Field effect transistor with an atomically thin channel

Abstract

Production of a transistor, the channel structure of which comprises at least one finned channel structure, the method comprising: forming, from a substrate (1), a molding block (3), forming, on the molding block, a thin layer (7) made from a given semiconductor or semi-metallic material, and consisting of one to ten atomic or molecular monolayers of two-dimensional crystal, withdrawing the molding block while retaining a portion (7a) of the thin layer extending against a lateral face of the molding block, said retained portion (7a) forming a fin that is capable of forming a channel structure of the transistor, producing a coating gate electrode against said fin.

Claims

1. A method for producing a transistor wherein a channel structure includes one or more fin(s), the method comprising the steps of: forming on a substrate one or more blocks referred to as moulding blocks, forming on said one or more moulding blocks a thin layer based on a given semiconductor or semi-metallic material, and consisting of one to ten atomic or molecular monolayer(s) of two-dimensional crystal, then, forming one or more holding blocks respectively at one or more ends of the moulding blocks, then forming a masking covering the moulding blocks and the holding blocks, partially removing the masking, a thickness of the masking being removed so as to reveal a top portion of the moulding blocks and the holding blocks, and removing said thin layer at the level of the top portion of the moulding blocks, removing said one or more moulding blocks while retaining one or more portion(s) of the thin layer extending against at least one lateral face of said moulding blocks, said one or more retained portions forming one or more fin(s) for forming a channel structure of the transistor, the removal of said moulding blocks being carried out by selective etching with respect to the material of said one or more holding blocks, said one or more holding blocks being for ensuring mechanical strength of said one or more fin(s) during the removal of said one or more moulding blocks, then removing the rest of the masking, then producing a gate electrode against said one or more fin(s).

2. The method according to claim 1, wherein the thin layer is formed by catalytic growth, said one or more moulding blocks being based on a catalytic material for the growth of said given material.

3. The method according to claim 1, wherein the holding blocks are source and drain electrodes.

4. The method according to claim 3, wherein the source and drain electrodes are based on a conductive material.

5. The method according to claim 1, wherein said one or more moulding blocks are made of a dielectric material.

6. The method according to claim 5, wherein said dielectric material is SiO.sub.2.

7. The method according to claim 5, wherein the given material of the thin layer is MoS.sub.2, WS.sub.2, WSe.sub.2, or MoSe.sub.2.

8. The method according to claim 1, wherein the given material of the thin layer is graphene or a metal chalcogenide.

9. The method according to claim 1, wherein the gate is a surrounding gate produced by forming a gate stack comprising at least one layer of gate dielectric layer and at least one layer of gate material, the method further comprising localised removal of one or more zones of the surrounding gate so as to reveal one or more given regions of said one or more fins.

10. The method according to claim 9, wherein after the localised removal so as to remove zones of the gate stack, at least one capture layer based on a material suitable for absorbing and/or adsorbing at least one chemical species may be formed on the revealed given regions of said one or more fins.

11. The method according to claim 1, wherein said partial removal of a thickness of the masking and of said thin layer at the level of said top portion of the moulding blocks is performed using plasma etching.

12. The method according to claim 1, wherein the gate is a surrounding gate.

13. A method for fabricating a chemical or biological sensor provided with at least one transistor comprising the production of the transistor according to the method according to claim 1.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) The present invention will be understood more clearly on reading the description of embodiment examples given merely by way of indication and not limitation with reference to the appended drawings wherein:

(2) FIGS. 1A-1G and 2A-2F serve to illustrate an example of a method for producing a transistor with fins of very reduced dimensions and which may attain that of an atom or a molecule;

(3) FIGS. 3A-3B serve to illustrate an alternative embodiment of a method for producing the fin transistor;

(4) FIG. 4 illustrates an example of a finFET transistor with fin of very reduced dimension and as used according to an embodiment of the present invention;

(5) FIG. 5 illustrates an example of fin transistor as used according to an embodiment of the present invention and further provided with a capture layer for detecting gas or biological elements;

(6) FIG. 6 illustrates a particular example of a transistor with orthogonal fins in relation to one another and joining together in at least one contact point;

(7) The different parts represented in the figures are not necessarily represented according to a uniform scale, in order to render the figures more legible.

(8) The different options (alternative embodiments and embodiments) should be understood as not being mutually exclusive and may be combined with one another.

(9) Furthermore, in the description hereinafter, terms which are dependent on the orientation of the structure such as “vertical”, “horizontal”, “bottom”, “lateral”, apply considering the structure as oriented as illustrated in the figures.

DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS

(10) An example of a method for producing a transistor, in particular of the finFET type, wherein the channel structure is formed from one or more fins of nanometric size and as used according to an embodiment of the present invention will not be described with reference to FIGS. 1A-1G (giving a transversal cross-sectional view) and FIGS. 2A-2F (wherein the structure being produced is represented as a perspective view).

(11) A bulk semiconductor substrate, for example made of silicon, may be envisaged as the starting material. Alternatively, it is also possible to start from a semiconductor-on-insulator type substrate 1 formed from a semiconductor base layer, for example silicon, the base layer being covered with an insulating layer made of dielectric material, typically silicon oxide, in turn covered with a superficial semiconductor layer.

(12) The substrate 1 may be provided with alignment marks, for example formed by etching parts not protected by a photosensitive resin mask or indeed by metal deposition followed by etching.

(13) On the substrate 1, one or more blocks 3, referred to as molding blocks (FIG. 1A), are produced, which, in the particular example illustrated in FIG. 2A, are long blocks of parallelepipedal shape. In this example, several molding blocks 3 are disposed parallel with one another, according to a predetermined interval, which may be for example between 10 nm and 100 μm.

(14) The blocks 3 have a critical dimension Dc1 typically between 30 nm and 100 μm. The critical dimension Dc1 may be less than 30 nm, this dimension generally corresponding to a limit dimension imposed by the lithographic methods currently available, in particular those using an electron beam (e-beam).

(15) “Critical dimension” denotes the smallest dimension of a pattern measured parallel with the principal plane of the substrate 1. The “principal plane” is defined as a plane passing through the substrate 1 and which is parallel with the plane [O;x;y] of an orthogonal reference [O;x;y; z] given in the figures.

(16) The molding blocks 3 may be formed for example by deposition of a layer based on a material 4 followed by photolithography and etching of this material 4 in order to define patterns.

(17) Alternatively, to produce these blocks 3, openings are formed in a masking layer, in particular a photosensitive resin masking, then the masking openings are filled with a material 4. The masking is removed typically by means of a stripping method commonly referred to as “lift-off” when this masking is made of resin.

(18) A thin layer 7 based on semiconductor material or semi-metal is then formed, preferably of Van der Waals or two-dimensional (2D) type, from which it is envisaged to produce the channel structure of the transistor. The thin layer 7 of nanometric thickness is produced by growth on the molding blocks 3. “Nanometric” denotes a thickness of less than 10 nanometres.

(19) Typically, the thin layer 7 has a thickness of less than 3 nm.

(20) The thin layer 7 particularly consists of 1 to 10 atomic monolayer(s) of a crystalline two-dimensional material (2D) such as graphene or of 1 to 10 molecular monolayer(s) of a two-dimensional material (2D) of a material based on chalcogenides and transition metals.

(21) “Atomic monolayer” denotes a lamina, otherwise referred to as a layer, wherein the thickness consists of a single atom. “Molecular monolayer” denotes a lamina, otherwise referred to as a layer, wherein the thickness consists of a single molecule, this layer being formed from a repeated pattern of molecules bonded together by means of covalent bonds.

(22) Preferably, the thin layer includes at most 5 atomic or molecular monolayers of crystalline two-dimensional material, in particular of a two-dimensional semiconductor also referred to as 2D semiconductor.

(23) Advantageously, the thin layer 7 deposited on the blocks 3 is an atomic monolayer or a molecular monolayer.

(24) For example, the thin layer 7 may be formed from an atomic monolayer made of graphene, i.e. a two-dimensional lamina, composed of carbon atoms with a thickness of a single carbon atom.

(25) According to a further example, the thin layer 7 may be formed from a molecular monolayer of transition metal dichalcogenides, i.e. a two-dimensional lamina of MoS.sub.2 or WS.sub.2 or WSe.sub.2 or MoSe.sub.2 molecules, forming a 2D crystal, with a thickness corresponding to a single molecule.

(26) The growth of the thin layer 7 is carried out for example by CVD (Chemical Vapour Deposition). For example, when a graphene layer is formed, the growth may be carried out at a temperature between 900° C. and 1,200° C. Alternatively, the thin layer 7 may be formed by PVD (Physical Vapor Deposition).

(27) In the example illustrated in FIGS. 1B and 2B, the thin layer covers the molding blocks 3 and extends both on a top face 3a, and on lateral faces 3b, 3c of the blocks 3. According to an alternative embodiment (not shown), it is also possible to envisage producing the thin layer 7 only on a portion of the blocks 3, in particular on a lateral face of the blocks 3 by envisaging to mask further zones of these molding blocks 3.

(28) Preferably, the molding blocks 3 are envisaged in a material 4 suitable for serving as a catalyst for the growth of the thin layer 7.

(29) For example, in the case where it is sought to grow a thin layer 7 of graphene, copper or platinum blocks 3 may be envisaged. According to a further example, blocks 3 made of silicon oxide (SiO.sub.2) or gold or indeed sapphire may be envisaged when it is sought to form a thin layer 7 of MoS.sub.2 or WS.sub.2. The material 4 of the blocks 3 may be SiO.sub.2 when it is sought to grow a thin layer of WSe.sub.2 or MoSe.sub.2.

(30) The choice of the material 4 of the blocks 3 may also be governed by that of the material based on which further elements referred to as “holding blocks” are envisaged and which are intended to support the thin layer 7 following a subsequent at least partial removal of the molding blocks 3.

(31) Advantageously, the holding blocks 9, 10 correspond to the source and drain electrodes. Thus, a material 4 which can serve as a catalyst for the thin layer 7 and can be removed selectively in relation to that or those envisaged to form the source and drain electrodes is preferably chosen.

(32) In the embodiment example illustrated in FIG. 2C, source 9 and drain blocks 10 are formed, serving as holding blocks at the ends of the molding blocks 3. The source and drain blocks 9, 10 join herein the ends of several molding blocks 3.

(33) The source and drain blocks 9, 10 are based on a conductive material 8 which, as described above, is suitable for withstanding selective etching of the molding blocks 3. For example, in the case where the molding blocks 3 are based on copper, or sapphire or SiO.sub.2, it may be envisaged to produce the blocks 9, 10 based on gold or palladium. According to a further example, molding blocks 3 based on platinum or gold may be formed whereas the blocks 9, 10 are based on titanium.

(34) In FIGS. 1C and 2D, a subsequent step of forming a masking 13 is represented. The masking 13 is typically a resin masking and may be produced by spin coating. Typically, the masking is deposited so as to cover the molding blocks 3 as well as the source and drain blocks 9, 10.

(35) A masking 13 thickness is then removed.

(36) The partial removal of the masking 13 is carried out for example using plasma etching, in particular of the ICP (Inductively Coupled Plasma) or RIE (Reactive Ion Etching) type when the masking is a resin.

(37) In the example illustrated in FIGS. 1D and 2E, this removal is carried out so as to remove once again a top portion of the molding blocks 3, the thin layer 7 being also removed at the level of this portion of the molding blocks 3. The material 4 forming the molding blocks 3 is thus revealed.

(38) A removal of the mould blocks 3 is then performed as illustrated in FIG. 1E. The molding blocks 3 are in this embodiment example sacrificial blocks intended to be entirely removed. This removal is carried out by selective etching of the material 4 with respect to the material 8 of the holding blocks 9, 10, herein forming source and drain electrodes. The etching of the blocks 3 is typically carried out by dipping the structure in an etching solution.

(39) Examples of etching solutions suitable for use are listed in a table given hereinafter, the solutions being associated respectively with different examples of catalyst materials 4, as well as materials 8 of the blocks 9, 10 suitable for withstanding selective etching of the molding blocks 3.

(40) TABLE-US-00001 Material of Graphene MoS.sub.2 WS.sub.2 WSe.sub.2 MoSe.sub.2 the thin layer 7 Catalyst Copper (Cu) Platinum SiO.sub.2 Gold Sapphire SiO.sub.2 SiO.sub.2 material 4 Etching Iron chloride Ammonium Aqua regia Hydrofluoric Aqua regia Phosphoric Sodium HF HF solution of persulfate acid acid hydroxide material 4 (NH.sub.4).sub.2S.sub.2O.sub.8 (HF) (H.sub.3PO.sub.4) (NaOH) Material of Gold (Au) Gold Titanium Gold Titanium Gold Gold Gold Gold the source Platinum (Pt) Platinum Palladium Palladium Palladium Palladium Palladium and drain (Pd) blocks 9, 10

(41) Thus, according to a particular embodiment example, when platinum molding blocks 3 are envisaged to grow a thin layer 7 of graphene, the etching solution used to remove the molding blocks 3 may be an aqua regia solution, in other words a mixture of hydrochloric acid and nitric acid. To be able to withstand such etching, it may in this case be envisaged to produce titanium blocks 8, 9.

(42) On removing the molding blocks 3, the blocks 9, 10 situated at the ends thereof, make it possible to ensure mechanical strength of non-removed portions 7a of the thin layer 7.

(43) The embodiment example given in FIG. 2F shows the structure following the etching, with the blocks 9, 10 envisaged to form source and drain regions suitable for holding portions 7a of the thin layer 7 which extend vertically, in other words in a direction forming an angle different to zero, for example of the order of 90°, with the principal plane of the substrate 1.

(44) The rest of the masking 13 may then be removed. For example, when the masking 13 is made of resin, the removal is typically performed by dipping in a solvent followed by drying.

(45) In FIG. 1F, the structure is represented following the removal of the masking 13. The remaining vertical portions 7a of the thin layer 7 form fins envisaged to produce a channel structure for the transistor. The critical dimension Dc2 of the fins 7a corresponds to the thickness of the thin layer 7.

(46) Due to the holding of the fins 7a by the blocks 8, 9, fins 7a of extremely small critical dimension may advantageously be envisaged. Thus, when a thin layer 7 is deposited in the form of an atomic (or molecular) monolayer, a critical dimension Dc2 may thus be obtained of the order of the size of an atom (respectively of a molecule). Fins having a high aspect ratio Dc2/H in other words of the critical dimension Dc2 over the height H (dimension measured parallel with the z axis) while retaining good mechanism strength may also be envisaged. The distribution interval of the fins 7a is dependent on that initially envisaged for the molding blocks 3 and on the critical dimension Dc1 thereof.

(47) A surrounding gate is then produced on the fins 7a. For this, a layer of dielectric material 16 such as for example silicon oxide or HfO.sub.2 is first deposited. The fins 7a are then covered with a gate material 18 such as for example a metal such as gold (FIG. 1G).

(48) For certain applications, in particular for the use of a sensor, it may be required to remove certain portions of the gate stack at the level of given regions of the fins and thus reveal regions of the channel structure.

(49) Steps (not shown) of producing contacts and connections may then be envisaged.

(50) According to an alternative embodiment of the example of the method described above, after revealing the top portion of the molding blocks 3, the removal of the resin masking 13 may be carried out (FIG. 3A). The masking 13 is thus removed before removing the molding blocks 3 (FIG. 3B).

(51) In one or the other of the embodiment examples described above, the molding blocks 3 form a mould wherein the thin layer 7 substantially reproduces the relief. It is thus possible to produce fins of different shape to that described above particularly by adapting the shape of the molding blocks 3 which is not necessarily parallelepipedal.

(52) In one or the other of the embodiment examples described above, the molding blocks 3 are entirely removed. A partial removal of the molding blocks may alternatively be envisaged. Preferably, when a portion of the molding blocks is retained, these blocks are envisaged in a dielectric material so as particularly to prioritise conduction via the fin structure.

(53) In one or the other of the examples described above, the holding blocks 9, 10 have advantageously the further function of forming source and drain electrodes. It is also possible to envisage producing these electrodes after forming holding blocks 9, 10.

(54) A transistor as described above may find applications in various types of circuits for example in integrated circuits, processors, logic circuits.

(55) An example of finFET transistor produced using a method of the type of that described above is illustrated in FIG. 4. The gate stack 16, 18 is only retained on a central portion of a fin 7a, whereas end regions 7a1 situated between the central portion and the source and drain blocks 9, 10 are revealed.

(56) Such a structure may serve as a base for example for the embodiment of a detection structure for a chemical and/or biological sensor of enhanced detection sensitivity and of reduced size.

(57) The revealed regions 7a1 of the fin 7a may thus serve to accommodate at least one capture layer 22, also referred to as “functionalisation layer” and which is based on a material suitable for absorbing and/or adsorbing and/or reacting with at least one chemical or biological species to be detected.

(58) Chemical or biological elements to be detected are suitable for being absorbed or adsorbed or reacting with the capture layer 22, inducing a modification of the electrical properties of the channel structure.

(59) The detection sensitivity is especially enhanced as the fin 7a is thin.

(60) A fin based on a 2D crystal such as MoS2 or graphene has a very high surface-to-volume ratio while having very good electrical performances which renders the transistor provided with such a structure particularly advantageous for the use of high-sensitivity sensors.

(61) Various types of capture layers and the associated applications thereof are listed in the table hereinafter by way of example.

(62) TABLE-US-00002 PHYSICAL APPLICATIONS CAPTURE LAYER 22 CAPTURE PRINCIPLE Gas detection Porous materials Adsorption, (Porous alumina, Porous Physisorption, silicon, nanoporous Chemisorption amorphous carbon) Metal oxides (Cr.sub.2O.sub.3, Chemical reaction Mn.sub.2O.sub.3, Co.sub.3O.sub.4, NiO, (reduction oxidation CuO, SrO, In.sub.2O.sub.3, WO.sub.3, of capture layer atoms) TiO.sub.2, V.sub.2O.sub.3, Fe.sub.2O.sub.3) Non-biological Capture layers similar to those molecule of gas detectors may be applied detection in Molecular fingerprint Physical conformation liquid medium polymers (Composed of recognition a functional monomer polymerised around the molecule to be detected, the monomers suitable for being used for example: Methyl methacrylate,N,N′- methylenediacrylamide,3,5- bis(acryloylamido)benzoic acid Biological Antibodies (IgG, IgM, IgE) Specific biological molecule recognition detection DNA (complementary Strand strands of a DNA target hybridisation to be detected) Enzyme (Lactase, Specific biological Amylase, etc.) recognition

(63) According to a particular embodiment example wherein the sensor is a gas sensor, a capture layer 22 made of porous material such as porous alumina or porous silicon may be envisaged. According to a further particular embodiment example wherein the sensor is dedicated to non-biological molecule detection, a capture layer 22 made of polymer such as Methyl methacrylate may be envisaged.

(64) Given the dimensions of the fin(s) 7a forming the channel structure and the embodiment whereof is described above, such a transistor is particularly suitable for detecting very small quantities of biological and/or chemical elements, in particular compounds present at a concentration corresponding to several ppb or to several hundred ppb (parts per billion).

(65) A method as described above wherein the critical dimension of the fin is dependent on the thickness of a thin layer of two-dimensional material and the arrangement whereof is dependent on the shape and the arrangement of at least one molding block whereon this fin is formed, may be suitable to produce a structure of the type of that illustrated in FIG. 6.

(66) This structure includes several fins 7a, 7′a oriented in different directions not parallel with one another. In the particular example of FIG. 6, the structure includes fins 7a, 7′a disposed perpendicularly to the substrate (plans parallel with the plane [O;x;y], the substrate not being represented) and orthogonal with one another. The fins 7a, 7′a join at a meeting or intersection point. The fins 7a, 7′a are connected to and secured mechanically respectively by electrodes 91, 92.