising interface elements and conductive tracks, the cavity must not have a plane close to the vertical, but only horizontal or inclined planes according to a tilt angle between 5 and 30 °, preferably between 15 and 20 °.

Thanks to these three-dimensional conductive ink printing techniques, it is possible to print, in a single step, both interface elements constituted by a connection terminal block and / or an antenna, and tracks conductive intended to allow the connection of the chip. In this case, the interface elements and the conductive tracks form a single pattern.

Different printing techniques allow different kinds of conductive ink to be used. Thus, the conductive ink can be constituted by a solvent ink, comprising a polymer resin dissolved in a solvent with conductive fillers (metal particles), which hardens by evaporation of the solvent. The ink can also be a one-component or two-component thermosetting ink, an UV polymerization ink, a solder paste type compound or a metal alloy.

The chip can, in turn, be transferred to the bottom of the cavity according to three different types of mounting. A first method consists in transferring the chip according to a “flip chip” type of assembly. This type of assembly is already well known and is shown in the diagrams in top view and in section of FIGS. 4A and 4B. In FIG. 4B, the contact pads 111 of the connection terminal block 110 and the conductive tracks 112 are represented by a thick black line to facilitate understanding. However, since they are obtained by printing conductive ink, their thickness is in reality negligible. The chip is transferred by turning it over, the active face comprising the contact pads 220 oriented towards the bottom of the cavity 120. The chip 200 is then connected by applying its contact pads 220 to the conductive pads 112 previously printed, without the use of conductive wires. In this case the interconnection tracks 112 must be printed with precision and they are brought to the exact location of the contact pads 220 of the integrated circuit chip 200.

In the example illustrated in FIG. 4B, the chip 200 is connected to the conductive tracks 112 by means of an adhesive 350 with anisotropic electrical conduction which is well known and often used for mounting passive components on the surface. This glue 350 in fact contains elastically deformable conductive particles, which make it possible to establish an electrical conduction along the z axis (that is to say along the thickness) when they are pressed between the contact pads 220 and the tracks. conductive 112, while ensuring insulation in the other directions

( ^χ , y) •

In an alternative embodiment, the electrical connection can be established by means of protrusions formed by a conductive adhesive, previously deposited on the contact pads 220 of the chip and reactivated hot when the chip is transferred.

Another way of establishing the electrical connection between the chip 200 and the conductive tracks 112 consists in making, on the contact pads 220 of the chip 200, bosses in conductive material, intended to improve the electrical contact, then in applying the chip on the previously printed pattern, before complete polymerization of the conductive ink used for printing the pattern. The fixing and the connection of the chip are then carried out simultaneously, during the polymerization of the conductive ink of the printed pattern. Finally, in the case where the conductive tracks 112 are produced by printing by ink jet, of a metal alloy, it is possible to fix and connect the chip in a single welding step. For this, bosses of metal alloy with a low melting point are produced on the contact pads 220 of the chip 200 and are remelted when the chip is transferred in order to weld them to the conductive tracks 112.

A final step in the manufacture of the chip card with flush contacts illustrated in FIG. 4B then consists in coating the chip with a protective resin 300. For this, a drop of resin is deposited in the cavity 120. In addition, to obtain a flat external surface, a resin of very low viscosity is preferably used. Furthermore, when a conductive adhesive is used for transferring the chip, the protective resin must be chosen so that it is compatible with this adhesive. A second method for carrying out the transfer of the chip consists in sticking the chip in the place with its active face, comprising the contact pads, facing upwards, that is to say towards the opening of the cavity 120 This type of assembly is illustrated by FIGS. 5A and 5B which show respectively a top view and a sectional view of a smart card with flush contact.

In this case, the interconnection tracks 112 are brought close to the location provided for the chip 200. The chip 200 is bonded to the bottom of the cavity 120, by the face opposite to the active face, using an adhesive. 500 insulating. The adhesive 500 used can for example be a crosslinking adhesive under the effect of exposure to ultraviolet radiation. The rate of this bonding operation can be particularly high, since it is possible, for example, to bond five to six thousand chips per hour.

In a second step, the electrical connections are made between the contact pads 220 of the chip 200 and the conductive tracks 112. These connections are made by dispensing a conductive resin 400 on the contact pads 220 of the chip and the tracks 112. The conductive resin 400 may for example be a polymerizable adhesive loaded with conductive particles such as silver particles. This second connection step can be carried out at the same high rate as the bonding step of the chip. In addition, these two bonding and connection steps can be carried out using the same equipment.

In the same manner as previously described, the chip 200 is then coated with a protective resin 300 which is deposited in the cavity 120 and is flush with the surface of the card support 100. this encapsulation resin thus makes it possible to protect the integrated circuit chip from climatic and mechanical constraints. It must also be compatible with the insulating adhesive 500 and with the conductive resin 400 used.

FIGS. 5A and 5B which have just been described show diagrammatically a configuration for which each contact pad is located opposite a pad of the chip which is associated with it. On the other hand, when the chip is mounted according to a third method consisting of conventional wired cabling, it will be necessary to use an interdigitated pattern as shown diagrammatically in FIGS. 5C and 5D and as described in patent application EP-A-0 753 827. This interdigitated pattern thus makes it possible to bring the connection tracks 112 of each contact pad 111, associated with a contact pad 220 of the chip 200, close to this pad, and thus to avoid entanglement of the connection wires 260 FIG. 5C represents more particularly the interdigitated pattern on which a chip 200 is transferred. FIG. 5D also represents the wire connections 260 between the connection tracks 112 and the contact pads of the chip.

The invention also applies to the manufacture of contactless smart cards. In this case, the interface elements consist of an antenna, the turns of which can be printed on the surface of the card and / or in the cavity. However, whatever the location of the turns, the ends of the antenna must always be positioned in the bottom of the cavity, in order to be able to connect them to the contact pads of the chip. FIGS. 6A and 6B diagrammatically show two electronic labels seen from above, during their manufacture. These two labels can possibly be used as a basis for the manufacture of contactless smart cards or else be used as such. They are referenced 100. They comprise a cavity 120 and an antenna 140. The antenna 140 is obtained by printing conductive ink using one of the printing techniques mentioned above, namely pad printing, offset printing or the inkjet.

The label of Figure 6A has an antenna

140, the turns of which are made both on the surface of the card and in the cavity 120. The conductive tracks associated with this interface element are formed by the antenna ends 141, 142, and terminate in the bottom of the cavity. A chip, not shown in this figure, is then transferred to the bottom of the cavity and connected to the ends 141, 142 of the antenna. The transfer of the chip can be done in the two ways described above. However, in the case of a "flip chip" assembly, it is preferable to avoid the use of an anisotropic conductive adhesive in order to avoid short circuits liable to occur due to the presence of the antenna turns in the bottom. of the cavity.

In this case, and more particularly in the case of the diagram of FIG. 6B described below, insofar as there are only two contacts to be connected, it will be possible to envisage making the connection by depositing two small drops of conductive glue then positioning the chip face down. Before transferring the chip, we also prefer protect the antenna turns located at the bottom of the cavity by covering them with an insulating varnish.

The antenna 140 of the label shown in Figure 6B differs from the antenna shown in Figure 6A in that the turns are fully printed on the surface of the card holder 100 and only the antenna ends 141, 142 , forming conductive tracks associated with the antenna, terminate in the bottom of the cavity. This embodiment facilitates the transfer of the chip into the bottom of the cavity. On the other hand, in this case the antenna turns overlap at at least one point C of the surface of the card. It is therefore necessary to apply an insulating varnish to this overlap point (s) in order to avoid the appearance of a short circuit.

When the antenna is thus produced on the surface of the card, a subsequent step consists in applying an insulating protective varnish to its turns.

It is also possible to drown the antenna, by applying another plastic sheet to the printed surface of the card, in order to produce a contactless smart card.

An encapsulation resin is also deposited in the cavity 120 in order to protect the chip, and the antenna when the latter is located in the cavity.

A hybrid card can also be produced in accordance with the method according to the invention. Figures 7A and 7B show schematically such a card. In this case, the interface elements consist of a connection terminal 110 and an antenna 140. The printed pattern, by printing of conductive ink, comprises on the one hand the connection terminal 110 which is extended by conductive tracks 112 ending in the bottom of the cavity, and on the other hand the antenna 140 whose ends 141,

142 at least terminate in the bottom of the cavity 120. A chip is then transferred to the bottom of the cavity 120 so that its contact pads are connected on the one hand to the ends 141, 142 of the antenna and on the other share to the conductive tracks 112 associated with the contact pads 111 of the connection terminal block 110.

For reasons of clarity, the chip is not shown in FIGS. 7A and 7B. FIG. 7A illustrates a preferred embodiment, according to which the antenna 140 is entirely produced in the cavity 120 so that only the contact pads 111 of the connection terminal block 110 are visible on the surface of the card support 100. In this case, the antenna tracks overlap and an insulating varnish

143 is deposited on the overlapping points to avoid the appearance of a short circuit.

However, it is of course conceivable to produce both the antenna turns and the connection terminal block on the surface of the card body, as shown diagrammatically in FIG. 7B. In this case, only the antenna ends 141, 142 terminate in the bottom of the cavity 120. An insulating varnish 143 is also deposited on the overlapping points of the antenna turns to avoid the appearance of a short circuit. .

In the example of FIG. 7A, it is preferable to protect the antenna turns 140 with an insulating varnish, before the transfer of the chip, in order to avoid the appearance of short circuits. The chip can be transferred using the different methods described above. However, we prefer to postpone it according to the second method, i.e. the active side facing upwards and the contact pads connected by means of a resin. conductive, like silver glue for example. This mode of transfer indeed requires less precision concerning the position of the conductive tracks formed by the ends 141, 142 of the antenna, and of the conductive tracks 112 associated with the terminal block 110, relative to the contact pads of the chip; and it makes it possible to avoid possible short circuits by bonding, in the bottom of the cavity and on the antenna turns 140, the inactive face of the chip by means of an insulating adhesive. Thanks to the method according to the invention, it is possible to manufacture smart cards in large mass because the production rate is considerably increased. Indeed, the intermediate steps of manufacturing a micromodule are eliminated and the realization of the connection terminal block, and / or the antenna, and the interconnection tracks is done in a single and same step consisting of an impression of conductive ink. This results in a significant reduction in the cost price.

In addition, the method according to the invention is inexpensive because the conductive ink is less expensive than copper, nickel and gold which are used in conventional methods for producing metallizations and connections.

In addition, the invention does not use expensive equipment which reduces manufacturing costs.

On the other hand, the cavity being produced at a sufficiently shallow depth to allow a printing of good quality conductive ink on the inclined walls and on the bottom (typically on a depth less than 400 μm), it remains on the back of the chip, that is to say in the lower part of the card located under the cavity, a greater amount of material than in traditional smart cards. The thickness remaining under the cavity is indeed between 350 and 500 μm. This remaining thickness makes it possible to considerably reduce the risks of formation of cracks likely to occur. The mechanical strength of the chip in the card body is therefore improved. In addition, this geometry is extremely easy to manufacture by injection molding with a fixed core of simple design.

Claims

1. A method of manufacturing an electronic device, such as a smart card, comprising at least one integrated circuit chip (200) which is embedded in a card holder (100) and which comprises contact pads (220 ) connected, via conductive tracks (112; 141, 142), to interface elements constituted by a connection terminal block (110) and / or an antenna (140), characterized in that it consists of : - make, in the card holder (100), a cavity (120) having inclined walls,

- printing conductive ink, in three dimensions, to form a pattern comprising the interface elements (110; 140) and the conductive tracks (112; 141, 142), said pattern extending from the surface of the support card (100), along the inclined walls of the cavity (120) as far as the bottom of the latter,

- transfer and connect the chip (200) to the bottom of the cavity (120), and

- coat the chip (200) in a protective resin (300).

2. A method of manufacturing a smart card with flush contacts according to claim 1, characterized in that the interface elements consist of a connection terminal block (110) which is printed on the surface of the card support (100 ) and is extended by the conductive tracks (112) which terminate in the bottom of the cavity (120).

3. Method of manufacturing a contactless smart card or an electronic label according to claim 1, characterized in that the interface elements consist of an antenna (140), the turns of which are printed on the surface from the card holder (100) and / or in the cavity (120), and in that the conductive tracks (141, 142) are formed by the antenna ends and terminate in the bottom of the cavity.

4. A method of manufacturing a hybrid smart card according to claim 1, characterized in that the interface elements consist firstly of a connection terminal block (110), which is printed on the surface of the support. card (100) and is extended by conductive tracks (112) terminating in the bottom of the cavity (120), and on the other hand by an antenna (140) whose ends (141, 142) at least terminate in the bottom of the cavity (120).

5. Method according to one of claims 1 to 4, characterized in that the cavity (120) is produced over a depth between 100 and 600 .mu.m.

6. Method according to one of claims 1 to 5, characterized in that the inclined walls of the cavity (120) are produced at an angle of inclination between 5 and 30 °.

7. Method according to one of claims 1 to 6, characterized in that the printing of conductive ink is carried out by pad printing, using a deformable ink pad capable of adapting to the shape of the cavity.

8. Method according to claim 7, characterized in that the pad used is a pad with vertical or rotary movement.

9. Method according to one of claims 1 to 6, characterized in that the printing of conductive ink is carried out by an offset technique using a blanket-type roller for the transfer of the ink onto the card support, the roller being deformable to adapt to the shape of the cavity (120).

10. Method according to one of claims 1 to 6, characterized in that the printing of conductive ink is carried out by ink jet.

11. Method according to one of claims 7 to 10, characterized in that the conductive ink used is a solvent ink, a bicomponent thermosetting ink, a poly erisable ink under ultraviolet radiation, a solder paste or a metal alloy.

12. Method according to claim 1, characterized in that the chip (200) is transferred to the bottom of the cavity (120) in a "flip chip" arrangement.

13. Method according to claim 1, characterized in that the chip (200) is transferred to the bottom of the cavity (120), by bonding of the face opposite to the active face by means of an insulating adhesive (500), and connected by means of a conductive resin (400) dispensed at the same time on the contact pads (220) of the chip (200) and on the conductive tracks (112; 141, 142).

14. Method according to claim 3, characterized in that the chip (200) is transferred to the active face turned downwards in the bottom of the cavity (120), after having deposited two small drops of conductive adhesive on the antenna ends (141, 142).