ELECTRONIC COMPONENT WITH AT LEAST ONE LAYER OF A FERROELECTRIC OR ANTIFERROELECTRIC MATERIAL

20230200085 · 2023-06-22

    Inventors

    Cpc classification

    International classification

    Abstract

    An electronic component with at least one layer of a ferroelectric or antiferroelectric material. The layer may be provided for setting an imprint with a chemical element as a dopant which has a different number of free outer electrons than a non-oxide element of the ferroelectric or antiferroelectric material, and is introduced into the layer in a locally inhomogeneous distribution.

    Claims

    1-9. (canceled)

    10. An electronic component with at least one layer of a ferroelectric or antiferroelectric material, wherein the layer is provided, for setting an imprint, with a chemical element as a dopant which has a different number of free outer electrons than a non-oxide element of the ferroelectric or antiferroelectric material, and is introduced into the layer in a locally inhomogeneous distribution.

    11. The electronic component according to claim 10, wherein the chemical element is introduced at a proportion with respect to the material with which the layer is formed of between 1 percent and 10 percent.

    12. The electronic component according to claim 10, wherein said layer is formed of a ferroelectric material having a fluorite structure.

    13. The electronic component according to claim 10, wherein the ferroelectric material is hafnium oxide or zirconium oxide.

    14. The electronic component according to claim 10, wherein the chemical element is introduced into the layer in an asymmetric distribution.

    15. The electronic component according to claim 10, wherein the chemical element is selected from aluminum, lanthanum and yttrium.

    16. The electronic component according to claim 10, wherein the layer is formed with an antiferroelectric material in which the chemical element used for doping is contained in the layer material in such a way that when the external electrical potential is neutral, the polarization of the layer does not switch back.

    17. A buffer capacitor as an electronic component according to claim 16.

    18. A method of manufacturing an electronic component having at least one layer of a ferroelectric or antiferroelectric material, wherein the layer is provided, for setting an imprint, with a chemical element as a dopant which has a different number of free outer electrons than a non-oxide element of the ferroelectric or antiferroelectric material, and is introduced into the layer in a locally inhomogeneous distribution.

    Description

    DESCRIPTION OF THE FIGURES

    [0016] Examples of the invention are shown in the drawings and are explained below with reference to FIGS. 1-4.

    [0017] Shown are:

    [0018] FIG. 1 a schematic sectional view of an electronic component with a ferroelectric layer;

    [0019] FIG. 2 a diagram of the polarization-voltage curve for the component shown in FIG. 1;

    [0020] FIG. 3 a view corresponding to FIG. 1 of a further component with ferroelectric layer, and

    [0021] FIG. 4 a view corresponding to FIG. 2 of the polarization-voltage curve for the electronic component shown in FIG. 3.

    DETAILED DESCRIPTION

    [0022] FIG. 1 shows a schematic sectional view of an electronic component in which a multilayer system is arranged between an upper electrode 1 and a lower electrode 5. One of the layers, the top layer in the embodiment example shown in FIG. 1, is made of a ferroelectric material, thus forming ferroelectric layers 2, 3, and 4. The two further layers 3 and 4 are only shown as several layers by way of example. Typically, titanium nitride, TiN, or tantalum nitride, TaN, are used as electrode materials for electrodes 1 and 5. Alternatively, semiconductor materials such as silicon, silicon germanium, SiGe, indium gallium zinc oxide, IGZO, or two-dimensional materials such as graphene can be used.

    [0023] In the embodiment shown in FIG. 1, the ferroelectric layer is made of hafnium oxide, but in further embodiments it can also be formed of zirconium oxide or generally have a ferroelectric material with a fluoride structure. Alternatively, the ferroelectric layers 2, 3 and 4 may be replaced by an antiferroelectric or antiferroelectric-like layer.

    [0024] In the embodiment shown, the ferroelectric layer 2 has a dopant added to it, for example the chemical element aluminum, to thus selectively create oxygen vacancies that generate an internal field and thus an imprint. This is achieved by using dopants whose electric charge differs from that of hafnium or zirconium (both 4+). The dopant is introduced into the ferroelectric layers in a locally inhomogeneous distribution, which can be understood to mean in particular an asymmetric distribution. Here, an asymmetric distribution is understood to mean that there is no axis of symmetry in the ferroelectric layer 2 parallel to the direction of current flow, i.e. from electrode 1 to electrode 2 (vertical in this embodiment example). The dopant concentration is typically 2 percent to 4 percent, i.e., the chemical element aluminum is introduced into the ferroelectric layer 2 at a level between 2 percent to 4 percent (mass percent or volume percent) relative to hafnium oxide in the example shown. The materials used can be easily integrated into existing CMOS processes, so that nothing stands in the way of manufacturing such components on an industrial scale.

    [0025] FIG. 2 shows a diagram in which the polarization of the ferroelectric layer 2 is plotted against the applied electrical voltage (labeled “voltage”). The polarization-voltage curve achieved by the locally inhomogeneous distribution of the dopant can be seen. Since a steeper switching behavior can be achieved with such a component or a process in which a locally inhomogeneous doping of the ferroelectric layer 2 is performed, such a component is also favorable for memory applications since a smaller variability can be achieved.

    [0026] FIG. 3 shows a cross-sectional view of another electronic component corresponding to FIG. 1, but in this case the positions of layers 2 and 4 have now been exchanged, i.e. ferroelectric layer 2 is now in direct contact with lower electrode 5. Recurring features are marked with identical reference signs in all figures. FIG. 4 shows the resulting polarization-voltage curve in a representation corresponding to FIG. 2, which is now different according to the changed arrangement of the layers.

    [0027] In a process for fabricating the illustrated electronic components, the layers are typically deposited starting from the bottom electrode 5, for example by atomic layer deposition, chemical vapor deposition, or physical vapor deposition. The ferroelectric layer 2 or the ferroelectric layers 2, 3, and 4, if several of these layers are introduced (possibly also deposited one on top of the other to ultimately form a single layer in the component), are usually generated by dopants of the same charge or homogeneously distributed dopants of different electrical charge, which is also referred to as codoping, since two or more dopants are present in the entire layer. That is, the first dopant is used to adjust the ferroelectric or antiferroelectric properties (and may be present in a homogeneous distribution), the second dopant with a non-oxide element of the ferroelectric or antiferroelectric material with a different number of free outer electrons is introduced in a spatially inhomogeneous distribution for targeted local influencing of the properties of the respective layer.

    [0028] The electronic components shown in FIGS. 1 and 3, or components with a correspondingly similar structure, can be used in all areas in which (anti-)ferroelectric layers with a fluoride structure (in particular HfO.sub.2, ZrO.sub.2, Hf.sub.xZr.sub.1−xO.sub.2) are used, for example to compensate for unwanted imprints. Often, when integrating ferroelectric material stacks into existing semiconductor manufacturing technologies, different processes or process conditions are used for the deposition of electrodes, which causes an intrinsic asymmetry and thus imprint, which should be compensated. The local doping approach presented here is a cost-effective alternative to adapting a larger manufacturing flow with numerous dependencies.

    [0029] On the other hand, this can be used specifically for antiferroelectric devices, which should behave in a non-volatile manner. Local doping allows the antiferroelectric switching operating point to be shifted such that no switching back occurs at neutral external potential. This solution represents a CMOS-compatible variant to the approach mentioned at the beginning by means of asymmetric exit work of non-CMOS-compatible electrode materials. Another field of application are components, which should behave differently depending on the direction of the voltage change, both electrically and in mechanical expansion (inverse piezoelectric effect) as well as pyroelectrically.

    [0030] In the case of antiferroelectric buffer capacitors, local doping can be used both to shift the point of application of antiferroelectric (energy) storage and to improve reliability by moving away from the breakdown voltage. The described electronic component can thus be used as a buffer capacitor or be designed as a buffer capacitor. Local doping and imprint generation can be used selectively to optimize the reliability, especially data retention, of ferroelectric devices. One example here is a targeted compensation of depolarization fields in ferroelectric field-effect transistors.

    [0031] Directed doping can simplify the peripheral circuitry of arrays of nonvolatile memory elements because operations can be shifted toward a voltage polarity. For example, it is often undesirable to pass high negative voltages.

    [0032] The project that led to this application was funded by the ECSEL Joint Undertaking (JU) under Grant Agreement No. 826655. JU receives support from the European Union's Horizon 2020 research and innovation program and from Belgium, France, Germany, the Netherlands and Switzerland.