Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
DETAILED ACTION
Election/Restrictions
Applicant's election filed in response to the Restriction Requirement of April 23, 2026, has been considered.
Applicant elected, without traverse, Group I, drawn to the two-step etch process, and further elected, without traverse, Species I-A, drawn to the HCl-based acidic etch. Applicant identified claims 1–3, 21–23, 28, and 29 as readable on the elected Species I-A, and identified claims 1, 21, and 28 as generic claims.
Accordingly, claims 1–3, 21–23, 28, and 29 are examined on the merits in this action.
Claims 8–20 have been canceled.
Claims 4–7 and 24–27 are withdrawn from consideration as being directed to nonelected species.
Newly added claims 30–33 were not identified by Applicant as readable on the elected Species I-A. In addition, claims 30–33 recite limitations directed to dielectric thickness variation in corner regions and/or transition-metal-nitride material intermixing and structural distribution. Such limitations are directed to the structural subject matter corresponding to the nonelected Group III invention identified in the Restriction Requirement.
Accordingly, claims 30–33 are withdrawn from consideration as being directed to a nonelected invention. See 37 CFR 1.142(b) and MPEP § 821 et seq.
Therefore, the status of the claims is as follows:
• Claims 1–3, 21–23, 28, and 29 are examined on the merits.
• Claims 4–7, 24–27, and 30–33 are withdrawn from consideration.
• Claims 8–20 are canceled.
The election of Group I and Species I-A was made without traverse.
Specification
The specification is objected to because number of figures submitted does not match the number of figures listed under Brief Description of Drawings in the specification. All of the figures with alphabets should be listed separately. For example, ‘Figs. 1A-1C’ should be ‘Figs. 1A, 1B and 1C’.
In particular, ‘Figures 2A-13’ in the paragraph [0004] is objected.
See MPEP 500 - Receipt and Handling of Mail and Papers, MPEP 507 - Drawing Review in the Office of Patent Application Processing (OPAP). This labeling convention ensures clarity and consistency in referencing figures throughout the patent application and publication. Improper labeling may result in an objection from OPAP and require correction.
Appropriate correction is required.
The specification is objected to because of a grammatical error in paragraph [0004].
Paragraph [0004] recites “an IC device of at various stages of fabrication.” The word “of” appears to be extraneous.
The phrase should be corrected to “an IC device at various stages of fabrication.”
Appropriate correction is required.
The specification is objected to because of a grammatical error in paragraph [0011].
Paragraph [0011] recites “In other example.” The phrase should be corrected to “In another example.”
Appropriate correction is required.
The specification is objected to because of a grammatical error in paragraph [0013].
Paragraph [0013] recites “which can beneficial or detrimental effects on threshold voltage....” The phrase is grammatically incomplete and appears to be missing a verb.
Applicant is required to amend the specification to provide a grammatically complete and clear statement, for example, “which can have beneficial or detrimental effects on threshold voltage....”, if such wording reflects Applicant's intended disclosure.
Appropriate correction is required.
The specification is objected to because of an apparent typographical inconsistency in paragraph [0049].
Paragraph [0049] states that the nanostructures 24 are selectively recessed. However, the paragraph subsequently states that the inner spacer layer fills “recesses in the nanostructures 22 formed by the previous selective etching process.”
Since the preceding selective etching process is described as recessing the nanostructures 24, the reference to “nanostructures 22” appears to be inconsistent and should be corrected to “nanostructures 24.”
Appropriate correction is required.
The specification is objected to because of improper chemical-formula notation and an apparent typographical error in paragraph [0066].
Paragraph [0066] recites “HfCl4 and/or H2O as precursors.” The chemical formulas should be properly presented with subscripts.
Accordingly, “HfCl4” should be corrected to “HfCl₄,” and “H2O” should be corrected to “H₂O.”
Appropriate correction is required.
The specification is objected to because of an inconsistency in paragraph [0078].
Paragraph [0078] refers to the “second region 620,” but subsequently states that the second region “may be a PFET region 610 of the device 10.” Thus, the same region appears to be identified by two different reference numerals, 620 and 610.
Applicant is required to amend paragraph [0078] to provide the correct and consistent reference numeral for the referenced region.
Appropriate correction is required.
The specification is objected to because of an inconsistency in paragraph [0080].
Paragraph [0080] refers to the “first region 620.” However, elsewhere in the specification, including within paragraph [0080], the first region is identified by reference numeral 610, whereas the second region is identified by reference numeral 620.
Accordingly, “first region 620” appears to be inconsistent and should be corrected to “first region 610,” if this reflects Applicant's intended disclosure.
Appropriate correction is required.
The specification is objected to because of a grammatical error in paragraph [0091].
Paragraph [0091] states:
“The third etch operation may be performed for a length of time is shorter than that of the first etch operation.”
The phrase “for a length of time is shorter than” is grammatically incorrect and should be corrected to “for a length of time that is shorter than.”
Accordingly, the sentence should read:
“The third etch operation may be performed for a length of time that is shorter than that of the first etch operation.”
Appropriate correction is required.
Claim Objections
Claim Objection – Claim 22:
Claim 22 is objected to because of an informality. Claim 22 recites “HCl:H2O2,” which contains typographical errors in the chemical formulas.
“HCl:H2O2” should be corrected to “HCl:H₂O₂”, consistent with the specification.
Appropriate correction is required.
Claim Objection – Claim 29:
Claim 29 is objected to because of an informality. Claim 29 recites “HCl:H2O2,” which contains typographical errors in the chemical formulas.
“HCl:H2O2” should be corrected to “HCl:H₂O₂”, consistent with the specification.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 1 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20210210388; “Zhang”; in the IDS on 1/30/25) in view of Lee et al. (US 20050000942; “Lee-942”).
Regarding claim 1, Zhang teaches a method comprising forming a stack of alternating first semiconductor channels and second semiconductor layers on a substrate, wherein nanosheet stack 20 is formed over semiconductor substrate 10 and includes alternating semiconductor layers 22 and 24 (Fig. 1; [0027], [0032]). Semiconductor layers 22 may comprise SiGe and semiconductor layers 24 may comprise Si.
Zhang teaches releasing the first semiconductor channels by removing the second semiconductor layers, wherein sacrificial semiconductor layers 22, e.g., SiGe layers, are removed while semiconductor layers 24, e.g., Si layers, remain as released semiconductor nanosheets/channels (Fig. 2; [0033]–[0035]). Accordingly, Zhang's semiconductor layers 24 correspond to the claimed first semiconductor channels, and Zhang's sacrificial semiconductor layers 22 correspond to the claimed second semiconductor layers.
Zhang teaches forming a gate dielectric on the first semiconductor channels, wherein interfacial layer (IL) 30 and high-k gate dielectric 32 are formed adjacent the remaining semiconductor layers 24 after removal of sacrificial semiconductor layers 22 (Fig. 2; [0034]–[0035]).
Zhang teaches forming a transition metal nitride layer on the gate dielectric, wherein first work-function-metal (WFM) layer 34 is deposited on high-k gate dielectric 32 (Fig. 3; [0037]–[0039]). Zhang expressly teaches that WFM layer 34 may comprise titanium nitride (TiN) or tantalum nitride (TaN) and, in a preferred embodiment, comprises TiN.
Zhang further teaches exposing the gate dielectric in a first region of the substrate by removing the transition metal nitride layer, wherein first WFM layer 34 is selectively removed from selected regions, including structures 5 and 5′ (Fig. 4; [0042]–[0045]). Removal of WFM layer 34 results in exposure of IL 30 and high-k gate dielectric 32 in structures 5 and 5′.
Thus, Zhang teaches selectively removing a transition-metal-nitride layer, such as TiN or TaN, from a selected transistor region to expose the underlying gate dielectric.
But, Zhang does not expressly teach that the removing of the transition metal nitride layer includes performing a first etch using an acidic etchant and, after performing the first etch, performing a second etch using an oxidizing etchant.
However, Lee teaches the missing sequential two-step wet-etch technique in semiconductor fabrication.
Lee teaches removing unreacted cobalt by wet etching and expressly teaches that continued etching removes portions of the underlying metal nitride layer 24 [0027]. Thus, Lee expressly associates its wet-etch processing with removal of an underlying metal nitride layer, rather than merely removal of an overlying metal.
Lee further teaches that the wet etching may be carried out using “two solutions in a two step process,” wherein a first solution includes a mineral acid and a peroxide, followed by a second solution including a peroxide [0028].
Lee teaches that HCl is a preferred mineral acid [0031] and that H₂O₂ is a preferred peroxide [0032]. Thus, Lee teaches a first etch employing an acidic etchant and, thereafter, a second etch employing an oxidizing peroxide-containing etchant.
Lee further teaches that the second solution of the two-step process is used for removal of metal nitride portions and comprises peroxide and deionized water [0035].
Lee additionally provides a specific embodiment in which a first solution including HCl and H₂O₂ is employed, followed by a second solution including H₂O₂, wherein the second solution is used to etch TiN [0046].
Lee is also directed to semiconductor transistor fabrication rather than an unrelated chemical-treatment environment. For example, [0049] expressly describes fabrication of a local interconnect for an SRAM device including source region 87, drain region 85, and gate structure 86, wherein gate structure 86 includes gate oxide 91, polysilicon region 92, metal silicide region 93, and spacers 97.
Accordingly, Lee teaches, in the context of semiconductor transistor fabrication, a wet-etch process that proceeds to removal of a metal nitride layer and teaches carrying out the wet etching using a sequential two-step process comprising an acidic first etchant followed by an oxidizing peroxide-containing second etchant, with the second etchant expressly being used to remove metal nitride, including TiN.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Zhang's process for selectively removing transition-metal-nitride WFM layer 34 to employ the sequential two-step wet-etch process taught by Lee.
First, Zhang and Lee are in the same field of endeavor, namely fabrication of semiconductor transistor devices. Zhang concerns fabrication of gate-all-around nanosheet transistors having gate dielectric and work-function-metal layers. Lee likewise concerns semiconductor device fabrication and expressly describes transistor structures having source and drain regions and a gate structure including a gate oxide, including an SRAM implementation [0049].
Second, and more particularly, the references address closely related metal-nitride removal operations. Zhang expressly requires selective removal of WFM layer 34, which may comprise TiN or TaN, to expose the underlying gate dielectric. Lee expressly teaches wet etching that proceeds to removal of an underlying metal nitride layer [0027] and specifically identifies TiN as a metal nitride removed by its wet-etch chemistry [0046]. Thus, Lee's teaching is directly pertinent to the material-removal operation required by Zhang and is not merely a general teaching concerning unrelated wet etching.
Third, Lee provides a known manner of carrying out such wet processing as a two-solution, two-step etch sequence. Lee teaches an acidic mineral-acid/peroxide first solution followed by an oxidizing peroxide-containing second solution ( [0028], [0031]–[0032]), with the second solution being used for metal-nitride removal [0035] and specifically for TiN etching [0046].
Therefore, one of ordinary skill in the art, faced with Zhang's express requirement to selectively remove a TiN WFM layer 34 to expose the underlying gate dielectric, would have had reason to look to and employ Lee's known wet-etch technique for metal-nitride/TiN removal. Applying Lee's sequential wet-etch technique to Zhang's TiN-removal operation would have resulted in performing a first etch using an acidic etchant followed by a second etch using an oxidizing etchant, as claimed.
The proposed modification would have involved applying Lee's known sequential wet-etch technique to the same type of material—TiN—that Zhang expressly identifies as a WFM material to be selectively removed. A person of ordinary skill would therefore have reasonably expected Lee's known TiN wet-etch technique to be applicable to Zhang's TiN WFM removal and to result predictably in removal of the WFM layer and exposure of the underlying gate dielectric.
Accordingly, Zhang in view of Lee renders claim 1 obvious under 35 U.S.C. 103.
Claims 2-3 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20210210388; “Zhang”; in the IDS on 1/30/25) in view of Lee et al. (US 20050000942; “Lee-942”), and further in view of Lee et al. (US 20020060307; “Lee-307”).
Regarding Claim 2. Zhang in view of Lee-942 teaches or suggests the method of claim 1. But Zhang in view of Lee-942 does not expressly disclose the first etch is performed using an acidic etchant including HCl:H₂O₂:H₂O in a ratio of about 1:10:50.
However, Lee-307 teaches aqueous wet-etch compositions comprising HCl, H₂O₂, and H₂O for semiconductor processing and, more particularly, for etching a transition metal nitride layer comprising TiN.
In particular, Lee-307 [0045] teaches an aqueous solution containing hydrogen peroxide and HCl having a ratio of:
0.1:1:6 (HCl:H₂O₂:H₂O)
which is equivalently expressed as:
1:10:60 (HCl:H₂O₂:H₂O).
Lee-307 further teaches that this composition etches a TiN layer against cobalt silicide at a rate of about 90 Å/min to about 180 Å/min at 65°C.
Thus, Lee-307 expressly teaches the same three components recited in claim 2—HCl, H₂O₂, and H₂O—in substantially the same relative proportions and for the same pertinent purpose of wet etching TiN. The HCl:H₂O₂ proportion disclosed by Lee-307 is identical to that recited in claim 2, namely 1:10, with the only difference being 60 parts water in Lee-307 compared with about 50 parts water recited in claim 2.
Lee-307's expressly disclosed 1:10:60 composition therefore closely approximates the claimed ratio of about 1:10:50. The term “about” indicates that the claimed ratio is not restricted to mathematical exactness.
Moreover, even if “about 1:10:50” is construed sufficiently narrowly that Lee-307's 1:10:60 ratio does not fall within the scope of the claim, the claimed ratio nevertheless would have been obvious to one of ordinary skill in the art.
Lee-307 teaches that concentrated mineral-acid and peroxide solutions are diluted by volume in deionized water in desired proportions and discloses various proportions for the mineral-acid/peroxide/deionized-water compositions. Lee-307 also teaches determining processing conditions empirically based upon desired etch rates, demonstrating that the etchant conditions were recognized parameters subject to adjustment.
Accordingly, Lee-307 establishes that 1:10:60 HCl:H₂O₂:H₂O is a known composition that actually etches TiN and that the relative dilution/proportions of the etchant components are parameters that may be adjusted.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, when implementing the acidic first etch of Zhang as modified by Lee-942, to employ the HCl/H₂O₂/H₂O TiN wet-etch composition taught by Lee-307 and to adjust the amount of diluent water from 60 parts to about 50 parts, while maintaining the expressly taught 1:10 HCl:H₂O₂ proportion, through routine optimization to obtain a desired etching characteristic.
Such modification represents a relatively small adjustment in dilution of a known 1:10:60 HCl:H₂O₂:H₂O TiN etchant, using the same components for the same pertinent purpose of etching TiN. Lee-307 itself teaches varying the proportions/dilution of those components.
Accordingly, Lee-307 teaches a 1:10:60 HCl:H₂O₂:H₂O composition closely approximating the claimed “about 1:10:50” composition and expressly demonstrates that the composition etches TiN. Even if 1:10:60 is considered outside the scope of “about 1:10:50,” selection of about 1:10:50 would have been an obvious optimization of the known TiN wet-etch composition.
Regarding claim 3, Zhang in view of Lee-942 teaches or suggests the method of claim 1, including performing a first etch using an acidic etchant. But Zhang in view of Lee-942 does not expressly disclose the acidic etchant has HCl at a concentration in a range of about 0.1 wt% to about 50 wt%.
However, Lee-307 teaches HCl as a suitable and preferred mineral acid for the disclosed wet-etch compositions and teaches HCl at a concentration of approximately 36% by weight in deionized water [0036]. Lee-307 further teaches that the mineral-acid/peroxide/deionized-water ratios disclosed therein are particularly applicable when the mineral acid is HCl and the peroxide is H₂O₂ [0042].
Accordingly, Lee-307 expressly teaches an HCl concentration of approximately 36 wt%, which falls within the claimed range of about 0.1 wt% to about 50 wt%.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, when implementing the acidic first etch of Zhang as modified by Lee-942, to employ HCl at the concentration taught by Lee-307 because Lee-307 teaches HCl as a suitable mineral acid for semiconductor wet-etch processing involving metal nitride materials.
Claim 21 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20210210388; “Zhang”; in the IDS on 1/30/25) in view of Lee et al. (US 20020060307; “Lee-307”), and further in view of Tai (US 7105458).
Regarding claim 21, Zhang teaches forming a gate dielectric on first semiconductor channels released from a stack of alternating the first semiconductor channels and second semiconductor layers. Zhang teaches nanosheet stack 20 comprising alternating semiconductor layers 22 and 24 formed over substrate 10, selectively removing sacrificial semiconductor layers 22 while leaving semiconductor layers 24 as released nanosheets/channels, and thereafter forming interfacial layer 30 and high-k gate dielectric 32 adjacent the released semiconductor layers 24 (Figs. 1–2; [0027], [0032]–[0035]).
Thus, Zhang's semiconductor layers 24 correspond to the claimed first semiconductor channels, sacrificial semiconductor layers 22 correspond to the claimed second semiconductor layers, and high-k gate dielectric 32 corresponds to the claimed gate dielectric.
Zhang further teaches forming a transition metal nitride layer on the gate dielectric, wherein first work-function-metal (WFM) layer 34 is formed on high-k gate dielectric 32 (Fig. 3; [0037]–[0039]). Zhang expressly teaches that WFM layer 34 may comprise TiN or TaN, and preferably TiN.
Zhang further teaches exposing the gate dielectric in a first region by removing the transition metal nitride layer, wherein WFM layer 34 is selectively removed from structures 5 and 5′, thereby exposing IL 30 and high-k gate dielectric 32 in those regions (Fig. 4; [0042]–[0045]).
Thus, Zhang teaches the claimed nanosheet transistor process, including selective removal of a transition-metal-nitride WFM layer, such as TiN, to expose an underlying gate dielectric.
But Zhang does not expressly teach that the removing comprises performing a first etch using an acidic etchant, wherein the first etch removes at least about 90% of the transition metal nitride layer, followed by a second etch using an oxidizing etchant.
However, Lee-307 teaches wet etching of TiN using an acidic HCl/H₂O₂/H₂O composition. Specifically, Lee-307, [0045] teaches an aqueous etchant having:
HCl:H₂O₂:H₂O = 0.1:1:6
and teaches that this composition, at 65°C, etches a TiN layer at approximately 90 Å/min to 180 Å/min.
Lee-307 therefore establishes that an acidic HCl-containing wet etchant removes TiN, the same transition-metal-nitride material expressly disclosed by Zhang for WFM layer 34.
Lee-307 further teaches empirical selection of processing conditions based upon desired etch rates, demonstrating that TiN removal using the disclosed wet chemistry is controllable through conventional etching parameters.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to employ Lee-307's acidic HCl/H₂O₂/H₂O wet etchant in carrying out Zhang's selective removal of TiN WFM layer 34 because Lee-307 expressly demonstrates that such an acidic etchant removes TiN at known etch rates.
But Zhang in view of Lee-307 does not expressly disclose oxidizing Second Etch.
However, Tai discloses staged TiN Removal and Oxidizing Second Etch. Specifically, Tai further teaches staged removal of a TiN layer using sequential etching operations.
In Tai's second embodiment, a first etching operation removes a portion of TiN film 62, after which the remaining titanium nitride film 62 is removed using an H₂O₂/H₂O etchant.
The subsequent H₂O₂/H₂O etchant constitutes an oxidizing etchant and is expressly used to remove the TiN remaining after the first etching operation.
Tai also demonstrates that the first and second TiN etchants have different removal characteristics. Tai reports a TiN etch rate of approximately 107 Å/min for the first chemistry and approximately 37.4 Å/min for the subsequent H₂O₂/H₂O chemistry.
Tai explains that performing initial TiN removal with the faster first etchant and thereafter removing the remaining TiN with the H₂O₂/H₂O etchant can curtail production time, while the H₂O₂/H₂O chemistry provides protection/selectivity with respect to underlying materials.
Tai further teaches that, within the disclosed concentration range, TiN can be removed by the H₂O₂/H₂O etchant without removing underlying cobalt or CoSi even when the etchant contacts the cobalt or CoSi, thereby permitting the second etch to proceed to the TiN endpoint while protecting the underlying material.
Thus, Tai teaches the relevant process architecture:
first TiN etch → controlled partial TiN removal → oxidizing H₂O₂/H₂O second etch → removal of remaining TiN.
At Least About 90% — Routine Optimization of a Result-Effective Variable
Tai expressly teaches controlling the amount of TiN removed during the first etch through etching time.
In its exemplary process, Tai considers the TiN thickness and TiN etch rate and selects an etching time sufficient to remove a predetermined amount of TiN before changing to the second etchant. Tai specifically calculates the first-etch duration from the known TiN etch rate and the desired amount of TiN to be removed.
Accordingly, Tai expressly recognizes that the extent of TiN removal during the first etch is controllable by adjustment of etching time based upon known film thickness and etch rate.
One of ordinary skill in the art implementing Lee-307's acidic TiN etchant as the first etch in Zhang's TiN WFM removal process therefore would have recognized the percentage of TiN removed during the first etch as a result-effective process variable.
Tai provides recognized reasons for adjusting that variable: performing more TiN removal using the faster first etch can reduce the amount of material that must be removed by the slower subsequent etch and thereby improve processing efficiency, while the subsequent H₂O₂/H₂O etch provides selective removal of the remaining TiN and protection of the underlying material.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, through routine optimization of the first-etch duration, to select the relative amount of TiN removed during the first etch in view of these known considerations of etch rate, processing time, remaining TiN thickness, and selectivity to underlying material.
Selecting the first-etch duration such that the first etch removes at least about 90% of the transition metal nitride layer, leaving about 10% or less for removal by the subsequent oxidizing etch, would have been an obvious selection of the recognized and controllable first-etch removal amount to obtain the predictable result of performing substantially all bulk TiN removal during the first etch while retaining the selective oxidizing second etch for removal of the remaining TiN.
The claimed “at least about 90%” does not change the function of either known etching operation; rather, it specifies the degree to which the known TiN removal is allocated to the first etch. No criticality or unexpected result attributable specifically to the claimed approximately 90% threshold is recited in claim 21.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Zhang's selective removal of TiN WFM layer 34 by employing the acidic TiN wet-etch chemistry taught by Lee-307 together with the staged TiN-removal technique taught by Tai.
The references are directed to closely related semiconductor fabrication operations involving wet-etch removal of TiN. Zhang expressly requires selective removal of a TiN/TaN WFM layer to expose an underlying gate dielectric. Lee-307 expressly teaches an acidic HCl/H₂O₂/H₂O wet etchant that removes TiN at known etch rates. Tai expressly teaches dividing TiN removal between an initial TiN etch and a subsequent oxidizing H₂O₂/H₂O etch that removes the remaining TiN.
Accordingly, one of ordinary skill seeking to perform Zhang's selective TiN WFM removal would have had reason to employ Lee-307's known acidic TiN etchant as the first etching operation and thereafter employ Tai's oxidizing H₂O₂/H₂O etchant to remove remaining TiN. The combination applies known TiN wet-etch techniques to the same material, TiN, for the same general purpose of controlled TiN removal during semiconductor fabrication, with predictable results.
Further, because Tai expressly teaches controlling the first-etch removal amount through known relationships among TiN thickness, etch rate, and etch time, one of ordinary skill would have been able to routinely optimize the first-etch endpoint, including selecting an endpoint at which at least about 90% of the TiN had been removed before performing the subsequent oxidizing etch.
Regarding claim 22, Zhang in view of Lee-307 and Tai teaches or suggests the method of claim 21, including removal of a transition metal nitride layer using a first acidic etch that removes at least about 90% of the transition metal nitride layer, followed by a second oxidizing etch. Claim 22 further requires that:
“the acidic etchant comprises HCl:H₂O₂:H₂O in a ratio of about 1:10:50.”
Lee-307 expressly teaches an aqueous acidic etchant comprising HCl, H₂O₂, and H₂O for wet etching TiN. Specifically, Lee-307 ¶ [0045] teaches:
HCl:H₂O₂:H₂O = 0.1:1:6
which is equivalently expressed as:
HCl:H₂O₂:H₂O = 1:10:60.
Lee-307 further teaches that this composition, at 65°C, etches a TiN layer at a rate of about 90 Å/min to about 180 Å/min.
Thus, Lee-307 expressly teaches the same three components recited in claim 22—HCl, H₂O₂, and H₂O—in substantially the same relative proportions and for the same pertinent purpose of wet etching TiN. The HCl:H₂O₂ proportion disclosed by Lee-307 is identical to that recited in claim 22, namely 1:10, with the only difference being 60 parts water in Lee-307 compared with about 50 parts water recited in claim 22.
Lee-307's expressly disclosed 1:10:60 composition therefore closely approximates the claimed ratio of about 1:10:50. The term “about” indicates that the claimed ratio is not restricted to mathematical exactness.
Moreover, even if “about 1:10:50” is construed sufficiently narrowly that Lee-307's 1:10:60 ratio does not fall within the scope of the claim, the claimed ratio nevertheless would have been obvious.
Lee-307 expressly teaches varying the relative proportions of the mineral acid, peroxide, and deionized water and teaches that the concentrated mineral-acid and peroxide solutions are diluted by volume in deionized water in desired proportions. Lee-307 also teaches determining processing conditions empirically based upon desired etch rates.
Accordingly, Lee-307 establishes both that 1:10:60 HCl:H₂O₂:H₂O is a known composition that actually etches TiN and that the relative proportions/dilution of the etchant components are known adjustable process parameters.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, when implementing the acidic first etch of claim 21 as suggested by the combined teachings of Zhang, Lee-307, and Tai, to adjust the amount of diluent water in Lee-307's known TiN etchant from 60 parts to about 50 parts, while maintaining the expressly taught 1:10 HCl:H₂O₂ proportion, through routine optimization to obtain desired TiN etching characteristics.
Such modification represents only a relatively small adjustment in dilution of a known 1:10:60 HCl:H₂O₂:H₂O TiN etchant, using the same components for the same purpose of wet etching TiN.
Accordingly, Lee-307's 1:10:60 composition closely approximates the claimed “about 1:10:50” composition. Alternatively, even if 1:10:60 is determined not to fall within the reasonable scope of “about 1:10:50,” selection of about 1:10:50 would have been an obvious optimization of Lee-307's known TiN wet-etch composition.
Regarding claim 23, Zhang in view of Lee-307 and Tai teaches or suggests the method of claim 21, including removal of a transition metal nitride layer using a first acidic etch that removes at least about 90% of the transition metal nitride layer, followed by a second oxidizing etch.
Claim 23 further requires that:
“the acidic etchant comprises HCl at a concentration in a range of about 0.1 wt% to about 50 wt%.”
Lee-307 teaches use of HCl as a mineral acid in the disclosed wet-etch compositions. In particular, Lee-307 teaches that commercially available concentrated HCl is approximately 36% by weight in deionized water (Lee-307, ¶ [0036]).
Lee-307 further teaches that the concentrated mineral-acid and peroxide solutions may be diluted by volume in deionized water in desired proportions and that the disclosed mineral-acid/peroxide/deionized-water compositions are particularly applicable when the mineral acid is HCl and the peroxide is H₂O₂.
Accordingly, Lee-307 expressly teaches HCl at a concentration of approximately 36 wt%, which falls directly within the claimed range of about 0.1 wt% to about 50 wt%.
Lee-307 further establishes the applicability of the HCl-containing acidic etchant to the transition metal nitride being removed. Specifically, Lee-307 teaches an aqueous solution containing HCl and H₂O₂ for removal of a metal nitride layer and expressly demonstrates that an HCl:H₂O₂:H₂O composition etches TiN, reporting a TiN etch rate of about 90 Å/min to about 180 Å/min [0045].
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, when implementing the acidic first TiN etch of Zhang as modified by Lee-307 and Tai, to employ HCl at a concentration as taught by Lee-307 because Lee-307 expressly identifies HCl as a suitable mineral acid for the wet-etch composition and demonstrates that an HCl-containing acidic etchant is effective for etching TiN, the same transition metal nitride expressly disclosed by Zhang.
Therefore, Lee-307 teaches an HCl concentration falling squarely within the claimed range and teaches use of the HCl-containing acidic wet-etch chemistry for the same pertinent TiN-removal operation.
Claim 28 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20210210388 A1; “Zhang”; in the IDS on 1/30/25) in view of Lee et al. (US 20050000942; “Lee-942”), and further in view of Wang et al. (US 20200083051; “Wang”).
Regarding claim 28, Zhang teaches a method comprising releasing first semiconductor channels from a stack of alternating the first semiconductor channels and second semiconductor layers by removing the second semiconductor layers. Zhang teaches a nanosheet stack comprising alternating semiconductor layers 22 and 24 formed over substrate 10 and selectively removing sacrificial semiconductor layers 22 while retaining semiconductor layers 24 as released semiconductor channels (Figs. 1–2; [0027], [0032]–[0035]).
Zhang further teaches forming a gate dielectric on the first semiconductor channels, wherein interfacial layer 30 and high-k gate dielectric 32 are formed adjacent the released semiconductor layers 24.
Zhang further teaches forming a transition metal nitride layer on the gate dielectric, wherein first work-function-metal (WFM) layer 34 is formed on high-k gate dielectric 32. Zhang expressly teaches that WFM layer 34 may comprise TiN or TaN, and preferably comprises TiN (Fig. 3; [0037]–[0039]).
Zhang further teaches exposing the gate dielectric in a first region by removing the transition metal nitride layer, wherein WFM layer 34 is selectively removed from selected regions, thereby exposing the underlying IL 30 and high-k gate dielectric 32 (Fig. 4; [0042]–[0045]).
But Zhang does not expressly teach that removal of the transition metal nitride layer comprises performing a first etch using an acidic etchant followed by a second etch using an oxidizing etchant.
However, Lee-942 teaches wet removal of a metal nitride layer during semiconductor fabrication and teaches carrying out such removal using sequential wet-etch chemistries.
Lee-942, [0027] teaches that after removal of unreacted cobalt by wet etching, continued etching removes portions of underlying metal nitride layer 24.
Lee-942 further teaches that the wet etching may be carried out using “two solutions in a two step process,” wherein a first solution includes a mineral acid and a peroxide, followed by a second solution including a peroxide [0028].
Lee-942 identifies HCl as a preferred mineral acid [0031] and H₂O₂ as a preferred peroxide [0032]. Lee-942 further teaches that the second solution of the two-step process is used for removal of metal nitride portions and comprises peroxide and deionized water [0035].
Accordingly, Lee-942 teaches the claimed sequential etching concept comprising a first etch using an acidic etchant followed by a second etch using an oxidizing etchant in connection with semiconductor metal-nitride removal.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Zhang's selective removal of transition-metal-nitride WFM layer 34 to employ the sequential wet-etch process taught by Lee-942.
Zhang and Lee-942 are directed to semiconductor fabrication and address removal of metal nitride materials by wet etching. Zhang expressly teaches selective removal of a WFM layer comprising TiN or TaN, while Lee-942 expressly teaches wet etching that proceeds to removal of an underlying metal nitride layer and teaches carrying out such wet etching using a sequential acidic/peroxide-containing first solution followed by a peroxide-containing second solution.
Thus, one of ordinary skill seeking a wet-etch technique for selectively removing Zhang's TiN WFM layer would have had reason to employ Lee-942's known two-step wet-etch technique because Lee-942 teaches that such sequential wet processing is suitable for metal-nitride removal.
The proposed modification would have amounted to applying a known metal-nitride wet-etch technique to the same type of material being removed in Zhang, with the predictable result of removing the transition-metal-nitride WFM layer and exposing the underlying gate dielectric.
But Zhang in view of Lee-942 does not expressly teach that:
“after the removing the transition metal nitride layer, thickness of the gate dielectric in corner regions of an uppermost channel of the first semiconductor channels is less than thickness of the gate dielectric outside the corner regions.”
Neither Zhang nor Lee-942 expressly addresses the localized effect that wet gate-metal removal can have on the gate dielectric at the corners of nanosheet channels, particularly the uppermost nanosheet.
However, Wang teaches this nanosheet-specific gate-dielectric phenomenon.
Wang is directed to gate-metal patterning/removal in nanosheet semiconductor devices and teaches a stack of nanosheets surrounded by gate dielectric 15 and gate metal 20. Wang further teaches removal of the gate metal using wet chemical etching.
Significantly, Wang teaches that during wet removal of the gate metal, the gate dielectric at the top sheet surface of the uppermost nanosheet and at the corners/edges E1 of the nanosheets can become exposed before gate-metal removal has been completed elsewhere in the nanosheet stack.
Wang further teaches that continued wet etching to remove gate metal remaining in difficult-to-access regions results in the already-exposed gate dielectric at these locations being more aggressively attacked by the over-etch. In particular, Wang identifies the gate dielectric at the edges E1 and the top surface of the uppermost nanosheet as regions susceptible to such excessive wet-etch attack.
Thus, Wang expressly recognizes that wet gate-metal removal in a nanosheet structure produces preferential gate-dielectric erosion at the nanosheet corner regions and at the uppermost nanosheet surface relative to portions of the gate dielectric that are not subjected to the same degree of wet-etch attack.
One of ordinary skill in the art would have understood that preferential etching or erosion of the gate dielectric at these corner regions would predictably reduce the thickness of the gate dielectric at those regions relative to portions of the gate dielectric outside the preferentially attacked corner regions.
Accordingly, Wang's teaching of preferential gate-dielectric attack at the corners of the nanosheets, including at the uppermost nanosheet, would have predictably resulted in the claimed condition wherein, after removal of the transition metal nitride layer, the thickness of the gate dielectric in corner regions of the uppermost channel is less than the thickness of the gate dielectric outside the corner regions.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply Wang's teaching concerning nanosheet gate-dielectric behavior during wet gate-metal removal to the wet removal of Zhang's transition-metal-nitride WFM layer as modified by Lee-942.
Zhang and Wang concern closely related nanosheet transistor gate structures having gate dielectric disposed around semiconductor nanosheets and gate material disposed around the gate dielectric. More particularly, both references concern removal/patterning of gate material in the confined geometry associated with stacked nanosheets.
Wang teaches that the geometry of stacked nanosheets causes portions of the gate dielectric at the corners/edges and uppermost nanosheet surface to become exposed to the wet etchant before removal of gate material has been completed elsewhere. Continued wet etching therefore preferentially attacks those already-exposed dielectric regions.
Thus, Wang would have informed one of ordinary skill in the art of a known and predictable consequence of performing the wet gate-metal-removal process of Zhang as modified by Lee-942 in a nanosheet structure: preferential erosion and consequent thinning of the gate dielectric at exposed corner regions of the uppermost nanosheet.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to perform Zhang's selective transition-metal-nitride removal using the sequential wet-etch process taught by Lee-942, with the predictable nanosheet-specific result taught by Wang that continued wet gate-metal removal preferentially attacks the gate dielectric at the corner regions of the uppermost nanosheet, thereby producing a gate-dielectric thickness at those corner regions that is less than the gate-dielectric thickness outside the corner regions.
Claim 29 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20210210388; “Zhang”; in the IDS on 1/30/25) in view of Lee et al. (US 20050000942; “Lee-942”), and Wang et al. (US 20200083051; “Wang”), and further in view of Lee et al. (US 20020060307; “Lee-307”).
Regarding claim 29, As discussed above with respect to claim 28, Zhang in view of Lee-942 and Wang teaches or suggests all of the method of claim 28.
But Zhang in view of Lee-942 and Wang does not expressly teach that the acidic etchant comprises:
HCl:H₂O₂:H₂O in a ratio of about 1:10:50.
However, Lee-307 teaches an aqueous acidic wet etchant comprising HCl, H₂O₂, and H₂O for semiconductor processing and, specifically, for etching TiN.
In particular, Lee-307 [0045] teaches an aqueous HCl/H₂O₂ solution having a ratio of:
0.1:1:6 (HCl:H₂O₂:H₂O),
which is equivalently expressed as:
1:10:60 (HCl:H₂O₂:H₂O).
Lee-307 further teaches that this composition at 65°C etches a TiN layer at a rate of about 90 Å/min to about 180 Å/min.
Thus, Lee-307 teaches the same three components recited in claim 29—HCl, H₂O₂, and H₂O—in substantially the same relative proportions and for the same pertinent purpose of wet etching TiN.
The HCl:H₂O₂ proportion disclosed by Lee-307 is identical to that recited in claim 29, namely 1:10. The only difference is 60 parts water in Lee-307 compared with about 50 parts water recited in claim 29.
Lee-307's expressly disclosed 1:10:60 composition therefore closely approximates the claimed ratio of about 1:10:50. In this regard, claim 29 expressly employs the term “about,” such that the claimed ratio is not restricted to mathematical exactness.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to employ the HCl/H₂O₂/H₂O acidic TiN etchant taught by Lee-307 when carrying out the acidic first etch of Zhang as modified by Lee-942 and Wang.
Lee-307 is particularly pertinent because it expressly teaches that its HCl:H₂O₂:H₂O = 1:10:60 composition actually etches TiN, the same transition metal nitride expressly disclosed by Zhang for WFM layer 34.
Thus, one of ordinary skill seeking an acidic wet-etch composition for removal of Zhang's TiN WFM layer would have had reason to employ Lee-307's known HCl/H₂O₂/H₂O TiN etchant because it was known to etch the same material for the same general semiconductor-processing purpose.
Moreover, even if the claimed term “about 1:10:50” were construed sufficiently narrowly that Lee-307's 1:10:60 ratio does not itself fall within the scope of the claim, the claimed ratio nevertheless would have been obvious.
Lee-307 teaches varying the relative proportions of the mineral acid, peroxide, and deionized water and teaches dilution of the concentrated mineral-acid and peroxide solutions with deionized water in desired proportions. Lee-307 also teaches selecting processing conditions empirically based upon desired etch rates.
Accordingly, Lee-307 establishes that the relative amount of water/dilution in an HCl/H₂O₂/H₂O TiN etchant was a known adjustable process parameter.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to adjust the amount of diluent water in Lee-307's known 1:10:60 HCl:H₂O₂:H₂O TiN etchant from 60 parts to about 50 parts, while maintaining the expressly taught 1:10 HCl:H₂O₂ proportion, through routine optimization to obtain desired TiN etching characteristics.
Such modification represents only a relatively small adjustment in dilution of a known TiN wet-etch composition, employing the same chemical components for the same pertinent purpose of etching TiN.
Accordingly, Lee-307's 1:10:60 composition closely approximates the claimed “about 1:10:50” composition. Alternatively, even if 1:10:60 is determined not to fall within the reasonable scope of “about 1:10:50,” selection of about 1:10:50 would have been an obvious optimization of Lee-307's known TiN wet-etch composition.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Changhyun Yi whose telephone number is (571)270-7799. The examiner can normally be reached Monday-Friday: 10A-3P.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Davienne Monbleau can be reached on 571-272-1945. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Changhyun Yi/Primary Examiner, Art Unit 2812
Prosecution Insights