The background description provided here is for the purpose of generally presenting the context of the disclosure. Information described in this background section, as well as aspects of the description that may not otherwise qualify as prior art at the time of filing, are neither expressly nor impliedly admitted as prior art against the present disclosure.
The disclosure relates to methods of forming semiconductor devices on a semiconductor wafer. More specifically, the disclosure relates to the selective etching of silicon oxide (SiO2) with respect to other materials. In the specification and claims the etching of SiO2 includes the etching of SiO2 based materials, such as borophosphosilicate glass (BPSG, a silicon oxide glass doped with boron and phosphorous) or carbon doped silicon oxide, such as SiCOH or other low-k silicon oxide based materials. Such SiO2 based materials may be other types of doped SiO2.
The smallest feature dimensions of semiconductor devices are constantly shrinking to follow Moore's law. One of these features is the contact between the 1st metal layer and the silicon layer with gates and source/drains (S/D). Such features are typically prepared by a process called self-aligned contact (SAC). In a SAC etch, SiO2 is etched to form a hole or a trench. Later, the hole or trench is filled with a contact metal. A selective SiO2 etch is employed to selectively etch SiO2 with respect to a spacer material. The spacer material is typically a lower oxygen containing silicon material such as silicon nitride, silicon oxy-nitride, or silicon oxy-carbon-nitride. Many selective etch processes are not selective enough. As a result, too much of the spacer material or corners of the spacer material may be etched increasing electric leakage and device failure rates.
To achieve the foregoing and in accordance with the purpose of the present disclosure, a method for selectively etching at least one feature in a silicon oxide region with respect to a lower oxygen containing region is provided. An etch gas is provided comprising a fluorocarbon gas and at least one of a metalloid halide gas or metal halide gas. The etch gas is formed into a plasma. At least one feature is selectively etched in the silicon oxide region with respect to the lower oxygen containing region, while simultaneously forming a metalloid or metal containing hardmask over the lower oxygen containing region.
These and other features of the present disclosure will be described in more detail below in the detailed description and in conjunction with the following figures.
The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements and in which:
The present disclosure will now be described in detail with reference to a few exemplary embodiments thereof as illustrated in the accompanying drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. It will be apparent, however, to one skilled in the art, that the present disclosure may be practiced without some or all of these specific details. In other instances, well-known process steps and/or structures have not been described in detail in order to not unnecessarily obscure the present disclosure.
The hole or trench of the contacts needs to be very accurately placed in respective to the underlying gate and S/D. Current photolithography tools can only partially meet the placement requirement of the contacts. Therefore, the contact etch can expose the spacer around the gate. Such exposure often leads to corner loss of the space material resulting in an electric leakage.
The selective SiO2 etch can be achieved by the protection of the spacer against an etch using the selective deposition of a carbon-based polymer on the spacer material. However, the lateral dimensions of the hole and trench in the 5 nanometers (nm) node and below can be smaller than 10 nm. More typically, such features may be in the of 5-15 nm range. The aspect ratio can be larger than 6:1. In some embodiments, the aspect ratio may range from 6:1 to 12:1. In addition, the use of a carbon-based polymer mask has a limited etch resistance. The limited etch resistance requires deposition of a thick mask to protect the spacer during the SiO2 etch. Placing thick polymer material on the spacer can clog up and/or pinch-off the contact and block any further etch. This blocked etch would result in an electrical open contact, causing a device failure. Additionally, thick polymer material tends to have deposit different quantities in different parts of the pattern layout, depending on the local shading of the features. This effect, which can be described as pattern loading, leads to undesirable variation in the final results of the etch including the critical dimension (CD) of the trench or hole.
The process and apparatus of this disclosure enable the selective deposition of a metalloid or metal containing mask such as metal carbide on the spacer material during the etch process. This mask is not formed by the initial photolithography but in-situ during the etch process. Due to the higher etch selectivity of a metalloid or metal containing mask compared to the traditional carbon-based polymer mask, a thinner, less than 5 nm hardmask, is sufficient. This metalloid or metal containing hardmask can prevent the corner loss of the spacer while avoiding the pinch-off issues of the traditional polymer material. This thin metalloid or metal containing hardmask may also introduce low variability from pattern loading.
In order to facilitate understanding,
After the structure is provided, a selective pre-etch is provided (step 108) that selectively etches the silicon oxide region 204 and the native silicon oxide layer 216 not covered by the photolithographic mask 220. In this embodiment, the silicon oxide region 204 is selectively etched using an atomic layer etch (ALE). In this embodiment, the ALE provides a reactant gas of hexafluoro-1,3-butadiene (C4F6). The C4F6 forms a polymer deposition layer over the silicon oxide region 204 and the native silicon oxide layer 216. The reactant gas is purged, and an activation gas of argon (Ar) is provided. The Ar activates the deposition layer causing deposited fluorine to selectively etch the silicon oxide region 204 and the native silicon oxide layer 216. The ALE process of a selective deposition and selective etch steps may be repeated for a plurality of cycles.
After the selective pre-etch is completed (step 108), a selective etch of the silicon oxide region 204 is provided while depositing a metalloid or metal containing hardmask (step 112).
After the etching of the silicon oxide region 204 is complete, the metalloid or metal containing hardmask 224 is removed. In this embodiment, a wet clean is used to remove the metalloid or metal containing hardmask 224. In the embodiment, an aqueous solution of ammonia (NH3) and hydrogen peroxide (H2O2) is used to selectively remove the metalloid or metal containing hardmask 224 with respect to the silicon oxide region 204.
This embodiment provides a highly selective etch of silicon oxide with respect to lower oxygen containing regions. In various embodiments, the depth to width aspect ratio of the feature 228 etched in the silicon oxide region 204 is at least 6:1. For example, the feature 228 has a depth to width aspect ratio is between 6:1 and 12:1. In this example, the feature 228 has a width of less than 10 nm. In various embodiments, the feature 228 has a width of between 6 to 15 nm. The use of the metalloid or metal containing hardmask 224 prevents or reduces the etching of the SiOCN region 208, so that corner loss of the SiOCN region 208 is prevented or reduced. In addition, since the metalloid or metal containing hardmask 224 is kept thin, clog up and pinch-off of the feature 228 is prevented. In some embodiments, the partial etch of the pre-etch step (step 108) provides a geometry, where the silicon oxide region 204 is recessed. In some embodiments, the recessed initial geometry of the silicon oxide region 204 allows for more selective deposition of the metalloid or metal containing hardmask 224 and selective etch of the silicon oxide region 204.
The etch resistance of the metalloid or metal containing hardmask 224 is high because a metalloid or metal containing material such as metal carbide has a very high melting and boiling point. In other embodiments, other metalloid or metal containing hardmasks made of metal nitride, metal boride, or metal silicide can be used instead of metal carbide or together with metal carbide as long as the metalloid or metal containing hardmasks have a high etch resistance. Since this embodiment provides a metalloid or metal containing hardmask 224 with a high etch resistance, the mask thickness may be less than 5 nm. This thin hardmask enables the etch of very narrow features <15 nm by avoiding the clogging risk.
In alternative embodiments, near the end of the etching of the silicon oxide region 204, the flow of the metalloid or metal containing precursor is reduced or stopped so that the deposition of the metalloid or metal containing hardmask 224 is reduced or stopped. In one embodiment, at the end of the etch of the silicon oxide region 204 the metalloid or metal containing hardmask 224 is etched away so that removal of the metalloid or metal containing hardmask (step 116) is not needed. In another embodiment, a dry etch is used to selectively remove the metalloid or metal containing hardmask (step 116) with respect to the silicon oxide region 204.
In another embodiment, the metalloid or metal containing precursor and a halogen containing component is molybdenum hexafluoride (MoF6). In other embodiments, the metalloid or metal containing precursor may be other metalloid halides or metal halides. In such embodiments, metalloids, such as boron (B), silicon (Si), germanium (Ge), antimony (Sb), or metals, such as tin (Sn), titanium (Ti), zirconium (Zr), hafnium (Hf), vanadium (V), niobium (Nb), tantalum (Ta), aluminum (Al), gallium (Ga), indium (In), iron (Fe), ruthenium (Ru), rhenium (Re), tungsten (W), molybdenum (Mo), and bismuth (Bi), may be used if a metalloid or metal halogen exists to allow the metalloid or metal halide to be flowed as a gas or vapor. In some embodiments, a liquid metal halide may be heated to form a metal halide vapor. In some embodiments, a solid metal halide may be heated to sublime to form a metal halide vapor. Metalloid or metal bromides may be used in other embodiments. In other embodiments, the precursor may be metalloid or metal oxyhalides.
In other embodiments, other lower oxygen regions may be used in place of the lower oxygen silicon containing regions. For example, silicon germanium (SiGe), germanium (Ge), elemental metal or metal nitrides may form the lower oxygen regions and may be protected, so that SiO2 may be selectively etched with respect to these materials.
In some embodiments, the etch gas further comprises an oxygen containing component. In some embodiments, the oxygen containing component comprises at least one of oxygen (O2), ozone (O3), carbon dioxide (CO2), carbon monoxide (CO), nitric oxide (NO), nitrogen dioxide (NO2), nitrous oxide (N2O), sulfur dioxide (SO2), sulfur trioxide (SO3), and carbonyl sulfide (COS). In various embodiments, the etch gas further comprises an inert gas. In some embodiments, the inert gas is selected from the group consisting of nitrogen, helium, argon, krypton, xenon, and neon. In various embodiments, a plasma is formed at a pressure of 1-500 millitorr with a power of 20-1000 Watts at radio frequencies such as one or more of 13.56 megahertz (MHz), 60 MHz, 27 MHz, 2 MHz, 1 MHz, 400 kilohertz (kHz), or combinations thereof.
In various embodiments, the selective pre-etch (step 108) allows the removal of any native silicon oxide layer 216. If the lower oxygen containing region does not form a native oxide layer, the selective pre-etch (step 108) may not be needed. The hardmask may be selectively deposited without the selective pre-etch (step 108).
Another embodiment is shown by the flow chart shown in
In this embodiment, after the silicon oxide region has been partially etched leaving a metalloid or metal containing hardmask, the silicon oxide region is further or additionally etched by an atomic layer etch (ALE) process utilizing the metalloid or metal containing hard mask (step 416). In this embodiment, the ALE provides a reactant gas of C4F6. The C4F6 forms a polymer deposition layer over the silicon oxide region and the metalloid or metal containing hardmask. The reactant gas is purged, and an activation gas of Ar is provided. The Ar activates the deposition layer causing deposited fluorine to selectively etch the silicon oxide region with respect to the metalloid or metal containing hard mask. The ALE process of a selective deposition and selective etch steps may be repeated for a plurality of cycles.
Since the ALE process does not deposit the metalloid or metal containing hardmask, the metalloid or metal containing hardmask may be consumed during the ALE process. If the metalloid or metal containing hardmask is sufficiently consumed before etching of the silicon oxide region is completed, the ALE process (step 416) may be stopped and the selective etch of the silicon oxide region while depositing a metalloid or metal containing hardmask (step 412) may be repeated (step 420) in order to replenish the metalloid or metal containing hardmask. After the metalloid or metal hardmask has been replenished, the selective etching of the silicon oxide region while depositing a metalloid or metal containing hardmask (step 412) is stopped. The ALE process may be used to further etch the silicon oxide region (step 416). The steps of selectively etching the silicon oxide while depositing the metalloid or metal containing hardmask (step 412) and the ALE using the metalloid or metal containing hardmask (step 416) may be cyclically repeated until the feature in the silicon oxide region is sufficiently etched. Once the feature in the silicon oxide region is sufficiently etched the cyclical process may be stopped. Since the ALE process (step 416) consumes the metalloid or metal hardmask in this embodiment, a separate metal or metalloid hardmask removal step is not needed.
In some embodiments, the deposition of the hardmask and etching may be done in separate sequential steps. However, simultaneously depositing the hardmask and etching in a single continuous step may provide a faster process.
It has been unexpectedly found that providing a hydrogen free selective etch of the silicon oxide region while depositing a metalloid or metal containing hardmask reduces pattern loading, while providing sufficient protection of corners of regions with lower concentration of oxygen. Preventing pattern loading provides the ability to achieve similar final top CD for a range of differently shaped features within a single die. In practice, there is typically smaller CD in the more exposed features and larger CD in the features which are locally shaded, e.g., by a cut mask. Providing a hydrogen free selective etch provides a CD that is less dependent on shading. Higher aspect ratio features generally have more shading. So, providing a process where CD is less dependent on shading provides an improved process for higher aspect ratio features. Without wishing to be bound by theory, it is proposed that a process gas which comprises the hydrogen element in one or more gases tends to produce carbon-bearing polymer precursors with a higher sticking coefficient. Neutral deposition from these precursors tends to have more feature-dependent and aspect-ratio dependent effects. Conversely, a hydrogen free process tends to have carbon-bearing precursors with a lower sticking coefficient, with less resultant loading effects. Using a fluorocarbon instead of a hydrofluorocarbon or hydrocarbon helps to provide a hydrogen free process. In other embodiments, other fluorocarbons (CxFy), such as C4F6, perfluorocyclobutane (c-C4F8), octafluorocyclopentene (c-C5F8), tetrafluoroethylene (C2F4), carbon tetrafluoride (CF4), octofluoropropane (C3F8), and hexafluoroethane (C2F6) may be used. Some embodiments allow for thinner passivation layers that help to control breadloafing issues in more exposed features. In some embodiments, 59 nm of SiO may be etched with less than 3 nm of SiN etched.
To provide an example of a controller 535 in an embodiment,
Information transferred via communications interface 614 may be in the form of signals such as electronic, electromagnetic, optical, or other signals capable of being received by communications interface 614, via a communication link that carries signals and may be implemented using wire or cable, fiber optics, a phone line, a cellular phone link, a radio frequency link, and/or other communication channels. With such a communications interface, it is contemplated that the one or more processors 602 might receive information from a network or might output information to the network in the course of performing the above-described method steps. Furthermore, method embodiments may execute solely upon the processors or may execute over a network, such as the Internet, in conjunction with remote processors that share a portion of the processing.
The term “non-transient computer readable medium” is used generally to refer to media such as main memory, secondary memory, removable storage, and storage devices, such as hard disks, flash memory, disk drive memory, CD-ROM, and other forms of persistent memory, and shall not be construed to cover transitory subject matter, such as carrier waves or signals. Examples of computer code include machine code, such as one produced by a compiler, and files containing higher level code that are executed by a computer using an interpreter. Computer readable media may also be computer code transmitted by a computer data signal embodied in a carrier wave and representing a sequence of instructions that are executable by a processor.
While this disclosure has been described in terms of several exemplary embodiments, there are alterations, modifications, permutations, and various substitute equivalents, which fall within the scope of this disclosure. It should also be noted that there are many alternative ways of implementing the methods and apparatuses of the present disclosure. It is therefore intended that the following appended claims be interpreted as including all such alterations, modifications, permutations, and various substitute equivalents as fall within the true spirit and scope of the present disclosure. As used herein, the phrase “A, B, or C” should be construed to mean a logical (“A OR B OR C”), using a non-exclusive logical “OR,” and should not be construed to mean ‘only one of A or B or C. Each step within a process may be an optional step and is not required. Different embodiments may have one or more steps removed or may provide steps in a different order. In addition, various embodiments may provide different steps simultaneously instead of sequentially.
This application claims the benefit of priority of U.S. Application No. 63/284,877, filed Dec. 1, 2021, which is incorporated herein by reference for all purposes.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2022/051098 | 11/28/2022 | WO |
Number | Date | Country | |
---|---|---|---|
63284877 | Dec 2021 | US |