Semiconductor devices and manufacturing methods thereof
By forming specific metal stacked structures on two-dimensional materials and using CVD direct growth, the problems of delamination and stability of two-dimensional materials during manufacturing have been solved, improving the reliability and manufacturing efficiency of sensors and other devices.
Patent Information
- Application Number
- CN202580028674.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-03-22
- Filing Date
- 2025-03-21
- Publication Date
- 2026-11-17
AI Technical Summary
In existing technologies, two-dimensional materials are prone to delamination during manufacturing, especially when there is weak interaction with the underlying substrate. They also exhibit poor stability when used in liquid environments, affecting the reliability and manufacturing efficiency of devices such as sensors.
A specific metal stack structure is adopted, including titanium and metal layers such as chromium, aluminum, gold, and palladium. Stable electrical contacts are formed by etching, and two-dimensional material layers are grown directly on the substrate by CVD, avoiding impurities and defects in the transfer process and improving adhesion and mechanical stability.
It improves the integration stability of two-dimensional materials and metal contacts, enhances the reliability and manufacturing efficiency of devices such as sensors, reduces the risk of delamination, and is suitable for mass production.
Smart Images

Figure CN123125486A_ABST
Abstract
Description
[0001] This invention relates to semiconductor devices, and more particularly to semiconductor devices comprising a two-dimensional material layer structure and a gold layer thereon (providing electrical contacts to the two-dimensional material layer structure). The invention also relates to a method of manufacturing such a device. The device of this invention is particularly suitable for use as a sensor.
[0002] Two-dimensional (2D) materials, particularly graphene, and their electronic devices are currently a hot topic of research and development worldwide. 2D materials have proven to possess remarkable properties both theoretically and practically, leading to a plethora of products incorporating such materials (to name just a few, including coatings, batteries, and sensors). Graphene stands out and is being investigated for a range of potential applications. Most notably, graphene is being used in electronic devices and their components, including transistors, diodes, LEDs, photovoltaic cells, Hall effect sensors, current sensors, biosensors, gas sensors, and more.
[0003] Due to the extremely thin nature of two-dimensional materials and their typically minimal interactions with the underlying substrate / wafer (generally based primarily on weak van der Waals interactions), the inventors have found that such materials are particularly prone to delamination. This problem can arise during fabrication and when various layers are deposited on and over the two-dimensional material. It is also particularly pronounced in subsequent solution-based processing, and in other environments and situations where the two-dimensional material is re-exposed to liquids and solutions (i.e., both in device fabrication and in the use of the final device). One such example of a device having a surface of two-dimensional material exposed for receiving liquid samples is a biosensor.
[0004] WO 2023 / 148149 relates to graphene sensors, particularly graphene biosensors, in which the graphene layer structure has an exposed sample surface for receiving a test sample. The sensor includes a first electrical contact and a second electrical contact arranged on opposite sides of the sample surface in contact with the graphene layer structure, and each electrical contact is separated from the sample surface by a directly adjacent metal oxide layer for isolating the electrical contact from the sample surface. The electrical contacts are made of a metal (e.g., chromium, titanium, aluminum, nickel, and / or gold).
[0005] WO 2022 / 246261 relates to field-effect transistor arrays, and more particularly to integrated circuits having two-dimensional field-effect transistors for direct and indirect target signal measurements.
[0006] US 2017 / 0365474 relates to forming a trap on a graphene sheet by depositing a passivation layer on top of the graphene sheet.
[0007] US 2017 / 0365477 discloses a method for providing a temporary layer on a graphene sheet, comprising: transferring the graphene sheet to a selected wafer; depositing a metal layer onto a first surface of the graphene sheet; and removing the metal layer.
[0008] US 2017 / 0365562 discloses a method for patterning graphene, comprising: placing a graphene sheet on a wafer; depositing a metal layer on the graphene sheet; depositing a photoresist layer on the metal layer; and etching a pattern on the photoresist layer to expose the metal layer.
[0009] “Systematic Comparison of Metal Contacts on CVD Graphene” (45 th The European Solid State Device Research Conference (ESSDERC) 184-187 (IEEE, 2015) detailed experimental studies for forming high-quality ohmic contacts with graphene. Metal contacts of platinum / gold (Pt / Au), nickel / gold (Ni / Au), palladium (Pd), Ni, and Au with monolayer chemical vapor deposition graphene were investigated. Experimental data showed that pure Au and Ni / Au provide highly reproducible low-resistance ohmic contacts.
[0010] The study, “Contact Resistance Study of Various Metal Electrodes with CVD Graphene” (Solid-State Electronics 2016, 125, 234-239), investigated the contact resistance of various metals with chemical vapor deposition monolayer graphene. Transfer length method (TLM) structures with different widths and contact pitches were fabricated and electrically characterized under ambient air and vacuum conditions. The electrical contacts were made of five metals: gold, nickel, nickel / gold, palladium, and platinum / gold.
[0011] The article “The role of contact resistance in graphene field-effect devices” (arXiv:1705.04025v1 [cond-mat.mes-hall] and Progress in Surface Science 2017,92(3), 143-175) reviews experimental and theoretical work (including through different metals, surface treatments or device architectures) focusing on reducing contact resistance in graphene transistors.
[0012] "Recent Progress in 1D Contacts for 2D-Material-Based Devices" (Adv. Mater. 2023, 34, 2202408) provides a detailed overview of the electrical characteristics of 2D devices with 1D edge contacts fabricated using both top-down and bottom-up methods.
[0013] US 2014 / 151632 discloses a graphene film current spreading layer applied to a vertically structured GaN-based LED, wherein existing metal electrodes are replaced by graphene material with high light transmittance and electrical conductivity as the current spreading layer, thereby improving the luminous efficiency of the LED device while reducing its manufacturing cost.
[0014] The paper “High Frequency Performance of Graphene Transistors Grown by Chemical Vapor Deposition for Mixed Signal Applications” (Jpn. J. Appl. Phys. 2011, 50, 070114) discloses the use of Ti / Pd / Au-based ohmic contacts and a mixed gate dielectric stack of 5 nm SiO2 and 15 nm Al2O3 grown by atomic layer deposition, and demonstrates the high frequency performance of the associated graphene transistors.
[0015] CN 116435373 belongs to the field of semiconductor devices, and specifically relates to normally-off two-dimensional gallium oxide heterojunction field-effect transistors and their fabrication methods.
[0016] The present invention aims to overcome or at least reduce the aforementioned problems in the prior art to improve the reliability of device manufacturing (particularly sensors involving exposed two-dimensional materials) and to improve the manufacturing efficiency of integrating two-dimensional materials and desired metal contacts into robust electronic devices, or at least to provide a commercially viable alternative.
[0017] This disclosure will now be described further. In the following paragraphs, different aspects / implementations of this disclosure are defined in more detail. Unless expressly stated otherwise, the aspects / implementations so defined may be combined with any other one or more aspects / implementations. In particular, any feature indicated as preferred or advantageous may be combined with any other one or more features indicated as preferred or advantageous.
[0018] Therefore, a first aspect of the present invention provides a semiconductor device comprising:
[0019] It has a substrate with a patterned two-dimensional material layer structure;
[0020] A gold layer on the portion of a two-dimensional material layer structure near its edge; and
[0021] A contact stack adjacent to the edge of a two-dimensional material layer structure and overlapping with a gold layer, the contact stack comprising:
[0022] (i) A first metal layer on a portion of a substrate and on a gold layer, wherein the first metal is selected from titanium and chromium;
[0023] (ii) A second metal layer on the first metal layer, wherein the second metal is selected from aluminum, copper, gold, nickel, and palladium; and
[0024] (iii) A third metal layer on the second metal layer, wherein the third metal is selected from cobalt, chromium, iridium, iron, magnesium, niobium, platinum, ruthenium, silver, tantalum, titanium, titanium nitride and tungsten.
[0025] A second aspect of the present invention provides a method for forming a semiconductor device, the method comprising:
[0026] (a) Forming a precursor, said precursor comprising:
[0027] It has a substrate with a patterned two-dimensional material layer structure;
[0028] A precursor gold layer on a two-dimensional material layer structure; and
[0029] A contact stack adjacent to the edge of the two-dimensional material layer structure and overlapping with the precursor gold layer, the contact stack comprising:
[0030] (i) A first metal layer on a portion of a substrate and on a precursor gold layer, wherein the first metal is selected from titanium and chromium;
[0031] (ii) a second metal layer on a titanium and / or chromium layer, wherein the second metal is selected from aluminum, copper, gold, nickel, and palladium; and
[0032] (iii) A third metal layer on the second metal layer, wherein the third metal is selected from cobalt, chromium, iridium, iron, magnesium, niobium, platinum, ruthenium, silver, tantalum, titanium, titanium nitride, and tungsten; and
[0033] (b) Etching the precursor gold layer with a gold etchant to form the apparatus according to the first aspect.
[0034] The present invention provides a semiconductor device, preferably a sensor, more preferably an electrochemical sensor. The semiconductor device includes a two-dimensional material layer structure as an electronic component, and unique metal contacts formed by layers of gold and a first metal, a second metal, and a third metal described herein.
[0035] The device comprises a patterned two-dimensional material layer structure on a substrate. Two-dimensional materials are well known, with graphene being the most preferred for this invention. Two-dimensional materials include monoelemental two-dimensional materials such as graphene, silicene, phosphorene, and boronene (whether doped or undoped); and heteroelemental two-dimensional materials such as h-BN and transition metal dichalcogenides (TMDCs), such as MoS2, WS2, and MoSe2 (in the art, a monolayer is known to refer to a layer with an MX2 stoichiometry). Therefore, the device comprises a semiconducting / semi-metallic two-dimensional material, preferably graphene, phosphorene, or TMDC, because the unique electronic properties of such materials are desirable for electronic devices, particularly for at least the uppermost layer of a layered structure. For many devices, such as sensors, undoped (e.g., pristine or unintentionally doped) two-dimensional materials are preferred.
[0036] The two-dimensional material layer (which may be referred to herein as a two-dimensional material layer structure, such as a graphene layer structure) preferably has one to ten monolayers of two-dimensional material. Preferably, the two-dimensional material layer consists of a single monolayer. A preferred multilayer structure will have two or three monolayers. In some embodiments, all monolayers are of the same two-dimensional material (e.g., bilayer graphene), but in some preferred embodiments, the multilayer structure can be a heterogeneous structure. By way of example only, the two-dimensional material layer structure can be a sandwich structure of graphene sandwiched between two layers of hBN, or a TMDC on graphene.
[0037] The known thickness of a monolayer of graphene is about 0.34 nm, the known thickness of a monolayer of phosphorene is about 0.85 nm, and the known thickness of a monolayer of TMDC can range from about 0.6 nm to about 0.7 nm. As described in more detail herein, the combination of metals used to manufacture electrical contacts provides an advantageous means of resisting the delamination of such thin materials from the surface of the substrate.
[0038] Two-dimensional material layer structures are patterned. Patterning is a well-known technique in the fields of microfabrication and semiconductor manufacturing (i.e., semiconductor device manufacturing). Patterned two-dimensional material layer structures have shapes with one or more edges (typically multiple straight edges of a defined polygon, such as four to six outer edges). Patterned two-dimensional material layer structures are typically “solid” or “filled” shapes with only outer edges (i.e., the layer structure does not have holes, etc., similar to loops).
[0039] Quadrilaterals (i.e., four straight edges) are preferred, and the patterned two-dimensional material layer structure can preferably be substantially rectangular, as such shapes are simple to design and well-suited for many subsequent device applications (including sensors). Preferably, the angles of the patterned shape are about 90° (e.g., rectangular) or greater.
[0040] A two-dimensional material layer is disposed on a substrate, preferably on a non-metallic surface of the substrate, and in some embodiments on an insulating surface. The properties of the substrate are not particularly limited. Such substrates or wafers are known in the art to provide a substantially flat upper surface for fabricating devices by growth / deposition and patterning of additional layers. Preferably, the non-metallic surface on which the two-dimensional material layer is disposed is silicon (Si), silicon carbide (SiC), silicon nitride (Si3N4), silicon dioxide (SiO2), sapphire (Al2O3), aluminum gallium oxide (AGO), hafnium dioxide (HfO2), zirconium dioxide (ZrO2), yttrium-stabilized hafnium oxide (YSH), yttrium-stabilized zirconium oxide (YSZ), magnesium aluminate (MgAl2O4), yttrium orthoaluminate (YAlO3), titanate, etc. Strontium (SrTiO3), cerium oxide (Ce2O3), scandium oxide (Sc2O3), erbium oxide (Er2O3), magnesium difluoride (MgF2), calcium difluoride (CaF2), strontium difluoride (SrF2), barium difluoride (BaF2), scandium trifluoride (ScF3), germanium (Ge), cubic boron nitride (c-BN), hexagonal boron nitride (h-BN), and / or group III / V semiconductors, such as aluminum nitride (AlN) and gallium nitride (GaN).
[0041] In some embodiments, the substrate may be composed of a material such that the non-metallic surface is silicon, silicon nitride, silicon dioxide, sapphire, aluminum nitride, YSZ, scandium oxide, germanium, calcium difluoride, and / or magnesium difluoride. Preferably, the non-metallic surface is sapphire, yttrium-stabilized zirconium oxide, scandium oxide, calcium difluoride, or magnesium difluoride. In some preferred embodiments, the substrate comprises a first layer providing the non-metallic surface and a "substrate support" layer, or is composed of a first layer providing the non-metallic surface and a "substrate support" layer, with sapphire being a preferred example (such embodiments are particularly suitable for insulating substrates, i.e., typically the oxides, nitrides, and fluorides described). In other preferred embodiments, the substrate support layer comprises silicon. The silicon support layer comprises a structure of a "pure" silicon wafer (consisting essentially of silicon, doped or undoped) or may be referred to as a CMOS wafer (including additional associated circuitry). A first layer comprising rare earth oxides (including yttrium and scandium) or composed of rare earth oxides (including yttrium and scandium) is particularly preferably disposed on the silicon substrate support. Scandium oxide is particularly preferred, for example. Such a substrate is preferred for devices such as gas sensors. The substrate may also include one or more layers (e.g., regions or channels of embedded waveguide materials such as silicon nitride suitable for electro-optic modulators and photodetectors). The thickness of the substrate support layer is typically much greater than the thickness of the first layer thereon. Typically, the thickness of the substrate support layer is 250 µm to 1.5 mm, for example, 400 µm to 1 mm. On the other hand, the thickness of the first layer of such a substrate is significantly thinner and can be formed on the substrate support by epitaxy, such as molecular beam epitaxy (MBE), or high-temperature sputtering. Preferably, the thickness is at least 2 nm, preferably at least 5 nm and / or less than 500 nm, preferably less than 100 nm. A suitable range for the thickness of the first layer is preferably 5 nm to 100 nm, preferably 10 nm to 50 nm.
[0042] Particularly preferred in this invention is that the two-dimensional material layer structure is disposed on the surface of the substrate by directly forming a two-dimensional material layer on the surface via CVD (i.e., CVD-grown two-dimensional material layer structure directly grown on the surface of the substrate). Direct formation of the two-dimensional material on the substrate avoids steps such as physical transfer, which could otherwise introduce impurities and / or defects, risking negating the benefits of the invention. For example, direct formation avoids the use of transfer polymers that are difficult to remove. Formation can be considered synonymous with synthesis, manufacturing, production, and growth. CVD is discussed in more detail for the methods described herein.
[0043] Those skilled in the art can readily determine whether a two-dimensional material is grown directly on a substrate via CVD. This can be determined using conventional techniques in the art, such as atomic force microscopy (AFM) and energy dispersive X-ray (EDX) spectroscopy. The two-dimensional material may be free of copper contamination and organic polymer residues because these materials are not brought into contact with the two-dimensional material during the process (e.g., graphene is typically grown indirectly on a sacrificial catalytic copper substrate via CVD and then transferred to a non-metallic surface). Furthermore, such a transfer process is generally unsuitable for large-scale manufacturing (e.g., on silicon-based substrates in a processing plant) or at least uneconomical.
[0044] The device comprises a gold layer on a portion of a two-dimensional material layer structure adjacent to its edges. Preferably, particularly for sensors, additional surface portions of the two-dimensional layer structure are exposed. The exposed surfaces of the two-dimensional material layer structure may preferably be functionalized with an analyte acceptor. Suitable acceptors for functionalizing two-dimensional materials, including graphene, are well known in the art. The analyte is a chemical substance of interest to be measured, for example, to detect its presence or, more quantitatively, its concentration in the sample to be tested.
[0045] A sensor is a device that generates a signal in response to a biological or chemical interaction with an analyte. A typical sensor includes an analyte receptor (e.g., a biosensor includes a biological receptor), a transducer, electronics, and a display to measure the analyte. Many different conventional configurations of such components exist. In some systems, they may be integrated into a single device, or a biosensor may be part of a system that includes a separate reader with a display and performs diagnostics. Alternatively, in other systems, cloud-based services are used for data analysis / diagnostics, and a remote computer provides the diagnostics and display.
[0046] A biological receptor is a molecule or other biological element or species used to recognize an analyte. Biological receptors include, but are not limited to, enzymes, cells, aptamers, DNA, RNA, and antibodies. The interaction between the analyte and the biological receptor is also known as biorecognition, and biorecognition events (such as changes in light, heat, pH, charge, or mass) are subsequently converted into a measurable signal by a transducer (i.e., a two-dimensional material layer structure). Electronic devices are used to process and prepare the transducer signal for display to a user. For example, a processor or signal processing unit can process the voltage signal generated by the transducer through signal conditioning, such as by amplification and conversion from analog to digital form. The processed signal is then quantized by the display unit of the biosensor. Such processing can be performed remotely. The user of the biosensor device may be different from the user of the analytical output.
[0047] In other technical fields, the sensors described herein can be used for gas sensing, pH sensing, or ion sensing. Therefore, the devices have particular applications in healthcare and diagnostics, and environmental monitoring in agriculture (e.g., for testing the presence of food and heavy metals (and / or their ions)).
[0048] The sensor includes first and second electrical contacts typically arranged on opposite sides of a sample surface (e.g., adjacent to opposing parallel edges) and configured to contact a two-dimensional material layer structure. The electrical contacts are metallic contacts made of gold as described herein, as well as stacks of a first, second, and third metal. Such opposing arrangements of contacts on opposite sides of the sample surface are well known in the art and serve to define the sample surface of the transducer for receiving an analyte composition between the contacts. The sensor may also include gate contacts as known for field-effect transistor-based sensors. Particularly preferred are gate contacts formed on a substrate by contact stacks as described herein, such gate contacts can be used as common gate contacts for two or more sensors fabricated on a common underlying substrate. The electronic device may include two or more independent devices (where independent devices may subsequently be referred to as units), which may then share common device features (e.g., common gate contacts) elsewhere on the substrate. The following description will focus on independent devices or units, but it should be understood that arrays of independent components can be fabricated simultaneously on a common substrate / wafer.
[0049] The gold layer on the two-dimensional material layer structure may be referred to herein as the primary gold layer, for example, to distinguish it from any gold included in the second metal layer of the contact stack (which itself may be referred to as the secondary gold layer). The gold layer is positioned adjacent to the edges of the two-dimensional material layer structure. The gold layer may include two portions, each adjacent to two opposite edges, to provide electrical contacts.
[0050] As described herein, the two-dimensional material layer structure and the gold layer are simultaneously patterned during fabrication to form a co-patterned stack. This can be done using conventional techniques such as photolithography and etching. Therefore, the gold layer is not on a substrate but only on the two-dimensional material, and during device fabrication, etching after the formation of the contact stack, preferably after the patterning of the dielectric layer thereon, can further pattern the gold layer to expose a portion of the two-dimensional material (and form, for example, two electrical contacts). Preferably, the thickness of the gold layer is 20 nm to 200 nm, more preferably 80 nm to 120 nm.
[0051] The device also includes a contact stack adjacent to the edge of the two-dimensional material layer structure and overlapping the gold layer. That is, the contact stack is disposed on the substrate, adjacent to the edge of the two-dimensional material layer structure. Preferably, the contact stack on the substrate further extends to form connection traces for the circuit, such as connecting source and drain contacts. In a particularly preferred embodiment, the circuit also includes gate contacts on the substrate surface, thus the gate contacts are formed by the metal contact stack. The contact stack advantageously allows the gate contacts to be exposed during the gold etching step, thereby reducing the number of manufacturing steps in embodiments where a dielectric layer is patterned thereon.
[0052] The contact stack includes a first metal layer on a portion of the substrate and on a gold layer, wherein the first metal is selected from titanium (Ti) and chromium (Cr). Preferably, the thickness of the first metal layer is less than 15 nm and / or greater than 1 nm, for example, 3 nm to 10 nm. The first metal layer is an adhesion promoter that facilitates the adhesion of the second metal to the substrate surface, particularly non-metallic surfaces such as silicon and metal oxides (e.g., sapphire and rare earth oxides). Therefore, only a relatively thin layer is required. For the fabrication of contacts, the use of titanium and / or chromium in the adhesive layer is well known; these metals typically form covalent interactions with the substrate to ensure good adhesion and have good adhesion properties to the metal suitable for the second layer.
[0053] The contact stack also includes a second metal layer on top of the first metal layer, wherein the second metal is selected from aluminum (Al), copper (Cu), gold (Au), nickel (Ni), and palladium (Pd). Preferably, the thickness of the second metal layer is at least 50 nm and / or at most 100 nm, preferably 70 nm to 90 nm. The second metal layer is typically deposited directly after the first metal layer and thus employs the same pattern as the first metal layer (therefore located on and across the first metal layer), as is the case with the subsequent third metal layer. As is known in the art, the circuit pattern can be provided by using a patterned photoresist for photolithography, and the circuit pattern can be removed in a stripping step after the deposition of each metal layer, leaving the patterned contact stack.
[0054] The adhesion of the metal layers is required to provide at least the desired low contact resistance (which is often the focus of the prior art). However, the inventors also found that the mechanical properties of the layers are important for forming contacts with two-dimensional materials because their interaction with the substrate is relatively weak, making them prone to delamination. Thermal expansion between the layers and the substrate can induce stress in the microfabricated layers. Such stress can alter adhesion properties, and the inventors found that appropriately selecting the metal and its thickness can mitigate such problems. Furthermore, the inventors identified a metal that provides the desired chemical properties to protect the contact stack during gold etching.
[0055] When selected from the metals described herein, the second metal layer is stress-relieved, such that it is typically the thickest layer of the contact stack, providing the desired conductivity for the contacts and associated circuitry. It is generally preferred that the second metal comprises gold or is substantially composed of gold, but in some embodiments, combinations of metals may be used. Exemplary combinations include gold on nickel or palladium. Nickel and palladium can act as diffusion barriers to prevent the first metal from diffusing into the second metal layer and can be set to a thickness of 10 nm to 20 nm (e.g., about 15 nm of Ni or Pd and about 60 nm of Au).
[0056] The contact stack also includes a third metal layer on the second metal layer, wherein the third metal is selected from cobalt (Co), chromium (Cr), iridium (Ir), iron (Fe), magnesium (Mg), niobium (Nb), platinum (Pt), ruthenium (Ru), silver (Ag), tantalum (Ta), titanium (Ti), titanium nitride (TiN), and tungsten (W). The third metal layer preferably has a thickness sufficient to maintain conformality (i.e., a complete layer free from any pinholes, etc.). Preferably, the thickness of the third metal layer is less than 35 nm and / or greater than 10 nm, more preferably 15 nm to 30 nm.
[0057] Preferably, the contact stack has a thickness (i.e., the total thickness of the first, second, and third metal layers) extending vertically from the substrate by 80 nm to 200 nm. Therefore, in a preferred embodiment, the thickness of the first metal layer is less than 15 nm, the thickness of the second metal layer is 50 nm to 100 nm, and the thickness of the third metal layer is less than 35 nm. As described herein, it is also particularly preferred that the second metal is gold and the third metal is platinum.
[0058] Therefore, another preferred embodiment is a semiconductor device, which includes:
[0059] It has a substrate with a patterned two-dimensional material layer structure;
[0060] The main gold layer on the portion of the two-dimensional material layer structure adjacent to its edge; and
[0061] A contact stack adjacent to the edge of the two-dimensional material layer structure and overlapping with the main gold layer, the contact stack comprising:
[0062] (i) A titanium and / or chromium layer on a portion of the substrate and on the gold layer;
[0063] (ii) A sub-gold layer on a titanium and / or chromium layer; and
[0064] (iii) A platinum layer on a gold sublayer, the thickness of which is less than 35 nm.
[0065] The third metal layer is resistant to gold etching solutions (particularly iodine / potassium iodide solutions), thereby protecting the underlying second metal layer, which is easily etched by gold etchants. Like the second metal layer, the third metal layer may comprise two or more of such metals, or preferably consist essentially of one metal. Preferred metals are those particularly resistant to gold etchants and oxidation. Therefore, preferred metals include chromium, iridium, niobium, platinum, ruthenium, silver, tantalum, titanium, and titanium nitride, more preferably platinum. Although titanium nitride (especially in thin layers) is a ceramic material, it is a known conductive ceramic nitride, commonly referred to in the art as a barrier metal to chemically isolate semiconductors from soft metal interconnects while maintaining electrical connections between them. Therefore, those skilled in the art will readily recognize titanium nitride as a suitable metal in the context of contact stacks. Platinum is also used as a reference in electrochemical applications and thus provides a particularly suitable metal for the third metal layer as a protective layer for gate contacts.
[0066] Due to the stress that may be exerted by the metal used in the third metal layer, the inventors found that using a thinner layer than the second metal layer (e.g., less than 35 nm) is particularly desirable. After metal deposition, layers at great thicknesses typically possess internal stress, which, if excessive, can propagate within the layer. The inventors discovered that if the internal stress exceeds the adhesive force of the two-dimensional material layer structure on the substrate surface, this can cause delamination, including delamination of the entire device structure.
[0067] The device may also include a patterned dielectric layer on and across the contact stack, the dielectric layer being patterned to expose a portion of the two-dimensional material layer structure and gate contacts. That is, the dielectric layer is disposed at least on and across contacts (e.g., source and drain contacts) and metal traces in the two-dimensional material. For sensors, the dielectric layer may be patterned to provide a window to the exposed sample surface of the two-dimensional material layer structure. In some preferred embodiments, the window provided by the dielectric layer extends onto the surface of the two-dimensional material, thereby enclosing the edges of the two-dimensional material and isolating the sample surface from the electrical contacts.
[0068] Therefore, during manufacturing, the dielectric layer can be patterned simultaneously to provide windows to the gold layer on the two-dimensional material and to other metal stacks (such as gate contacts), allowing the gold layer to be subsequently etched to expose the surface of the two-dimensional material layer structure, while the gate contacts are already exposed, thus improving manufacturing efficiency. The dielectric layer on the precursor gold layer can also help encapsulate the edges of the second metal layer during the gold etching step, although some undercutting of the third metal layer in the absence of a dielectric layer still allows for top-down etching of the gold layer to expose the two-dimensional material layer.
[0069] The methods described herein for forming semiconductor devices include:
[0070] (a) Forming a precursor, said precursor comprising:
[0071] It has a substrate with a patterned two-dimensional material layer structure;
[0072] A precursor gold layer on a two-dimensional material layer structure; and
[0073] Contact stacks as described herein, adjacent to the edges of the two-dimensional material layer structure and overlapping with the precursor gold layer; and
[0074] (b) Etching the precursor gold layer with a gold etchant to form the device.
[0075] Therefore, the method includes partially etching the precursor gold layer to expose a portion of the underlying two-dimensional material layer structure. The remaining portion of the precursor gold layer provides the (main) gold layer of the product as electrical contacts on the surface of the two-dimensional material layer structure. The pattern of the gold etching can be controlled by the pattern of the overlapping contact stack that protects the underlying gold layer and optionally patterned dielectric layers.
[0076] As described herein, it is particularly preferred that the method comprises directly forming a two-dimensional material layer structure on the growth surface of a substrate by CVD.
[0077] CVD generally refers to a series of chemical vapor deposition techniques, each involving the deposition of thin film materials (e.g., two-dimensional crystalline materials such as graphene), optionally performed under vacuum / depressurized pressure. Volatile precursors (those in the gas phase or suspended in a gas) are decomposed to release the substances required to form the desired material (carbon in the case of graphene). The description of CVD for graphene in this article can be modified accordingly to produce other two-dimensional materials.
[0078] As described herein, CVD refers to thermal CVD, where the formation of graphene from the decomposition of carbon-containing precursors is a result of the thermal decomposition of those precursors. CVD layers formed directly on a surface can be distinguished from layers transferred due to impurities or other defects such as cracks and wrinkles.
[0079] Preferably, the method involves forming graphene via thermal CVD, such that decomposition is a result of heating the carbon-containing precursor. Preferably, the temperature of the growth surface during CVD is 700°C to 1350°C, more preferably 800°C to 1250°C, and even more preferably 1000°C to 1250°C. The inventors have found that such temperatures are particularly effective for directly providing graphene growth on the materials described herein via CVD. Preferably, the CVD reaction chamber used in the methods disclosed herein is a cold-walled reaction chamber, wherein a heater coupled to the substrate is the sole heat source for the reaction chamber.
[0080] In a particularly preferred embodiment, the CVD reaction chamber includes a tightly coupled nozzle having a plurality of precursor entry points or an array of precursor entry points. Such CVD equipment including tightly coupled nozzles is known for use in MOCVD processes. Therefore, the method can also be described as being carried out using an MOCVD reactor including tightly coupled nozzles. In either case, the nozzle is preferably configured to provide a minimum spacing of less than 100 mm, more preferably less than 25 mm, and even more preferably less than 10 mm between the substrate surface and the plurality of precursor entry points. It should be understood that a constant spacing means that the minimum spacing between the substrate surface and each precursor entry point is substantially the same. A minimum spacing refers to the minimum spacing between the precursor entry point and the substrate surface. Therefore, such an embodiment involves a “vertical” arrangement, in which the plane of the precursor entry point is substantially parallel to the plane of the substrate surface (i.e., the growth surface).
[0081] The precursor inlet leading to the reaction chamber is preferably cooled. The inlet, or during use, the nozzle is preferably actively cooled by an external coolant such as water to maintain a relatively low temperature at the precursor inlet, such that the temperature of the precursor as it passes through multiple precursor inlets and enters the reaction chamber is below 100°C, preferably below 50°C. For the avoidance of doubt, adding the precursor at temperatures above ambient does not constitute a heated chamber, as this would result in heat loss within the chamber and is partly responsible for establishing a temperature gradient within it.
[0082] Preferably, a combination of sufficiently small spacing between the substrate surface and the plurality of precursor entry points, cooling of the precursor entry points, and heating of the substrate to the extent of precursor decomposition creates a sufficiently steep thermal gradient extending from the substrate surface to the precursor entry points to allow graphene formation on the substrate surface. As disclosed in WO 2017 / 029470, a very steep thermal gradient can be used to facilitate the direct formation of a high-quality and uniform two-dimensional material layer on a non-metallic substrate, preferably across the entire surface of the substrate. The substrate diameter can be at least 5 cm (2 inches), at least 15 cm (6 inches), or at least 30 cm (12 inches). Particularly suitable equipment for the methods described herein includes the Aixtron® Close-Coupled Showerhead® reactor and the Veeco® TurboDisk reactor.
[0083] Therefore, in a particularly preferred embodiment of the method of the present invention, which involves using the method disclosed in WO 2017 / 029470, the direct formation of a graphene layer structure on a substrate by CVD comprises:
[0084] The growth substrate is placed on a heating base in a tightly coupled reaction chamber, the tightly coupled reaction chamber having a plurality of cooling inlets, the plurality of cooling inlets being arranged such that, in use, the inlets are distributed across the growth surface and have a constant interval from the substrate;
[0085] Cool the inlet to below 100°C (i.e., ensure the precursor is cold when it enters the reaction chamber).
[0086] Carbonaceous precursors in the gaseous phase and / or suspended in gas are introduced through an inlet into a tightly connected reaction chamber; and
[0087] The substrate is heated to a growth surface temperature at least 50°C higher than the decomposition temperature of the precursor to provide a sufficiently steep thermal gradient between the substrate surface and the inlet, thereby allowing the carbon released from the decomposed precursor to form graphene.
[0088] The constant interval is less than 100 mm, preferably less than 25 mm, and even more preferably less than 10 mm. The rotational speed of the heating base in the tightly connected reaction chamber is typically less than 300 rpm, or even less than 200 rpm.
[0089] In another particularly preferred embodiment of the method described herein, which involves the use of a method disclosed in WO 2019 / 138231, the direct formation of a graphene layer structure on a substrate by CVD comprises:
[0090] A growth substrate is placed on a heating base in a reaction chamber, the reaction chamber having a plurality of inlets, the plurality of inlets being arranged such that, in use, the inlets are distributed across the growth surface and have a constant interval from the substrate;
[0091] The heating base is rotated at a speed of at least 600 rpm, preferably up to 3000 rpm;
[0092] Carbonaceous precursors in the gaseous phase and / or suspended in gas are introduced into the reaction chamber through an inlet; and
[0093] The substrate is heated to a growth surface temperature at least 50°C higher than the decomposition temperature of the precursor.
[0094] The constant interval is at least 12 cm, and preferably up to 20 cm.
[0095] The most commonly used carbon-containing precursor for graphene growth in this field is methane (CH4). The inventors have found that, preferably, the carbon-containing precursor for forming graphene is an organic compound, i.e., a compound or molecule containing two or more carbon atoms and comprising carbon-hydrogen covalent bonds. The carbon-containing precursor is preferably a C3-C structure consisting of carbon and hydrogen, and optionally oxygen, nitrogen, fluorine, chlorine, and / or bromine. 10Organic compounds, more preferably C6-C9 organic compounds, are preferred. In a preferred embodiment, the precursor does not contain heteroatoms, such that the precursor consists of carbon and hydrogen. In other words, preferably, the carbon-containing precursor is a hydrocarbon, preferably an alkane. Also preferably, the organic compound contains at least two methyl groups (-CH3). Suitable precursors for other two-dimensional materials are known. TMDC can be grown by a combination of metal precursors and chalcogenide precursors (introduced simultaneously or sequentially into the reaction chamber).
[0096] The high-quality materials produced by such CVD methods also exhibit less and weaker interaction with the substrate due to fewer defects in the two-dimensional material layers (especially compared to transferred materials or graphene grown by sublimating silicon from silicon carbide). This is advantageous in terms of the unique electronic properties of two-dimensional materials, and therefore, directly grown materials may be more easily delaminated. Thus, there are synergistic benefits associated with such a combination.
[0097] Depositing a gold layer on and across a two-dimensional material layer structure can be achieved using conventional techniques such as sputtering, thermal evaporation, or electron beam evaporation. The gold layer is preferably deposited at a thickness of 20 nm to 200 nm and typically spans the entire surface. Preferably, one or more of the precursor gold layer, the first metal layer, the second metal layer, and the third metal layer, and optionally all of them, are formed using electron beam, sputtering, or vapor deposition techniques.
[0098] Simultaneous patterning of a two-dimensional material layer structure and a gold layer to form a co-patterned stack can be performed using conventional techniques such as photolithography and etching. Co-patterning means that the gold layer and the two-dimensional material layer structure have overlapping shapes with aligned edges. Typically, a mask is patterned onto the surface of the stack using a photoresist, and the two layers of the stack can be etched in a single step or in two steps (one step for each layer). Prior to the etching step, the patterned photoresist can be heated (which may be referred to as “curing” or “baking”) to improve adhesion to the gold layer. Such a heating step for curing the photoresist is well known in the art and may include heating to about 180°C, or to 160°C, for example, 80°C to 140°C, for up to 10 minutes, for example, 1 minute to 3 minutes. After etching, the cured photoresist can be washed away using a suitable solvent.
[0099] At this stage, contact stacks can be deposited to contact the edges of the two-dimensional material layer structure and the gold layer. The contact stacks can provide the desired circuitry for the final device. That is, additional metal can be used to provide circuit "traces," and preferably to provide gate contacts on the substrate. Similarly, contact stacks can be formed by photolithography using photoresist to pattern the metal through a lift-off process, wherein a first metal layer, a second metal layer, and a third metal layer are sequentially deposited to their desired thicknesses.
[0100] Etching of the gold layer is preferably performed by a wet etching method. Suitable methods for etching and etching solutions for wet etching of gold are well known. Gold is particularly advantageous in this regard because it can be etched with relatively mild solutions. The most common are solutions of potassium iodide and / or iodine, which are resistant to etching of the third metal layer by such solutions. Various commercial etchants (sometimes referred to as developing solutions) are also readily available. Preferably, the pH of the etching solution is greater than 5 and / or less than 9, for example, from about 7 to about 8, thereby avoiding undue damage to the two-dimensional material upon exposure.
[0101] The etching time and temperature can be varied to control the degree of etching. Typically, wet etching can last up to 30 minutes, for example, from 1 to 10 minutes, or even 2 to 5 minutes. That is, wet etching can be performed by immersing the entire substrate in the etchant solution for such a time. Etching can be performed at room temperature (e.g., from about 15°C to about 25°C), or it can be gently heated, for example, up to 50°C. The etchant solution is then washed away, the device structure is rinsed (e.g., with deionized water), and the structure is dried. Drying can be accelerated by gentle heating (e.g., up to 50°C) and / or by a flow of a dry inert gas (e.g., nitrogen).
[0102] Attached Figure
[0103] The invention will now be further described with reference to the following non-limiting drawings, in which:
[0104] Figures 1 to 3 An embodiment of the method of the present invention is shown in cross section.
[0105] Figure 4 This is an image of an exemplary substrate on which a device precursor manufactured has been layered.
[0106] Figures 1 to 3 These are schematic diagrams illustrating the invention, and the relative dimensions and shapes of some features / layers may be enlarged and / or simplified, and are not presented to scale to improve clarity. Relative dimensions and sizes do not necessarily correspond to those of the actual embodiments of the invention. In all figures, the same features are given the same reference numerals.
[0107] Figures 1 to 3An exemplary method according to the invention for manufacturing apparatus 200 is shown in cross-section. A two-dimensional material layer structure (e.g., a monolayer 210 of graphene) is disposed on and spans the surface of a substrate 205 (e.g., sapphire). The graphene monolayer 210 can be formed directly on the substrate 205 by CVD, avoiding the need for physical transfer.
[0108] In step 100, a main gold metal layer 215 is deposited over and across the entire surface of graphene 210, for example, with a thickness of about 100 nm. Then, in step 105, a photoresist 220 is coated over and across the gold layer 215, and then patterned into a patterned photoresist 220' in step 110, followed by baking. The photoresist 220 can be patterned using conventional photolithography techniques, thereby exposing a portion of the underlying gold layer 215. In step 115, the exposed portion of the gold layer 215 and the corresponding underlying portion of the graphene 210 are then etched away to leave a co-patterned stack of the gold layer 215' and the graphene 210' with opposing edges 210a', 210b'. Finally, Figure 1 In step 120, the patterned photoresist 220' is washed away to leave the co-patterned stacks 210' and 215', whereby the patterned gold layer 215' has the same shape and overlapping edges as the patterned graphene 210'.
[0109] like Figure 2 As shown, in step 125, the stack on substrate 205 is then coated with an additional photoresist 225, which is subsequently patterned in step 130 to expose portions of the edges of the adjacent stack adjacent to the patterned gold layer 215' below, leaving the patterned additional photoresist layer 225'. Figures 1 to 3 As shown in the specific cross-section, the relative edges of the stack of patterned graphene 210' are exposed.
[0110] In step 135, a first metal 230 (formed of titanium and / or chromium) for contacting the stack is deposited to a thickness of, for example, about 10 nm. Subsequently, in step 140, a second metal layer 235 comprising gold is deposited to a thickness of about 80 nm, followed by the deposition of a platinum layer 240 with a thickness of about 30 nm.
[0111] like Figure 3As shown, the patterned additional photoresist layer 225' is subsequently removed in a conventional "stripping" step 150, thereby removing the stack of the first, second, and third metals deposited on the patterned additional photoresist layer 225' to leave a portion of the metal layers 230', 235', 240' forming the contact stack 230'. Each portion contacts the opposing edges 210a', 210b' of the graphene 210' via the first metal layer 230' and is located on the substrate 205, and can serve as source and drain contacts. Elsewhere on the substrate 205, the stacks 230', 235', 240' of the first, second, and third metals provide desired circuitry for integrating the structure into a larger device including gate contacts (not shown).
[0112] The stripping process exposes a window leading to the gold layer 215' that lies entirely above the graphene 210'. Finally, in step 155, the patterned gold layer 215' is etched using a gold etching solution (e.g., a solution of potassium iodide and iodine in water). Etching step 155 exposes the upper surface 210c' of the patterned graphene monolayer 210', thereby forming a device 200 suitable for use as a sensor.
[0113] Figure 4 This is an image of an exemplary wafer on which a precursor device precursor has been layered. For clarity and reproducibility, dark backgrounds (e.g., substrates) have been removed from the image. The precursor device includes a patterned graphene monolayer 310' co-patterned with a (primary) gold layer 315' thereon, and contact stacks 330', 335', 340' comprising a first titanium layer 330', a second (or secondary) gold layer 335', and a third platinum layer 340' with a thickness of 40 nm. In this example, the platinum layer is too thick and introduces too much stress, which overcomes the adhesion of the graphene monolayer 310' to the substrate. Therefore, in two units ( Figure 4 In the approximate upper right and lower left sections, the graphene precursor device is fully decomposed, exposing the substrate 305 between the two traces of the contact stack 330', 335', and 340', while in the other two units ( Figure 4 In the upper left and lower right of the image, the edges of the patterned graphene 310', which has a portion of the contact stacks 330', 335', and 340', are layered, exposing the underlying substrate 305.
[0114] Unless the context clearly indicates otherwise, as used herein, the singular forms “an,” “a,” and “described” include plural referents. The use of the term “comprising” is intended to be interpreted as including such features but not excluding other features, and also as including options for features necessarily limited to those described. In other words, unless the context clearly indicates otherwise, the term also includes the limitations of “consistently composed of” (intended to mean that certain additional components may be present, provided they do not substantially affect the essential characteristics of the described feature) and “composed of” (intended to mean that no other features may be included such that, if the components were expressed as percentages, they would add up to 100%, taking into account any unavoidable impurities).
[0115] It should be understood that while the terms “first,” “second,” “main,” etc., may be used herein to describe, for example, different metals, layers, and / or portions, the metals, layers, and / or portions should not be limited by these terms. These terms are used only to distinguish one metal, layer, or portion from another metal, layer, or portion, or further metals, layers, or portions. It should be understood that the term “on” is intended to mean “directly on,” such that when one material is said to be “on” another material, there is no intermediate layer between them. For ease of description, spatially relative terms such as “below,” “under,” “below,” “lower,” “above,” “upper,” “upper,” etc., may be used herein to describe the relationship of one element or feature to one or more other elements or features. It should be understood that spatially relative terms are intended to cover different orientations of the device in use or operation other than those depicted in the accompanying drawings. For example, if the device as described herein is flipped, an element described as “below” or “under” other elements or features will be oriented “above” or “above” other elements or features. Therefore, the exemplary term "under" can encompass both above and below orientations. Devices can be oriented in other ways, and the spatial relative descriptions used herein are interpreted accordingly.
[0116] The lower and upper limits of the values for the features described herein can preferably be combined to provide a closed range.
[0117] The foregoing detailed description has been provided by way of illustration and example and is not intended to limit the scope of the appended claims. Many variations of the preferred embodiments of the invention shown herein will be apparent to those skilled in the art and remain within the scope of the appended claims and their equivalents.
[0118] To avoid any ambiguity, the full contents of all documents identified in this document are incorporated herein by reference.
Claims
1. A semiconductor device, comprising: It has a substrate with a patterned two-dimensional material layer structure; A gold layer on the portion of the two-dimensional material layer structure adjacent to its edge; as well as A contact stack adjacent to the edge of the two-dimensional material layer structure and overlapping the gold layer, the contact stack comprising: (i) A first metal layer on a portion of the substrate and on the gold layer, wherein the first metal is selected from titanium and chromium; (ii) a second metal layer on the first metal layer, wherein the second metal is selected from aluminum, copper, gold, nickel, and palladium; and (iii) A third metal layer on the second metal layer, wherein the third metal is selected from cobalt, chromium, iridium, iron, magnesium, niobium, platinum, ruthenium, silver, tantalum, titanium, titanium nitride and tungsten.
2. The semiconductor device according to claim 1, wherein the thickness of the third metal layer is less than 35 nm, preferably 15 nm to 30 nm.
3. The semiconductor device according to claim 1 or claim 2, wherein the thickness of the second metal layer is at least 50 nm and / or at most 100 nm, preferably 70 nm to 90 nm.
4. The semiconductor device according to any of the preceding claims, wherein the thickness of the gold layer is from 20 nm to 200 nm, preferably from 80 nm to 120 nm.
5. The semiconductor device according to any of the preceding claims, wherein the thickness of the first metal layer is less than 15 nm, preferably 3 nm to 10 nm.
6. The semiconductor device according to any of the preceding claims, wherein the third metal is platinum.
7. The semiconductor device according to any of the preceding claims, wherein the second metal is gold.
8. The semiconductor device according to any of the preceding claims, wherein the third metal layer is conformal.
9. The semiconductor device according to any of the preceding claims, wherein the contact stack on the substrate further extends to form connection traces for circuitry.
10. The semiconductor device of claim 9, wherein the circuitry includes a gate contact.
11. The semiconductor device according to any of the preceding claims, wherein the two-dimensional material is graphene.
12. The semiconductor device according to any of the preceding claims, wherein the two-dimensional material layer structure is CVD grown on the substrate.
13. The semiconductor device according to any of the preceding claims, wherein the contact stack has a thickness extending vertically from the substrate by a distance of 80 nm to 200 nm.
14. The semiconductor device according to any of the preceding claims, wherein the substrate is sapphire or a rare earth oxide on silicon.
15. A method of forming a semiconductor device, wherein the method comprises: (a) Forming a precursor, said precursor comprising: It has a substrate with a patterned two-dimensional material layer structure; The precursor gold layer on the two-dimensional material layer structure; and A contact stack adjacent to the edge of the two-dimensional material layer structure and overlapping the precursor gold layer, the contact stack comprising: (i) A first metal layer on a portion of the substrate and on the precursor gold layer, wherein the first metal is selected from titanium and chromium; (ii) a second metal layer on a titanium and / or chromium layer, wherein the second metal is selected from aluminum, copper, gold, nickel, and palladium; and (iii) A third metal layer on the second metal layer, wherein the third metal is selected from cobalt, chromium, iridium, iron, magnesium, niobium, platinum, ruthenium, silver, tantalum, titanium, titanium nitride, and tungsten; and (b) Etching the precursor gold layer with a gold etchant to form the device according to any of the preceding claims.
16. The method of claim 15, wherein the semiconductor device has an exposed portion of the two-dimensional material layer structure.
17. The method of claim 15 or claim 16, wherein one or more of the precursor gold layer, the first metal layer, the second metal layer and the third metal layer, preferably all of them, are formed by electron beam, sputtering or vapor deposition techniques.
18. The method according to any one of claims 15 to 17, wherein the gold etchant is an aqueous solution containing potassium iodide and / or iodine.
19. The method of claim 18, wherein the pH of the etching solution is greater than 5 and / or less than 9.
20. The method of any one of claims 15 to 19, wherein the contact stack on the substrate further extends to form connection traces for a circuit including a gate contact, and the precursor further includes a patterned dielectric layer on and across the contact stack and the precursor gold layer, the dielectric layer being patterned to expose a portion of the precursor gold layer and the gate contact.
Citation Information
Patent Citations
Gan-based vertical structure LED applying graphene film current expansion layer
US20140151632A1
Depositing a passivation layer on a graphene sheet
US20170365474A1
Providing a temporary protective layer on a graphene sheet
US20170365477A1
Patterning graphene with a hard mask coating
US20170365562A1
A method of producing a two-dimensional material
WO2017029470A1
