Nonvolatile field-effect transistors based on two-dimensional electron gas.

Abstract

The present invention relates to a gate electrode including a first contact (C1); a source (14) having a second contact (C2); a drain (16) including a third contact (C3); a channel (20) made of a two-dimensional electron gas between the drain (16) and the source (14); a residual state subassembly (18) having two electrically controllable residual states and including at least one oxide layer; a reduced layer (22) made of at least one metal type reduced material and having an atomic concentration of metal elements of more than 50%; Equipped with The present invention relates to a nonvolatile field effect transistor (10) in which application of a voltage between a first contact (C1) and further contacts (C2, C3) results in nonvolatile modulation of the conductivity of a two-dimensional electron gas.

Description

[Technical field]

[0001] The present invention relates to a non-volatile field effect transistor that uses a two-dimensional electron gas. [Background technology]

[0002] Field effect transistors are often referred to by the abbreviation FET. Such transistors are often used as switches or amplifiers.

[0003] A field effect transistor is a unipolar three-terminal device with three terminals (drain, source, and gate) that relies on the action of an electric field on the conductivity of a channel located between the drain and source. In the channel, carriers such as electrons or holes can move freely between the source and drain under the action of a voltage applied between the two terminals. The conductance of the channel is controlled by applying a voltage to the gate. There are many types of field effect transistors, depending on the nature of the carriers (electrons or holes) or the nature of the electrical control applied on the channel.

[0004] Field effect transistors are currently constructed using a variety of semiconductor materials, particularly single crystal silicon materials, although other materials such as amorphous silicon, polycrystalline silicon, gallium arsenide, or gallium nitride are also used.

[0005] In a FET transistor, a power supply maintains the transistor in an on state (i.e., non-zero channel conductivity) or off state (zero channel conductivity) since the state depends on the voltage applied to the gate. If the power supply is interrupted, the voltage applied to the gate necessarily becomes zero and no state is preserved. Such a state is volatile. The constant application of a voltage to preserve the memory state results in power consumption to maintain the transistor state. Such consumption is called static because it is related to the generation of leakage currents and not to the dynamic process of writing or reading the memory.

[0006] Therefore, by creating non-volatile field effect transistors suitable for maintaining the transistor state without power supply, static energy consumption is eliminated and thus energy consumption is reduced. Non-volatility also makes it possible to consider transistors with additional functions, such as memory functions or logic calculation functions.

[0007] Based on semiconductor transistors, several non-volatile field effect transistors have been proposed. More specifically, research has focused on the development of ferroelectric field effect transistors, often referred to as Fe-FETs. Fe-FETs comprise a ferroelectric element between the gate and drain-source-channel that serves to maintain an on or off state in the absence of power supply.

[0008] Known Fe-FETs suffer from limited duration and contrast between the on and off states (sometimes called on / off contrast). Summary of the Invention [Problem to be solved by the invention]

[0009] Therefore, there is a need for non-volatile transistors that have better duration and improved contrast between the on and off states. [Means for solving the problem]

[0010] To this end, a non-volatile field effect transistor is described herein, comprising a first electrode, called gate electrode, including a first contact, a second electrode, called source electrode, where the source includes a second contact, a third electrode, called drain, where the drain includes a third contact, a channel between the drain and the source, the channel being made of a two-dimensional electron gas, a residual state subassembly, having at least two electrically controllable residual states, in contact with the channel, the residual state subassembly including at least one oxide layer, and a reduced layer made of at least one metal type reduced material, where each metal type reduced material has an atomic concentration of metal element of 50% or more. The first electrode is in contact with the residual state subassembly, and application of a voltage between the first contact and one of the second and third contacts results in a non-volatile modulation of the electrical conductivity of the two-dimensional gas forming the channel.

[0011] The reduced layer makes it possible to create a two-dimensional gas of electrons that forms channels at the interface between the reduced layer and the residual state subassembly.

[0012] According to particular embodiments, the non-volatile transistor has one or more of the following characteristics, taken individually or according to all technically possible combinations: The reduced layer acts to create two-dimensional gas forming channels at the interface between the reduced layer and the residual state subassembly. the reduced layer is a single layer of metal-type reduced material or multiple layers of metal-type reduced materials, each of which has an atomic concentration of a metal element of 50% or more. the residual state subassembly comprises at least one non-volatile dielectric layer or multilayer electrically controlled by a ferroelectric effect, a trapped charge effect, an ion migration effect, or a combination of a number of said effects; Each metal-type reduced material has an atomic concentration of the metal element of 80% or more. each metallic element is selected from the list consisting of aluminum, tantalum, yttrium, magnesium, and ruthenium. the reduced layer has a thickness of less than or equal to 15 nanometers, preferably less than or equal to 10 nanometers; the transistor comprises a substrate carrying the drain, the source and the reduced layer, the reduced layer, the channel and the residual state subassembly forming a stack in said order, and a first electrode carrying the residual state subassembly. A residual state subassembly is placed on the first electrode, a drain, a channel and a source are placed on the residual state subassembly, and a reduced layer is disposed on the channel. the non-volatile transistor further comprises a protective layer covering the reduced layer. - the residual state subassembly comprises an oxide layer made of a perovskite oxide, (Hf 1-x Zr x )O2, where x varies between 0 and 1, doped where appropriate, and a layer made from poly(vinylidene fluoride). the transistor comprises a read unit, the read unit comprising a subunit for applying a voltage between the drain and the source, and a subunit for measuring a current between the drain and the source. - The source and drain contact the ends of the channel.

[0013] The characteristics and advantages of the invention will become apparent from reading the following description, given by way of example only and without limitation, and from reference to the enclosed drawings, in which: [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a schematic diagram of an example of a non-volatile field effect transistor using a two-dimensional electron gas. [Diagram 2] FIG. 2 is a schematic diagram of another example of a non-volatile field effect transistor using a two-dimensional electron gas. [Diagram 3]1 is a graph showing the dependence of the resistance between two contacts as a function of the applied voltage. [Figure 4] 1 is a graph illustrating current-voltage characteristics of an example transistor in an on and off state. [Diagram 5] 1 is a schematic diagram of an example of a transistor having a read unit and a write unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] A transistor 10 is shown in FIGS.

[0016] Transistor 10 is a non-volatile field effect transistor.

[0017] The transistor 10 includes three electrodes, a first electrode referred to as the gate electrode 12 including a first contact C1, a second electrode referred to as the source 14 including a second contact C2, and a third electrode referred to as the drain 16 including a third contact C3.

[0018] The electrodes are advantageously metallic or made from highly doped semiconducting materials.

[0019] Transistor 10 further includes a residual state subassembly 18 , a channel 20 formed from a two-dimensional electron gas and disposed between drain 16 and source 14 , and a reduced layer 22 .

[0020] As can be seen in FIG. 5, the transistor 10 further includes a read unit 40 and a write unit 42 .

[0021] The reading unit 40 further includes a voltage application subunit 44 and a current measurement subunit 46 .

[0022] Two separate layouts of the layers of transistor 10 are proposed in FIGS. 1 and 2, respectively.

[0023] In the following, the layers of a stack are the layers that are stacked along the stack direction. The relative concepts of low and high with respect to the direction of the stack are also defined. If a layer is lower in the sheet representation of FIG. 1, it is lower than another layer.

[0024] In the example shown in FIG. 1, non-volatile transistor 10 includes a substrate 24 carrying thereon a stack 26 of layers comprising drain 16, source 14, and reduced layer 22, channel 20, and residual state subassembly 18, stack 26 being stacked in the aforementioned order when going from bottom to top (the bottom corresponding to the substrate in this case).

[0025] Finally, the thickness of a layer is defined as the dimension along the direction of the stack of layers, ie the distance between its two faces. The gate electrode 12 is in contact with the residual state subassembly 18 .

[0026] The order of the layers in stack 26 can be reversed as shown in FIG.

[0027] 2, the residual state subassembly 18 rests on the gate electrode 12. The drain 16, the channel 20, and the source 14 rest on the residual state subassembly 18.

[0028] The channels 20 and reduced layers 22 form a stack 28 in that order from the bottom through the stack 28 to the top.

[0029] Additionally, stack 28 optionally includes a protective layer 30 overlying reduced layer 22 .

[0030] The protective layer 30 is made from an insulating material that is characterized by zero or almost zero electrical conductivity and makes it possible to protect the stack 28, in particular against oxidation.

[0031] The protective layer 30 is made, for example, from SiO2.

[0032] In the two cases shown, the source 14 and the drain 16 contact respective ends of the channel 20 .

[0033] The residual state subassembly 18 has at least two electrically controllable residual states, where the electrical control is obtained by applying a voltage between the first contact C1 and the second contact C2 or the third contact C3.

[0034] In other words, the residual state subassembly 18 is characterized by a nonlinear relationship between applied voltage and apparent stored charge that follows a hysteresis cycle, and by creating at least two residual states.

[0035] Thus, the residual state subassembly 18 is a dielectric layer or a stack of non-volatile dielectric layers that is electrically controlled by ferroelectric effects, trapped charge effects, ion migration effects, or a combination of two or more of the aforementioned effects. The residual states correspond to changes in ferroelectric polarization, number of trapped charges, or positions of ions, or a combination of such changes.

[0036] More precisely, the subassembly is made of perovskite oxide, or (Hf 1-x Zr x )O2 or (Hf 1-x Ga x )O2 (x varies between 0 and 1), possibly doped with other elements, or alloys thereof, or polyvinylidene fluoride.

[0037] Perovskite oxides have a structure of type ABO3, where A and B are cations. BaTiO3, PZT (i.e. PbZr 1-x Ti x O3, where x varies between 0 and 1), PMN-PT (i.e., [1-x]Pb(Mg 1 / 3 Nb 2 / 3)O3-xPbTiO3, where x varies between 0 and 1), BiFeO3 (where appropriate, for example doped with rare earth elements on the Bi site or Mn on the Fe site), SrTiO3 (doped where appropriate), KTiO3 (doped where appropriate), Pr 0.7 Ca 0.3 MnO3 (doped where appropriate), or YMnO3 (doped where appropriate) are examples of such materials.

[0038] The use of a dielectric material to create the residual state subassembly 18 serves to electrically control the conductivity of the channel 20 in a non-volatile manner.

[0039] In each of the above examples, the presence of the residual state results from a ferroelectric effect, a trapped charge effect, an ion migration effect, or a combination of several such effects. In practice, the predominant effect depends on the material and on the conditions of deposition of the layers forming the residual state subassembly 18. According to the described examples, the coercive field of the dielectric element and its thickness are small enough to be suitable for the writing device 14 to write the residual state with voltages compatible with microelectronic technology, i.e. voltages less than ten volts (<10V). It is possible to obtain such properties with a thickness of less than 100 nm, advantageously less than 50 nm, in the abovementioned materials. The residual state subassembly 18 is also resistant to cycling, typically at least 10 4 Suitable for enduring cycles.

[0040] The channel 20 consists of a two-dimensional electron gas, ie, a confined electron gas formed at the interface between the two layers, while no gas is present in the volume of the separate material.

[0041] Thus, the channel 20 is an electron gas that forms at the interface between the residual state subassembly 18 and the reduced layer 22. It has a high carrier concentration (typically 10 10 cm -2The confinement is such that the gas can be considered to be strictly two-dimensional, since only the vicinity of the interface is conductive.

[0042] Since the electron gas of the channel 20 is created at the interface between the residual state subassembly 18 and the reduced layer 22, its properties, in particular its resistance, depend directly on the state of the residual state subassembly 18. The two-dimensional electron gas, and therefore the resistance of the channel 20, can thus be electrically modulated in a non-volatile manner according to the state of the residual state subassembly 18. Thereby, by selecting the residual state of the residual state subassembly 18, it is possible to modulate the resistance of the channel 20.

[0043] The reduced layer 22 is used to create a two-dimensional electron gas that forms a channel at the interface between the reduced layer 22 and the residual state subassembly 18 .

[0044] The reduced layer 22 is made of a reduced material having an atomic concentration of metallic elements of 50% or more, preferably 80% or more.

[0045] By definition, atomic concentration is the ratio between the number of atoms of a metallic element and the total number of atoms.

[0046] In this sense, the reduced layer 22 is a reduced layer that can be described as a metal-type reduced layer.

[0047] As a specific example, each metal element of the reduction layer 22 is selected from Al, Ta, Ru, Pt, W, Ir, Mo, Ti, Y, Au, or an alloy thereof, such as PtW.

[0048] Other metallic elements such as Mg, Cr, Mn, Cu, Pd, Ag, Hf, Bi, Co, and Fe are contemplated.

[0049] Thereby, each metallic element is selected from the list consisting of Mg, Al, Ti, V, Ni, Cr, Mn, Cu, Mo, Nb, Ru, Rh, Pd, Ag, Hf, Ta, W, Ir, Bi, Co, Y, Pt, W, Au, and Fe, or alloys thereof.

[0050] Preferably, each metallic element is selected from the list consisting of aluminum, tantalum, yttrium, magnesium, and ruthenium.

[0051] In a variant, the reduced layer 22 is multi-layered. In such a case, each layer of said multi-layered has, again, an atomic concentration of the metallic element greater than or equal to 50%, preferably greater than or equal to 80%.

[0052] The reduced layer 22 allows the formation of a two-dimensional gas of electrons in contact with the residual state subassembly 18. In this sense, the channel 20 formed by the two-dimensional electron gas is quite different from the channel of a conventional field-effect transistor, which does not have a reduced layer and in which carriers reversibly appear in a semiconductor material (typically made of doped Si) by applying a gate voltage that generates an electrostatic field.

[0053] The reduced layer 22 has a thickness that is less than or equal to 15 nanometers (nm).

[0054] This allows the reduced layer 22 to be partially or completely oxidized in contact with the residual state subassembly 18, so that it has a conductivity lower than that of the two-dimensional electron gas. The thickness of the reduced layer 22 is preferably selected so that it is completely oxidized in contact with the residual state subassembly 18, so that its conductivity is the lowest possible.

[0055] As noted above, the source 14 and drain 16 contact respective ends of the channel 20 and are thereby electrically connected to each other through the channel 20 .

[0056] Each of the electrodes 12, 14, and 16 is made of a conductive or semiconducting material, preferably doped metallic, so that it has high electrical conductivity.

[0057] The operation of transistor 10 will now be described with reference to FIGS.

[0058] In a first example of operation, the residual state subassembly 18 has two states, designated by A and B.

[0059] The above can be seen specifically in FIG. 3, where the voltage applied between the first contact C1 and the second contact C2 (gate voltage V g ) is shown the dependence of the resistance between the second contact C2 and the third contact C3 as a function of the gate voltage. It is indeed observed that the variation of the conductivity of the channel 20 as a function of the gate voltage has a hysteresis corresponding to the two residual states A and B of the subassembly.

[0060] According to a first example of such operation, writing is performed by applying a voltage between the first contact C1 and the second contact C2 by the writing unit.

[0061] Writing can also be performed by replacing the second contact C2 with a third contact C3.

[0062] From a physical point of view, the conductivity of the channel 20 is here modulated in a non-volatile manner according to the residual state of the residual state subassembly 18 by application of a voltage between the first contact C1 and the second contact C2.

[0063] In an example of such operation, application of a negative voltage between the first contact C1 and the second contact C2 causes the residual state subassembly 18 to start in residual state A, resulting in a low conductivity in the channel 20. On the other hand, application of a positive voltage causes a high conductivity in the channel 20, corresponding to residual state B of the residual state subassembly 18.

[0064] Depending on the materials used, the sign of the voltages that result in the high and low resistance residual states can be reversed, which may result in the application of a positive voltage between the first contact C1 and the second contact C2, which results in a residual state of low conductivity of the channel 20. In such a case, it is the application of a negative voltage that causes a residual state corresponding to a high conductivity of the channel 20.

[0065] In a first example of such operation, the reading unit 40 non-destructively reads the residual state by applying a voltage between the second contact C2 and the third contact C3 using the voltage application sub-unit 44 and measuring the resulting current between the two contacts C2 and C3 by the measurement sub-unit 46.

[0066] Thereby, the current-voltage characteristic between the second contact C2 and the third contact C3 corresponds to the output characteristic of the transistor 10, but is controlled by the residual state of the residual state subassembly 18.

[0067] Such characteristics are shown in FIG.

[0068] In a second example of operation, the voltage application subunit 44 applies a voltage between the first contact C1 and the second contact C2 to obtain any state of the hysteresis cycle. The conductivity of the channel 20 can have more than two states, which can be used to encode analog, non-binary information, such as synaptic weights in neural networks for artificial intelligence.

[0069] The transistor 10 described thus far is a non-volatile field effect transistor that uses a two-dimensional electron gas channel 20 at the interface between an oxide layer and a metallic type reduced layer 22. It associates the oxide layer of the residual state subassembly 18 with the reduced layer 22, which makes it possible to achieve non-volatile control of the conductivity of the channel 20.

[0070] This makes it possible to avoid the use of additional ferroelectric elements on the structure, since the combination of the residual state subassembly 18 and the reduced layer 22 creates both the two-dimensional electron gas and the residual state.

[0071] Furthermore, the presence of the assembly formed by the oxide layer, the two-dimensional electron gas, and the reduced layer 22 makes it possible to optimize the conductivity of the channel 20 and obtain better duration and improved contrast between the on and off states compared to prior art nonvolatile field effect transistors. The performance obtained by the two-dimensional electron gas channel 20 produced by using a reduced layer formed according to conventional microelectronic methods such as cathodic sputtering serves to make the transistor 10 suitable for industrial use. [Explanation of symbols]

[0072] 10 Non-volatile field effect transistor 12 First electrode, gate electrode 14 Second electrode, source, write device 16 Third electrode, drain 18 Residual State Subassembly 20 Channels 22 Reduction layer 24 Substrate 26 Stack 28 Stack 30 protective layer 40 reading units 42 Write Unit 44 Voltage application subunit 46 Current measurement subunit

Claims

1. A non-volatile field effect transistor (10), comprising: a first electrode (12), called the gate electrode, including a first contact (C1); a second electrode (14), called source (14), said source (14) comprising a second contact (C2); a third electrode (16), called the drain (16), said drain (16) including a third contact (C3); a channel (20) between the drain (16) and the source (14), the channel (20) being made of a two-dimensional electron gas; a residual state subassembly (18) having at least two electrically controllable residual states, in contact with said channel (20), and including at least one oxide layer; a reduced layer (22) which serves to create the two-dimensional gas forming channel (20) at an interface between the reduced layer (22) and the residual state subassembly (18), the reduced layer (22) being made of at least one metal type reduced material, each metal type reduced material having an atomic concentration of a metal element of 50% or more; Equipped with A nonvolatile field effect transistor (10), wherein the first electrode (12) is in contact with the residual state subassembly (18), and application of a voltage between the first contact (C1) and one of the second contact (C2) and the third contact (C3) results in a nonvolatile modulation of the conductivity of the two-dimensional gas forming the channel (20).

2. 2. The non-volatile transistor of claim 1, wherein the reduced layer (22) is a single layer of a metal-type reduced material or a multi-layer of metal-type reduced materials, one or more of the reduced materials each having an atomic concentration of a metal element of 50% or more.

3. 3. The non-volatile transistor of claim 1 or 2, wherein the residual state subassembly (18) comprises at least one non-volatile dielectric layer or multilayer electrically controlled by a ferroelectric effect, a trapped charge effect, an ion migration effect, or a combination of a plurality of said effects.

4. 4. The non-volatile transistor of claim 1, wherein each metal-type reduced material has an atomic concentration of the metal element of 80% or more.

5. 5. The non-volatile transistor of claim 1, wherein each metal element is selected from the list consisting of Mg, Al, Ti, V, Ni, Cr, Mn, Cu, Mo, Nb, Ru, Rh, Pd, Ag, Hf, Ta, W, Ir, Bi, Co, Y, Pt, W, Au, and Fe, or alloys thereof.

6. 5. The non-volatile transistor of claim 1, wherein each metallic element is selected from the list consisting of aluminum, tantalum, yttrium, magnesium, and ruthenium.

7. A non-volatile transistor according to any one of claims 1 to 6, wherein the reduced layer (22) has a thickness of less than or equal to 15 nanometers, preferably less than or equal to 10 nanometers.

8. the transistor (10) comprises a substrate (24) carrying the drain (16), the source (14), and the reduced layer (22); the reduced layer (22), the channel (20), and the residual state subassembly (18), in said order, form a stack; The non-volatile transistor of any one of claims 1 to 7, wherein the first electrode (12) rests on the residual state subassembly (18).

9. the residual state subassembly (18) rests on the first electrode (12); the drain (16), the channel (20), and the source (14) rest on the residual state subassembly (18); The non-volatile transistor of any one of claims 1 to 8, wherein the reduced layer (22) is disposed over the channel (20).

10. The non-volatile transistor of claim 9, further comprising a protective layer (30) covering the reduced layer (22).

11. The residual condition subassembly (18) an oxide layer made from a perovskite oxide; (Hf １－ｘ Zr ｘ ) O ２ where x varies between 0 and 1 and is doped where appropriate; and Layer made from poly(vinylidene fluoride) 11. The non-volatile transistor of claim 1, comprising a layer selected from the group consisting of:

12. 12. The non-volatile transistor of claim 1, wherein the transistor (10) includes a read unit, the read unit including a subunit for applying a voltage between the drain (16) and the source (14) and a subunit for measuring a current between the drain (16) and the source (14).

Images