Atomic Layer Etching of Tungsten for Enhanced Tungsten Deposition Fill

Abstract

The present invention relates to atomic layer etching of tungsten for enhancing tungsten deposition fill. The present invention provides a method for depositing tungsten into high aspect ratio features using a deposition-etch-deposition process that integrates multiple deposition techniques with pulses of surface modification and removal pulses that alternate during etching.

Description

[0001] This application is a divisional application of the invention patent application with the application number 201610643282.6, the application date of August 8, 2016, and the invention name of "Atomic Layer Etching of Tungsten for Enhancing Tungsten Deposition Filling". Technical Field

[0002] The present invention generally relates to the field of semiconductors, and more particularly to atomic layer etching of tungsten for enhancing tungsten deposition filling. Background Art

[0003] Semiconductor manufacturing processes typically involve depositing a metal (such as tungsten) into features (e.g., vias or trenches) to form contacts or interconnects. However, as devices are scaled down, the features become smaller and more difficult to fill, especially in advanced logic and memory applications. Summary of the Invention

[0004] The present invention provides a method for filling features on a substrate. One aspect relates to a method for filling features on a substrate by the following steps: (a) depositing a first amount of metal within the feature; and (b) directionally etching the metal at or near the opening of the feature relative to the interior region of the feature by: (i) modifying the surface of the deposited metal by exposing the metal to a halogen-containing gas; and (ii) exposing the modified surface to an activation gas to selectively etch the metal. The method may further include repeating (a) and (b).

[0005] In various embodiments, the metal comprises one of titanium, tantalum, nickel, cobalt, or molybdenum. In some embodiments, the metal comprises tungsten.

[0006] In some embodiments, the halogen-containing gas may be selected from the group consisting of chlorine, bromine, iodine, sulfur hexafluoride, silicon tetrafluoride, boron trichloride, and combinations thereof. In some embodiments, the activation gas is an inert gas, such as neon, krypton, argon, or combinations thereof.

[0007] The method may further include applying a bias during at least one of (i) and (ii). The bias power may be less than a threshold bias power. The bias power may be less than about 80 Vb.

[0008] In various embodiments, (b) includes a self-limiting reaction. In some embodiments, the substrate comprises features having openings of different sizes. The feature may have an aspect ratio of at least 3:1. In some embodiments, the width of the opening is less than 20 nm.

[0009] In some embodiments, (a) and (b) are performed without breaking vacuum. In some embodiments, (a) and (b) are performed in the same chamber. In some embodiments, (a) and (b) are performed in different chambers of the same tool.

[0010] The method may further include igniting a plasma during at least one of (i) and (ii). The plasma power may be between about 0 W and about 1000 W.

[0011] Another aspect may relate to a method that includes: (a) partially filling a feature with tungsten; (b) directionally etching tungsten at or near an opening of the feature by exposing the substrate to pulses of an alternating halogen-containing gas and pulses of an activation gas; and (c) filling the feature with tungsten.

[0012] In some embodiments, a bias is applied during (b). In some embodiments, the bias is applied at a threshold bias power during (b).

[0013] In various embodiments, (a) and (b) are performed without breaking vacuum. In some embodiments, (a) and (b) are performed in the same chamber. The method may further include repeating (a) and (b). Filling the feature may include repeating (a) and (b).

[0014] The tungsten may be deposited by CVD. In some embodiments, the tungsten is deposited by ALD. The tungsten may be deposited by exposing the substrate to pulses of an alternating tungsten precursor and pulses of a reducing agent. The tungsten may be deposited using a chlorine-containing tungsten precursor. In some embodiments, the tungsten is fluorine-free tungsten.

[0015] Another aspect relates to an apparatus for processing a semiconductor substrate, the apparatus including: a processing chamber that includes a showerhead and a substrate support, a plasma generator, and a controller having at least one processor and a memory, wherein the at least one processor and the memory are communicatively connected to each other, the at least one processor is at least operatively connected to flow control hardware, and the memory stores machine-readable instructions for the following operations: (i) introducing a tungsten precursor and a reducing agent into the chamber to deposit tungsten on the substrate, (ii) introducing a halogen-containing gas to modify the surface of the tungsten, and (iii) introducing an activation gas and igniting a plasma to etch at least a portion of the modified surface of the tungsten.

[0016] The substrate support may include a bias, and the memory may further store machine-readable instructions for setting a bias power of less than about 80 Vb during (iii). In some embodiments, the memory further stores machine-readable instructions for igniting a plasma during (ii).

[0017] In some embodiments, the memory further stores machine-readable instructions for repeating (ii) and (iii) in a loop. In some embodiments, the memory further stores machine-readable instructions for repeating (i) after performing (ii) and (iii).

[0018] Specifically, some aspects of the present invention can be described as follows:

[0019] 1. A method of filling features on a substrate, the method comprising:

[0020] (a) depositing a first amount of metal within the feature; and

[0021] (b) directionally etching the metal at or near an opening of the feature relative to an interior region of the feature by:

[0022] (i) modifying a surface of the deposited metal by exposing the metal to a halogen-containing gas; and

[0023] (ii) exposing the modified surface to an activation gas to selectively etch the metal.

[0024] 2. The method according to clause 1, wherein the metal comprises one of titanium, tantalum, nickel, cobalt, or molybdenum.

[0025] 3. The method according to clause 1, wherein the metal comprises tungsten.

[0026] 4. The method according to clause 1, further comprising applying a bias during at least one of (i) and (ii).

[0027] 5. The method according to clause 4, wherein the power of the bias is less than a threshold bias power.

[0028] 6. The method according to clause 1, wherein (b) comprises a self-limiting reaction.

[0029] 7. The method according to clause 1, wherein the substrate comprises features having openings of different sizes.

[0030] 8. The method according to clause 1, wherein (a) and (b) are performed without breaking vacuum.

[0031] 9. The method according to clause 1, wherein (a) and (b) are performed in the same chamber.

[0032] 10. The method according to clause 8, wherein (a) and (b) are performed in different chambers of the same tool.

[0033] 11. The method according to Clause 1, wherein the activation gas is selected from the group consisting of neon, krypton, and argon.

[0034] 12. The method according to any one of Clauses 1 - 11, further comprising igniting a plasma during at least one of (i) and (ii).

[0035] 13. The method according to any one of Clauses 1 - 11, wherein the feature has an aspect ratio of at least 3:1.

[0036] 14. The method according to any one of Clauses 1 - 11, wherein the width of the opening is less than 20 nm.

[0037] 15. The method according to any one of Clauses 1 - 11, further comprising: repeating (a) and (b).

[0038] 16. The method according to any one of Clauses 1 - 11, wherein the halogen - containing gas is selected from the group consisting of chlorine, bromine, iodine, sulfur hexafluoride, silicon tetrafluoride, boron trichloride, and combinations thereof.

[0039] 17. The method according to Clause 4, wherein the bias power is less than about 80 Vb.

[0040] 18. The method according to Clause 12, wherein the plasma power is between about 0 W and about 1000 W.

[0041] 19. A method, comprising:

[0042] (a) partially filling a feature with tungsten;

[0043] (b) directionally etching tungsten at or near an opening of the feature by exposing the substrate to pulses of an alternating halogen - containing gas and pulses of an activation gas; and

[0044] (c) filling the feature with tungsten.

[0045] 20. The method according to Clause 19, wherein a bias is applied during (b).

[0046] 21. The method according to Clause 19, wherein the bias is applied at a threshold bias power during (b).

[0047] 22. The method according to Clause 19, wherein (a) and (b) are performed without breaking the vacuum.

[0048] 23. The method according to Clause 19, wherein (a) and (b) are performed in the same chamber.

[0049] 24. The method according to any one of clauses 19 - 23 further comprises: repeating (a) and (b).

[0050] 25. The method according to any one of clauses 19 - 23, wherein filling the feature comprises repeating (a) and (b).

[0051] 26. The method according to any one of clauses 19 - 23, wherein the tungsten is deposited by CVD.

[0052] 27. The method according to any one of clauses 19 - 23, wherein the tungsten is deposited by ALD.

[0053] 28. The method according to any one of clauses 19 - 23, wherein the tungsten is deposited by exposing the substrate to pulses of an alternating tungsten - containing precursor and pulses of a reducing agent.

[0054] 29. The method according to any one of clauses 19 - 23, wherein the tungsten is deposited using a chlorine - containing tungsten precursor.

[0055] 30. The method according to any one of clauses 19 - 23, wherein the tungsten is fluorine - free tungsten.

[0056] 31. An apparatus for processing a semiconductor substrate, the apparatus comprising:

[0057] A processing chamber comprising a showerhead and a substrate support,

[0058] A plasma generator, and

[0059] A controller having at least one processor and a memory,

[0060] wherein the at least one processor and the memory are communicatively connected to each other,

[0061] the at least one processor is at least operatively connected to flow control hardware, and

[0062] the memory stores machine - readable instructions for the following operations:

[0063] (i) Introducing a tungsten - containing precursor and a reducing agent into the chamber to deposit tungsten on the substrate;

[0064] (ii) Introducing a halogen - containing gas to modify the surface of the tungsten; and

[0065] (iii) Introducing an activation gas and igniting a plasma to etch at least a portion of the modified surface of the tungsten.

[0066] 32. The apparatus according to clause 31, wherein the memory further stores machine - readable instructions for igniting a plasma during (ii).

[0067] 33. The apparatus according to clause 31, wherein the substrate support includes a bias, and the memory further stores machine-readable instructions for setting a bias power of less than about 80 Vb during (iii).

[0068] 34. The apparatus according to clauses 31-33, wherein the memory further stores machine-readable instructions for repeatedly cycling (ii) and (iii).

[0069] 35. The apparatus according to clauses 31-33, wherein the memory further stores machine-readable instructions for repeating (i) after executing (ii)

[0070] and (iii).

[0071] These and other features will be described below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 is a schematic diagram of an example of etching a film on a substrate by atomic layer etching.

[0073] Figure 2 is a schematic diagram of features undergoing operations of certain disclosed embodiments.

[0074] Figure 3 is a process flow diagram depicting operations performed according to certain disclosed embodiments.

[0075] Figure 4 is a graph of the normal incidence sputtering yield of tungsten calculated using argon ions.

[0076] Figure 5 is a timing schematic diagram of an example of operations performed according to certain disclosed embodiments.

[0077] Figure 6 is a schematic diagram of an exemplary processing chamber for performing certain disclosed embodiments.

[0078] Figure 7 is a schematic diagram of an exemplary processing apparatus for performing certain disclosed embodiments.

[0079] Figure 8 is a graph of the relationship between the etch rate of the collected tungsten and the chlorine bias power.

[0080] Figure 9A is an image of a feature having tungsten.

[0081] Figure 9B is an image of a feature having tungsten deposited according to certain disclosed embodiments. DETAILED DESCRIPTION

[0082] In the following description, numerous specific details are set forth to provide a thorough understanding of the presented embodiments. The disclosed embodiments may be practiced without some or all of these specific details. In other instances, well-known processing operations have not been described in detail to avoid unnecessarily obscuring the disclosed embodiments. Although the disclosed embodiments will be described in conjunction with specific embodiments, it should be understood that the intention is not to limit the disclosed embodiments.

[0083] Semiconductor manufacturing processes typically involve depositing metal into features such as vias or trenches to form contacts or interconnects. Tungsten is often deposited into such features using chemical vapor deposition (CVD), whereby a substrate containing the features to be filled is exposed to a tungsten precursor and a reducing agent to deposit tungsten into the features. However, as devices are scaled down, the features become smaller and more difficult to fill by CVD, especially in advanced logic and memory applications. For example, the features can have a high aspect ratio, such as at least about 3:1. Certain features can have small openings of less than about 20 nm. Certain features can also include re-entrant feature profiles, which are further described below with reference to Figure 2 For features in advanced technology nodes, the deposition rate at or near the opening of the feature can be faster than the deposition rate at the bottom of the feature, resulting in the opening closing before the entire feature is filled, thereby leaving voids or gaps in the feature. The presence of such gaps can be detrimental to the performance and reliability of semiconductor devices and ultimately to the performance and reliability of semiconductor products.

[0084] In addition, some substrates can include features of various sizes. As a result, the features are filled more quickly within smaller features, or the openings of the features are closed more quickly, compared to within larger features, and the larger features may not be completely filled. The small openings and high aspect ratios of the features can also result in non-conformal deposition of tungsten within the features. Additionally, in re-entrant feature profiles caused by conventional deposition techniques and in possible protruding ends of underlying barrier or adhesion layers due to non-conformal coverage over the features, the features can have a net re-entrant sidewall profile, which makes it challenging to completely fill the feature.

[0085] A method of depositing tungsten into a feature having a small opening includes exposing a partially filled feature to a reactive species (e.g., a fluorine-containing species) generated in a remote plasma generator and operating in a mass transfer limited process regime to remove tungsten previously deposited at the opening of the feature, and thereby opening the feature to enable further deposition of tungsten into the feature, thus facilitating complete void-free filling. However, while such a method would be effective in depositing void-free tungsten into small features, the number of deposition and etch processes used to fill larger features can vary. In addition, the fluorine-containing reactive species is very reactive and thus etches tungsten rapidly, so that the etch conditions are adjusted to prevent over-etching of tungsten. If all of the deposited tungsten is removed, it is then difficult to re-nucleate the exposed surface with tungsten to fill the feature. For example, prior to depositing any tungsten, the feature is typically lined with a barrier layer, such as a titanium nitride barrier layer, and the feature may be exposed to a precursor to deposit a tungsten nucleation layer by a method such as atomic layer deposition (ALD). However, if the fluorine-containing reactive species removes the entire nucleation layer due to its high reactivity and etch rate, the titanium nitride barrier layer is exposed, and tungsten nucleates on the surface a second time, thus reducing throughput. In some embodiments, the fluorine-containing reactive species can etch at least some or all of the barrier layer, which may result in an incomplete second tungsten deposition because some tungsten nucleation layer may be missing on the substrate.

[0086] The present invention provides a method of filling features with tungsten using an integrated deposition and atomic layer etch (ALE) process. ALE is a technique for removing thin layers of material using sequential self-limiting reactions. In general, any suitable technique can be used to perform ALE. Examples of atomic layer etch techniques are described in U.S. Patent No. 8,883,028, issued November 11, 2014, and U.S. Patent No. 8,808,561, issued August 19, 2014, which are hereby incorporated by reference herein for the purpose of describing exemplary atomic layer etch techniques. In various embodiments, ALE can be carried out using a plasma or can be carried out using heat. The concept of an "ALE cycle" is relevant to the discussion of various embodiments herein. Generally, an ALE cycle is the smallest set of operations for performing one etch process (e.g., etching a single layer). The result of one cycle is the etching of at least some of the film layers on the substrate surface. Generally, an ALE cycle includes a modification operation of forming a reaction layer, followed by a removal operation of removing or etching only this modified layer. The cycle can include certain auxiliary operations, such as sweeping away one of the reactants or by-products. Generally, a cycle includes an example of a unique series of operations. For example, an ALE cycle can include the following operations: (i) delivering a reactant gas, which can be in a plasma state, (ii) sweeping the reactant gas from the chamber, (iii) delivering a removal gas and an optional plasma, and (iv) sweeping the chamber. In some embodiments, the etch can be performed non-conformally on a substrate (e.g., a substrate having topography and / or features).

[0087] Figure 1 Two exemplary schematic diagrams of an ALE cycle are shown. Diagrams 171a - 171e show a general ALE cycle. In 171a, a substrate is provided. In 171b, the surface of the substrate is modified. In 171c, the chemicals used to modify the substrate are swept away. In 171d, the modified layer is etched. In 171e, the modified layer is removed. Similarly, diagrams 172a - 172e show an example of an ALE cycle for etching a tungsten film. In 172a, a tungsten substrate is provided, which includes a plurality of tungsten atoms. In 172b, the reactant gas chlorine is introduced to the substrate, thereby modifying the substrate surface. In some embodiments, the chlorine reactant can also be delivered as atomic chlorine in a plasma state that does not cause direct etching of the tungsten substrate. The schematic diagram of 172b shows as an example that some chlorine gas is adsorbed on the substrate surface. Although chlorine gas (Cl2) is described in Figure 1 , any chlorine-containing compound or other suitable reactant can be used. In 172c, the reactant gas chlorine is swept from the chamber. In 172d, the removal gas argon containing a directional plasma, such as Ar +The plasma species and arrows are shown to remove the modified substrate surface. Activated etching involves using inert ions (e.g., Ar + ) operating at energies below the sputtering threshold to excite adsorbed species (e.g., Cl species) to etch away one monolayer of the substrate at a time. During this operation, a bias is applied to the substrate to attract ions towards it. In 172e, the chamber is purged and by-products are removed.

[0088] The etch rate for the ALE process is lower than that of fluorine-based remote plasmas, but due to the self-limiting nature of the surface reaction, ALE etching is more uniform. Thus, the ALE method provides more control over the etching operation (especially in larger features), such that the amount of material removed in each cycle is limited and is not etched too quickly, preventing the material on the surface of the feature from being completely etched. The deposition processes described herein can be controlled by switching the pressure of the chamber and the temperature of the substrate, both of which affect the adsorption of the modifying chemicals during ALE. The process can also be controlled by adjusting the substrate bias during one or more operations performed in ALE and by adjusting the modifying chemical flow rate and chemical process. The deposition process can also depend on the chemical properties of the metal to be deposited into the feature.

[0089] The disclosed embodiments can involve depositing a metal, such as tungsten, into a feature by any suitable method, suitable methods including ALD, CVD, plasma-enhanced ALD (PEALD), plasma-enhanced CVD (PECVD), or physical vapor deposition (PVD); adsorbing a halogen-containing gas and optionally exposing to a plasma to modify the surface of the deposited metal; exposing to an activation gas to remove the modified surface; and further depositing metal to fill the feature. Figure 2 An exemplary schematic of a feature undergoing multiple operations is provided in accordance with the disclosed embodiments. In 201, substrate 210 is shown having a feature 212, feature 212 including a TiN barrier layer 214 conformally deposited in the feature and tungsten 216 conformally deposited on the TiN barrier layer 214 by ALD. In 203, after the feature 212 is exposed to a halogen-containing gas to modify the surface of the deposited tungsten 216, the feature 212 is exposed to an activation gas, such as a gas including argon ions, or neon, or krypton, which can etch the tungsten 216 at or near the opening 218a of the feature 212, for example, by using low-bias directional etching. In 205, the feature 212 has been opened, leaving a feature opening 218b. In 207, the feature 212 is then filled with tungsten by CVD to obtain a void-free tungsten-filled feature.

[0090] Figure 3A process flow diagram depicting operations of a method is provided in accordance with the disclosed embodiments. Although the following description focuses on tungsten feature fill, aspects of the present disclosure may also be implemented in filling features with other materials. For example, feature fill using one or more of the techniques described herein may be used to fill features with other materials, including other tungsten-containing materials (e.g., tungsten nitride (WN) and tungsten carbide (WC)), titanium-containing materials (such as titanium (Ti), titanium nitride (TiN), titanium silicide (TiSi), titanium carbide (TiC), and titanium aluminide (TiAl)), tantalum-containing materials (e.g., tantalum (Ta), and tantalum nitride (TaN)), molybdenum-containing materials, cobalt-containing materials, and nickel-containing materials (e.g., nickel (Ni) and nickel silicide (NiSi)). In various embodiments, the feature may be filled with another metal in place of tungsten or filled with a combination of another metal and tungsten. For example, cobalt or molybdenum may be used to fill the feature.

[0091] In Figure 3 operation 301, a substrate is provided to a chamber. The substrate may be a silicon wafer, e.g., a 200 mm wafer, a 300 mm wafer, or a 450 mm wafer, including wafers having one or more material layers, such as dielectric, conductive, or semiconductive materials deposited on the wafer. The patterned substrate may have "features", such as vias or contact holes, which may be characterized as one or more narrow and / or re-entrant openings, feature constrictions, and high aspect ratios. The features may be formed in one or more of the above layers. An example of a feature is a hole or via in a semiconductor substrate or a layer on the substrate. Another example is a trench in the substrate or layer. In multiple embodiments, the feature may have an underlying layer, such as a barrier layer or an adhesion layer. Non-limiting examples of underlying layers include dielectric layers and conductive layers, such as silicon oxide, silicon nitride, silicon carbide, metal oxides, metal nitrides, metal carbides, and metal layers.

[0092] Examples of applications include logic and memory contact fill, DRAM buried word line fill, vertical integrated memory gate / word line fill, and 3-D integration with through-silicon vias (TSVs). These methods described herein may be used to fill vertical features, such as in tungsten vias, as well as horizontal features, such as vertical NAND (VNAND) word lines.

[0093] In various embodiments, the type of substrate fabricated by performing the disclosed embodiments can depend on the aspect ratio of features on the substrate prior to performing the disclosed embodiments. In some embodiments, the features on the substrate provided in operation 301 can have an aspect ratio of at least about 2:1, at least about 3:1, at least about 4:1, at least about 6:1, at least about 10:1, or higher. The feature can also have dimensions approaching the opening, e.g., an opening diameter or line width between about 5 nanometers and 500 nanometers, e.g., an opening diameter or line width between about 25 nanometers and about 300 nanometers. The disclosed methods can be performed on a substrate having features with an opening less than about 20 nm. A "small" feature can be defined as a feature having an opening diameter or line width that is smaller than the opening diameter or line width of a "large" feature in the relative term. A large feature can have an opening diameter or critical dimension that is at least 1.5 times, or at least 2 times, or at least 5 times, or at least 10 times, or more than 10 times larger than the critical dimension of the small feature. Examples of "small" features include features having an opening diameter between about 1 nm and about 2 nm. Examples of "large" features include features having an opening diameter on the order of several hundred nanometers to about 1 micron.

[0094] Through-holes, trenches, or other recessed features can be referred to as unfilled features or features. According to various embodiments, the feature profile can taper and / or include a protruding end at the feature opening. A concave profile is a profile that tapers from the bottom, closed end, or interior of the feature towards the feature opening. The concave profile can be created by asymmetric etch kinetics during patterning and / or due to protrusions resulting from non-conformal film step coverage in a previous film deposition (e.g., deposition of a diffusion barrier layer). In different embodiments, the feature can have a width of the opening at the top of the feature that is smaller than the width of the middle and / or bottom of the feature.

[0095] In operation 303, tungsten is deposited on the substrate by, for example, exposing the substrate to a tungsten precursor and a reducing agent to partially fill the features on the substrate. Exemplary tungsten precursors include tungsten-containing halide precursors, which can include tungsten fluorides such as WF6; and tungsten chlorides such as WCl6, W(CO)6, and WCl5. In some embodiments, tungsten-containing organometallic precursors can be used. Exemplary reducing agents include hydrogen, borane (such as B2H6), silane (such as SiH4), and germane (such as GeH4).

[0096] In some embodiments, tungsten is deposited conformally. In some implementations, operation 303 involves the deposition of a tungsten nucleation layer, followed by bulk deposition.

[0097] Any suitable method for depositing tungsten can be used, such as ALD, CVD, PECVD, PEALD, or PVD. For the examples provided herein, tungsten can be conformally deposited into the features by ALD. For example, in some embodiments, a tungsten nucleation layer is deposited by sequentially applying pulses of a tungsten precursor and one or more reducing agents to form a tungsten nucleation layer by an ALD or pulsed nucleation layer (PNL) process. In some implementations, if, for example, the feature includes a lower layer that supports tungsten deposition, operation 303 may only involve bulk deposition without nucleation layer deposition. The bulk deposition can be deposited by chemical vapor deposition and is further described below.

[0098] In features that include constrictions or are otherwise prone to pinch-off, operation 303 can be performed at least until the feature is pinched off. Features of different sizes can be pinched off at different times. In conformal deposition, the deposition starts from each surface and grows gradually perpendicular to the surface. The tungsten growth in the feature starts from each sidewall and continues until the growth causes the feature to pinch off. In some implementations, the amount of tungsten deposited by operation 303 can be determined based on the narrowest feature dimension.

[0099] In various embodiments, operation 303 can be performed such that the opening of the feature is closed. In some embodiments, a seam can be formed at or near the opening of the feature. For the purposes of this specification, "near the opening" is defined as a general position or region within the feature that corresponds to between approximately 0 - 10% of the feature depth measured from the field region (i.e., along the sidewall of the feature). In certain embodiments, the region near the opening corresponds to the region at the opening. Additionally, "within the feature" or "the interior of the feature" is defined as a general position or region within the feature that corresponds to between approximately 20% - 60% of the feature depth measured from the field region at the top of the feature. Generally, when values for certain parameters (e.g., thickness) are specified as "near the opening" or "within the feature", these values represent the average of measured values or multiple measured values taken within these positions / regions.

[0100] In operation 305, the substrate is etched directionally or preferentially by atomic layer etching. The “directional” or “preferential” used here can be defined as etching more material at or near the top of the feature than in the rest of the feature (such as within or inside the feature). Atomic layer etching involves surface modification and activation operations. In some embodiments, a carrier gas (which may include N2, Ar, Ne, He, and combinations thereof) flows continuously during operation 305. In some embodiments, the carrier gas is used only during the removal process in operation 305. In some operations, the carrier gas can be used as a purge gas, as described below. In some embodiments, another reactive gas (such as oxygen) is used during operation 305 to remove the modified layer. In some embodiments, the carrier gas does not flow during the removal process.

[0101] In operation 315, the substrate is exposed to a modifying chemical to modify the surface of the substrate. The modifying chemical can be a gas or a plasma or a reactive substance. The modifying operation forms a thin reactive surface layer of a certain thickness, which is more easily removed than the unmodified material. The modifying operation can be performed to prevent spontaneous etching of the substrate.

[0102] In the modifying operation, a halogen-containing chemical can be used to modify the substrate. For example, the substrate can be chlorinated by introducing chlorine gas into the chamber. Chlorine gas is used as an exemplary modifying chemical in the disclosed embodiments, but it should be understood that in some embodiments, different modifying chemicals are introduced into the chamber. Examples include bromine, iodine, sulfur hexafluoride, silicon tetrafluoride, boron trichloride (BCl3). Further examples of ALE etching of metals are described in U.S. Patent Application No. 62 / 207,250, filed on August 19, 2015, entitled “ATOMIC LAYER ETCHING OF TUNGSTEN AND OTHER METALS” (Attorney Docket No. LAMRP209P / 3706-1US), which is incorporated herein by reference in its entirety.

[0103] In various embodiments, fluorine-containing chemicals are not used to prevent chemical etching that would not be monolayer etching. For example, nitrogen trifluoride (NF3) can be in a highly reactive plasma state and can etch the substrate spontaneously rather than etch the substrate conformally by layer. However, in some embodiments, highly reactive halogen-containing chemicals (such as ClF3) can be used to etch other materials, such as materials that are not easily etched spontaneously.

[0104] The modifying chemical can be selected according to the type and chemical properties of the substrate to be etched. In some embodiments, chlorine gas can react with the substrate or can be adsorbed on the substrate surface. In multiple embodiments, chlorine gas is introduced into the chamber in gaseous form and can optionally be accompanied by a carrier gas, which can be any of those described above.

[0105] In some embodiments, a chlorine-based plasma can be generated during this operation. The species generated from the chlorine-based plasma can be generated in-situ by forming a plasma in a processing chamber containing a substrate or they can be generated remotely in a processing chamber not containing a substrate (e.g., a remote plasma generator) and can be supplied to the processing chamber containing the substrate. In various embodiments, the plasma can be an inductively coupled plasma or a capacitively coupled plasma or a microwave plasma. The power for the inductively coupled plasma can be set between about 50 W and about 2000 W, such as about 900 W. The power can be set at a low enough level so as not to cause direct plasma etching of the substrate.

[0106] In some embodiments, no plasma is used and chlorine can be introduced into the chamber by thermal means. The energy for dissociating Cl2 into Cl is 2.51 eV. In some embodiments, this energy can be applied during this operation using a thermal or other radiant energy source. In some embodiments, chlorine can be heated to a high enough temperature to decompose chlorine gas into chlorine atoms that can adsorb onto the surface of the substrate.

[0107] In various embodiments, a bias is applied during operation 315. A low bias power can be used to prevent spontaneous etching of the substrate surface by the modifying chemical while allowing the modifying chemical to adsorb onto the surface of the deposited metal and enter seams that can form at or near the openings of the features. For example, a bias between about 0 V and about 200 V can be applied. It should be understood that the terms "bias power" and "bias voltage" are used interchangeably herein to describe the voltage at which the pedestal is set when a bias is applied to the pedestal. The threshold bias power or threshold bias voltage refers to the maximum voltage of the bias applied to the pedestal before the material on the surface of the substrate on the pedestal is sputtered. Thus, the threshold bias power depends in part on the material to be etched, the gas used to generate the plasma, the plasma power used to ignite the plasma, and the plasma frequency. The bias power or bias voltage as described herein is measured in volts, which is represented by the unit "V" or "Vb", where b refers to bias. An electron volt (eV) is the amount of energy acquired by an electron when accelerated through a potential difference of 1 volt. The electron volt described relative to Figure 4 refers to the amount of energy acquired by an electron when accelerated through a potential difference of 1 volt. For Figure 4 , the calculations for determining the sputtering rate (e.g., for each argon ion accelerated towards tungsten, how many tungsten atoms are sputtered off the surface) depend on the energy of the argon ions in electron volts.

[0108] This bias can be used to establish a gradient of the modifying chemical throughout the feature depth. By appropriately controlling the bias and other parameters, such as pressure, the degree of modification (and ALE) throughout the feature depth can be controlled. In one example, more chlorine is adsorbed at or near the top of the feature, or at or near the opening of the feature, compared to the bottom and sidewalls of the feature. The bias is applied in such a way as not to cause physical sputtering of the substrate. In some embodiments, the bias may not be used. In some embodiments, if the opening of the feature is large enough, the bias may not be used. In operation 315, an exemplary pressure range can be between about 30 millitorr and about 80 millitorr.

[0109] In some embodiments, the purge can be performed after the modification operation. In the purge operation, non-surface-bound reactive chlorine species can be removed from the processing chamber. This can be accomplished by purging and / or evacuating the processing chamber to remove unadsorbed modifying chemicals without removing the adsorbed layer. Species generated in a chlorine-based plasma can be removed by stopping the plasma and allowing the remaining species to decay, optionally in combination with purging and / or evacuating the chamber. The purge can be done using any inert gas (e.g., N2, Ar, Ne, He, and combinations thereof).

[0110] In operation 335, an activated removal gas, such as an activation gas, an ion bombardment gas, or a chemically reactive gas, is used to remove the modified layer from the substrate. The activated removal gas can be an inert gas. For example, argon can be used. In some embodiments, neon or krypton can be used. During the removal operation, the substrate can be exposed to an energy source (e.g., an activation or ion bombardment gas or a chemically reactive species that induces removal), such as argon or helium, to etch the substrate by directed ion bombardment. In some embodiments, the removal operation can be performed by low-energy ion bombardment. In some embodiments, the removal can be isotropic.

[0111] The amount of the removal gas can be controlled so as to etch only a targeted amount of material. In various embodiments, the pressure of the chamber can be changed between the modification and removal operations. The pressure of the removal gas can depend on the size of the chamber, the flow rate of the removal gas, the temperature of the reactor, the type of the substrate, the flow rate of any carrier gas, and the amount of tungsten to be etched. An exemplary pressure range during operation 335 can be between about 1 millitorr and about 15 millitorr.

[0112] During removal, a bias can optionally be applied to facilitate directed ion bombardment. The bias power is selected to prevent sputtering but to allow the removal gas to enter the feature and etch the tungsten at or near the opening of the feature, thus opening it. The bias power can be selected based on the threshold sputtering rate of the activated removal gas for the metal deposited on the substrate. Sputtering as used herein can refer to physically removing at least some of the surface of the substrate. Ion bombardment can refer to the physical bombardment of the surface of the substrate by a species.

[0113] Figure 4 Shows an example sputtering rate calculated based on "Energy Dependence of the Yields of Ion - Induced Sputtering of Monatomic Solids", N. Matsunami, Y. Yamamura, Y. Itikawa, N. Itoh, Y. Kazumata, S. Miyagawa, K. Morita, R. Shimizu, and H. Tawara, IPPJ - AM - 32 (Institute of Plasma Physics, Nagoya University, Japan, 1983).

[0114] This figure shows a curve of the calculated normal - incidence sputtering rate of tungsten sputtered by argon atoms versus the argon - ion energy (or threshold - bias power). The calculation uses a sputtering threshold of 32 eV. Slightly above the threshold, i.e., at an argon - ion energy of 40 eV, the sputtering rate appears to be about 0.001 atoms / ion. However, at an argon - ion energy of 80 eV, it has increased by a factor of 30. This exemplary curve indicates the maximum argon - ion energy sufficient to etch the metal while preventing sputtering of the substrate by argon. Although Figure 4 providing a qualitative representation of the sputtering - threshold curve, for a particular system and maximum allowable sputtering rate, the sputtering threshold can be determined experimentally. For one system, sputtering of tungsten by argon ions was observed at 80 Vb. It should be understood that the units "Vb" or "V b ” or “V bias ” are volts, and "b" or "bias" is used to characterize that the power is bias power. Thus, the bias power during tungsten removal using argon ions can be set to less than about 80 Vb, or less than about 50 Vb, or between about 50 Vb and 80 Vb. In some embodiments, if some amount of sputtering is tolerable, operation 335 can be performed at a bias power above the threshold. Additionally, there can be a removal - threshold voltage below which removal does not occur, depending on the particular process. It should be noted that the sputtering threshold varies depending on the metal, metal compound, or other material to be etched.

[0115] In some embodiments, the chamber can be purged after the removal operation. The purging process can be any of those purging processes used for purging after operation 315.

[0116] Return to Figure 3, Operations 315 and 335 can be optionally repeated as needed to fill the feature. In operation 307, it is determined whether the feature has been sufficiently filled. If not, operations 303 and 305 can be repeated. In some embodiments, operation 303 is repeated, and the feature can be sufficiently filled such that operation 305 may not be performed again. In some embodiments, operations 303 and 305 are performed until the feature is sufficiently filled. In some embodiments, the feature can be sufficiently filled after operation 303 is performed in one of the repeated operations, such that operation 305 is not performed after the feature is filled. In some embodiments, operations 303 and 305 are performed in the same chamber. In some embodiments, operations 303 and 305 are performed in the same tool. In some embodiments, operations 303 and 305 are performed without breaking the vacuum. In some embodiments, the cycle of repeated operation 303 may involve different deposition methods and precursors compared to the previous cycle of operation 303. For example, in one process, tungsten can be deposited into the feature by ALD, ALE can be performed to etch the deposited tungsten to open the feature, and then the deposition of tungsten can be repeated by CVD using a tungsten-containing precursor and a reducing agent to completely fill the feature. In another embodiment, tungsten is deposited by alternating pulses of WF6 and pulses of BH4, tungsten at or near the opening of the feature can be etched by alternating pulses of Cl2 and pulses of Ar in the presence of a plasma and applying a bias, and tungsten can be deposited by simultaneous exposure to WCl5 and H2.

[0117] Figure 5 An example diagram of a timing scheme that can be performed according to the disclosed embodiments is provided. Process 500 includes a deposition cycle 520A, an etch cycle 505A, and repeated deposition cycles 520B and etch cycles 505B. Deposition cycle 520A includes a CVD stage 503A of W, which can correspond to Figure 3 operation 503. Although CVD deposition is provided in Figure 5 , in some embodiments, this operation may involve, for example, cyclic deposition of a metal by ALD. In the CVD stage 503A of W, a carrier gas can flow while the modified chemical stream is shut off and the purge gas is closed. The CVD precursor can flow continuously to deposit tungsten and the bias is off. Etch cycle 505A can correspond to Figure 3 operations 315 and 335. Etch cycle 505A includes a surface modification 515A, which can correspond to Figure 3 operation 315. During surface modification 515A, the modified chemical flows with the carrier gas while the purge gas and CVD precursor streams are shut off. The bias can be turned on, as Figure 5As shown. After surface modification 515A, there can be a purge stage 525A, which is an optional operation as described above. During the purge stage 525A, the carrier gas flows continuously to remove any modified chemicals that have not been adsorbed onto the substrate. Accordingly, the modified chemical, removal gas, and CVD precursor flows are shut off, and the bias is also shut off. In the removal stage 535A, the carrier gas flows continuously while the removal gas is flowing, and the modified chemical and CVD precursor flows are shut off. The bias can also be turned on during the removal stage 535A. The removal stage 535A can correspond to Figure 3 operation 335. In various embodiments, plasma is ignited during this stage. The purge stage 545A can involve flowing the carrier gas while the modified chemical, removal gas, and CVD precursor flows are shut off, and the bias is also shut off.

[0118] According to Figure 3 operation 307, the operation can be repeated as Figure 5 shown. The deposition cycle 520B involves a CVD stage 503B for W, which in this example includes the same flows as in the CVD stage 503A for W. Here, the carrier gas and CVD precursor flow to deposit tungsten while the removal gas and modified chemical flows are shut off, and the bias is shut off. In some embodiments, this can further partially fill the feature. Although the same precursor as used in the CVD stage 503A for W can be used in the CVD stage 503B for W, in some embodiments, as described above, Figure 3 the repeated operation of 303 can involve different deposition techniques or precursors. The etch cycle 505B can correspond to Figure 3 operation 305 in the repeated cycle. The etch cycle 505B involves a surface modification 515B, whereby the carrier gas and modified chemical flow while the removal gas and CVD precursor flows are shut off, and the bias is turned on. The purge stage 525B includes flowing the carrier gas while all other flows are shut off and the bias is shut off. The removal stage 535B involves flowing the carrier gas and the removal gas while the modified chemical and CVD precursor flows are shut off. In various embodiments, plasma is ignited during this stage. The bias is turned on to direct the etching of the substrate. The purge stage 545B involves flowing the carrier gas without flowing the modified chemical, removal gas, and CVD precursor while the bias is shut off.

[0119] The embodiments described herein can be integrated with other processes. For example, ALE etching can be integrated on a Multi-Station-Sequential-Deposition (MSSD) chamber structure, where one of the deposition stations can be replaced by an ALE station so that deposition / etching / deposition can be integrated with similar chemicals, resulting in better filling and faster throughput. The disclosed embodiments can be carried out in some embodiments without breaking the vacuum. For example, in some embodiments, the disclosed embodiments can be carried out in the same chamber or in the same tool. Further embodiments of the apparatus suitable for performing the disclosed embodiments are described further below.

[0120] Apparatus

[0121] An inductively coupled plasma (ICP) reactor suitable for atomic layer etching (ALE) operations and atomic layer deposition (ALD) operations in certain embodiments is now described. Such an ICP reactor has also been described in U.S. Patent Application Publication No. 2014 / 0170853, filed on December 10, 2013, and entitled "IMAGE REVERSAL WITH AHM GAP FILL FOR MULTIPLE PATTERNING", which is hereby incorporated by reference in its entirety and for all purposes. Although an ICP reactor is described herein, it should be understood that in some embodiments, a capacitively coupled plasma reactor can also be used.

[0122] Figure 6 A cross-sectional view schematically showing an inductively coupled plasma integrated etching and deposition apparatus 600 suitable for implementing certain embodiments herein is an example of Reactor, manufactured by Lam Research Corp. of Fremont, California. The inductively coupled plasma device 600 includes a total processing chamber 624 structurally defined by a chamber wall 601 and a window 611. The chamber wall 601 can be made of stainless steel or aluminum. The window 611 can be made of quartz or other dielectric materials. An optional internal plasma grid 650 divides the total processing chamber 624 into an upper sub-chamber 602 and a lower sub-chamber 603. In most embodiments, the plasma grid 650 can be removed to utilize the chamber space formed by the sub-chambers 602 and 603. A chuck 617 is positioned in the lower sub-chamber 603 near the bottom inner surface. The chuck 617 is configured to receive and hold a semiconductor substrate or wafer 619 on which an etching and deposition process is performed. The chuck 617 can be an electrostatic chuck for supporting the wafer 619 when the wafer 619 is present. In some embodiments, an edge ring (not shown) surrounds the chuck 617 and has an upper surface substantially in the same plane as the top surface of the wafer 619 (when the wafer is above the chuck 617). The chuck 617 also includes electrostatic electrodes for clamping and releasing the wafer 619. A filter and a DC clamp power source (not shown) can be provided for this purpose. Other control systems can also be provided to lift the wafer 619 away from the chuck 617. The chuck 617 can be charged by an RF power source 623. The RF power source 623 is connected to a matching circuit 621 through a connection member 627. The matching circuit 621 is connected to the chuck 617 through a connection member 625. In this way, the RF power source 623 is connected to the chuck 617.

[0123] Elements for plasma generation include a coil 633 located above the window 611. In some embodiments, the coil is not used in the disclosed embodiments. The coil 633 is made of a conductive material and includes at least one full turn. In Figure 6An example of the coil 633 shown in [the figure] includes three turns. The cross-section of the coil 633 is shown by symbols, where a coil with an "X" symbol indicates that the coil 633 extends rotationally into the page, and conversely, a coil with a "●" symbol indicates that the coil extends rotationally out of the page. The element for plasma generation also includes an RF power source 641 configured to provide RF power to the coil 633. Generally, the RF power source 641 is connected to the matching circuit 639 through a connector 645. The matching circuit 639 is connected to the coil 633 through a connector 643. In this way, the RF power source 641 is connected to the coil 633. An optional Faraday shield 649 is positioned between the coil 633 and the window 611. The Faraday shield 649 is held in a spaced-apart relationship relative to the coil 633. The Faraday shield 649 is disposed directly above the window 611. The coil 633, the Faraday shield 649, and the window 611 are each configured to be substantially parallel to each other. The Faraday shield 649 can prevent metal or other substances from depositing on the window 611 of the processing chamber 624.

[0124] Process gases (such as metal precursors, e.g., tungsten-containing precursors, reducing agents, carrier gases, halogen-containing gases, chlorine gas, argon gas, etc.) can flow into the processing chamber through one or more main gas inlets 660 located in the upper sub-chamber 602 and / or through one or more side gas inlets 670. Similarly, although not explicitly shown, similar gas inlets can be used to supply process gases to the capacitively coupled plasma processing chamber. A vacuum pump 640, e.g., a single-stage or two-stage dry mechanical pump and / or a turbomolecular pump, can be used to evacuate the process gases from the processing chamber 624 and maintain the pressure inside the processing chamber 624. For example, the vacuum pump 640 can be used to evacuate the lower sub-chamber 603 during the ALE purge operation. A valve-controlled conduit can be used to fluidly connect the vacuum pump to the processing chamber 624 so as to selectively control the application of the vacuum environment provided by the vacuum pump. During the operation of the plasma processing, this can be done using a closed-loop controlled flow restriction device such as a throttle valve (not shown) or a pendulum valve (not shown). Similarly, a vacuum pump and a valve that are controllably fluidly connected to the capacitively coupled plasma processing chamber can also be used.

[0125] During operation of the apparatus 600, one or more process gases may be supplied through gas inlets 660 and / or 670. In certain embodiments, the process gas may be supplied only through the main gas inlet 660, or only through the side gas inlet 670. In some cases, the gas inlets shown in the figures may be replaced by more complex gas inlets, for example, by one or more showerheads. The Faraday shield 649 and / or optional grid 650 may include internal channels and holes that enable the process gas to be delivered to the interior of the processing chamber 624. One or both of the Faraday shield 649 and the optional grid 650 may serve as a showerhead for delivering the process gas. In some embodiments, a liquid evaporation and delivery system may be located upstream of the processing chamber 624 such that once the liquid reactant or precursor is evaporated, the evaporated reactant or precursor is introduced into the processing chamber 624 through the gas inlets 660 and / or 670.

[0126] RF power is supplied from an RF power source 641 to the coil 633 to cause an RF current to flow through the coil 633. The RF current flowing through the coil 633 generates an electromagnetic field around the coil 633. The electromagnetic field generates an induced current within the upper sub-chamber 602. The generated ions and radicals physically and chemically interact with the wafer 619 to etch features and deposited layers on the wafer 619.

[0127] Volatile etch and / or deposition by-products may be removed from the lower sub-chamber 603 through port 622. The chuck 617 disclosed herein may operate within an elevated temperature range between about 10°C and about 250°C. The temperature will depend on the process operation and specific recipe.

[0128] The apparatus 600 may be connected to a facility (not shown) when installed in a clean room or manufacturing facility. The facility includes piping that provides process gases, vacuum, temperature control, and environmental particle control. These facilities are connected to the apparatus 600 when installed in the target manufacturing facility. Additionally, the apparatus 600 may be coupled to a transfer chamber, thereby allowing for the transfer of semiconductor wafers into and out of the apparatus 600 by a robot using typical automation.

[0129] In some embodiments, a system controller 630 (which may include one or more physical or logical controllers) controls some or all of the operations of the processing chamber 624. The system controller 630 may include one or more memory devices and one or more processors. For example, the memory may include instructions to alternate between flows of a modifying chemical (such as a chlorine-containing modifying chemical) and a removal gas (such as argon), or instructions to ignite a plasma or apply a bias. For example, during some operations, the memory may include instructions to set the bias power between approximately 0 V and approximately 200 V. In some embodiments, the apparatus 600 includes a switching system for controlling flow rates and durations when performing the disclosed embodiments. In some embodiments, the apparatus 600 may have a switching time of up to approximately 500 ms or up to approximately 750 ms. The switching time may depend on the flowing chemicals, recipe selection, reactor architecture, and other factors.

[0130] In some embodiments, the disclosed embodiments may be integrated on an MSSD (Multi-Station - Sequential Deposition) chamber structure, where one of the deposition stations may be replaced by an ALE station such that a deposition / etch / deposition process can be integrated with similar chemicals, resulting in better fill and faster throughput capabilities.

[0131] In some implementations, the system controller 630 is part of a system that may be part of the above examples. Such systems may include semiconductor processing equipment that includes one or more processing tools, one or more processing chambers, one or more platforms for processing, and / or specific processing components (wafer pedestal, gas flow system, etc.). These systems may be integrated with electronics for controlling their operations before, during, and after processing semiconductor wafers or substrates. The electronics may be integrated into the system controller 630, which may control various elements or sub-components of one or more systems. Depending on the processing parameters and / or the type of system, the system controller 630 may be programmed to control any of the processes disclosed herein, including controlling process gas delivery, temperature settings (e.g., heating and / or cooling), pressure settings, vacuum settings, power settings, radio frequency (RF) generator settings, RF matching circuit settings, frequency settings, flow rate settings, fluid delivery settings, position and operation settings, wafer transfer in and out of the tool and other transfer tools, and / or load locks connected or interfaced to the specific system.

[0132] Broadly speaking, system controller 630 can be defined as an electronic device with various integrated circuits, logic, memory, and / or software that receives instructions, issues instructions, controls operations, enables cleaning operations, enables endpoint measurements, and so on. The integrated circuits can include chips in the form of firmware that stores program instructions, digital signal processors (DSPs), chips defined as application-specific integrated circuits (ASICs), and / or one or more microprocessors or microcontrollers that execute program instructions (e.g., software). The program instructions can be instructions communicated to the controller in the form of various individual settings (or program files), which define the operating parameters for performing a specific process on or for a semiconductor wafer or system. In some embodiments, the operating parameters can be part of a recipe defined by a process engineer for completing one or more processing steps during the preparation or removal of one or more layers, materials, metals, oxides, silicon, silicon dioxide, surfaces, circuits, and / or dies of a wafer.

[0133] In some implementations, system controller 630 can be part of a computer that is integrated with, coupled to, or network-connected to the system or a combination thereof. For example, the controller can be in the "cloud" or be all or part of a fab host system, which can allow remote access to wafer processing. The computer can enable remote access to the system to monitor the current progress of manufacturing operations, examine the history of past manufacturing operations, examine trends or performance criteria of multiple manufacturing operations, change the parameters of the current process, set processing steps to follow the current process, or initiate a new process. In some instances, a remote computer (e.g., a server) can provide a process recipe to the system via a network, which can include a local network or the Internet. The remote computer can include a user interface capable of inputting or programming parameters and / or settings, which are then communicated from the remote computer to the system. In some instances, system controller 630 receives instructions in the form of data that specify the parameters for each processing step to be performed during one or more operations. It should be understood that the parameters can be for the type of process to be performed as well as the type of tool that the controller is configured to connect to or control. Thus, as described above, system controller 630 can be distributed, for example, by including one or more discrete controllers that are connected together via a network and work towards a common goal (e.g., the processes and controls described herein). An example of a distributed controller for these purposes can be one or more integrated circuits on a chamber that communicate with one or more remote integrated circuits (e.g., at the platform level or as part of a remote computer) combined to control in-chamber processes.

[0134] Under non - restrictive conditions, an exemplary system can include a plasma etch chamber or module, a deposition chamber or module, a spin - clean chamber or module, a metal plating chamber or module, a cleaning chamber or module, a bevel - edge etch chamber or module, a physical vapor deposition (PVD) chamber or module, a chemical vapor deposition (CVD) chamber or module, an ALD chamber or module, an ALE chamber or module, an ion implantation chamber or module, an orbit chamber or module, and any other semiconductor processing system that can be associated with or used in the fabrication and / or manufacture of semiconductor wafers.

[0135] As described above, depending on one or more process steps to be performed by the tool, the controller can communicate with one or more other tool circuits or modules, other tool components, cluster tools, other tool interfaces, adjacent tools, adjoining tools, tools located throughout the factory, a host, another controller, or tools used in material handling for transporting a container of wafers between tool locations and / or load ports in a semiconductor manufacturing facility.

[0136] Figure 7 A semiconductor process cluster architecture is described, where each module interfaces with a vacuum transfer module 738 (VTM). The multiple modular devices that "transfer" wafers between multiple storage and processing modules can be referred to as a "cluster tool architecture" system. An airtight chamber 730 (also referred to as a load lock or transfer module) is connected to the VTM 738, which in turn is connected to four processing modules 720a - 720d. The four processing modules 720a - 720d can be individually optimized to perform various manufacturing processes. For example, the processing modules 720a - 720d can be implemented to perform substrate etching, deposition, ion implantation, wafer cleaning, sputtering, and / or other semiconductor processes. In some embodiments, ALD and ALE are performed in the same module. In some embodiments, ALD and ALE are performed in different modules within the same tool. One or more of the substrate etching processing modules (any of 720a - 720d) can be implemented as disclosed herein, i.e., for depositing conformal films, directionally etching films by ALE, etching patterns, and other suitable functions as described in the disclosed embodiments. The airtight chamber 730 and the processing modules 720a - 720d can be referred to as "stations". Each station has a facet 736 that connects the station to the VTM 738. Inside each facet, sensors 1 - 18 are used to detect the passage of the substrate 726 as it moves between stations.

[0137] The robot arm 722 transfers the wafer 726 between stations. In one embodiment, the robot arm 722 has one arm, while in another embodiment, the robot arm has two arms, and each arm has an end effector 724 for picking up the wafer (e.g., wafer 726) for transportation. In the atmospheric transfer module (ATM) 740, the front-end robot arm 732 is used to transfer the wafer 726 from the wafer cassette or front-opening unified pod (FOUP) 734 in the load port module (LPM) 742 to the airtight chamber 730. The module center 728 within the processing modules 720a - 720d is a location for placing the wafer 726. The aligner 744 in the ATM 740 is used to align the wafer.

[0138] In an exemplary processing method, the wafer is placed in one of the multiple FOUPs 734 in the LPM 742. The front-end robot arm 732 transfers the wafer from the FOUP 734 to the aligner 744, which allows the wafer 726 to be properly centered before being etched or processed. After alignment, the wafer 726 is moved by the front-end robot arm 732 into the airtight chamber 730. Since the airtight chamber 730 has the ability to match the environment between the ATM 740 and the VTM 738, the wafer 726 can move between the two pressure environments without being damaged. From the airtight chamber 730, the wafer 726 is moved through the VTM 738 by the robot arm 722 and into one of the processing modules 720a - 720d. To achieve this wafer movement, the robot arm 722 uses the end effector 724 on each of its arms. Once the wafer 726 has been processed, it is moved from the processing module 720a - 720d to the airtight chamber 730 by the robot arm 722. The wafer 726 can be moved from here by the front-end robot arm 722 into one of the multiple FOUPs 734 or to the aligner 744.

[0139] It should be noted that the computer controlling the wafer movement can be local to the cluster architecture, or it can be located outside the cluster architecture in a manufacturing factory, or at a remote location and connected to the cluster architecture through a network. As referred to above Figure 6 The controller described can be implemented with the tools in Figure 7 ...

[0140] Experiment

[0141] Experiment 1

[0142] The etching rate of tungsten is plotted against the chlorine bias power for etching without argon sputtering during chlorine adsorption, and for atomic layer etching (ALE) processing with chlorine adsorption and argon sputtering. The results are plotted in Figure 8The dashed line depicts the etch rate of tungsten as a function of chlorine bias (e.g., bias power during chlorine adsorption) for a process involving adsorbing chlorine and igniting the plasma at 900 W without argon sputtering. The solid line depicts the curve of the etch rate of tungsten as a function of chlorine bias for a process involving adsorbing chlorine and igniting the plasma at 900 W followed by argon bombardment at a bias power of 60 V. As shown in Figure 8 the chlorine bias threshold voltage is approximately 60 V. Note that if the chlorine bias is less than 60 V, tungsten will not be etched without ion bombardment with argon. If the chlorine bias is greater than 60 V, the etch rate of tungsten without argon ion bombardment is much lower than that with argon ion bombardment. These results indicate that argon ion bombardment can be used to regulate the rate of tungsten etching performed by the ALE method in various embodiments, whereby 1) during chlorination, when there is no etching, chlorine is adsorbed on the tungsten substrate, and 2) during argon ion bombardment, the bias power is controlled to reduce or prevent physical removal (or sputtering) by setting a bias power lower than the sputtering threshold.

[0143] Experiment 2

[0144] Experiments were conducted on substrates having features to be filled with tungsten. The features were lined with a titanium nitride (TiN) barrier layer. Tungsten nucleated on the surface of the features, and tungsten was deposited by atomic layer deposition (alternating pulses of WF6 and B2H6). Figure 9A Shows a 20-nanometer feature 912 in substrate 910 lined with a TiN barrier layer 914 and a conformal tungsten layer 916. An opening 918a is shown at the top of the feature.

[0145] In Figure 9A the substrate was exposed to 10 ALE cycles involving alternating pulses of (1) Cl2 / BCl3 (in-situ inductively coupled plasma power of 900 W, no bias, 60 °C) and (2) argon (at a lower pressure ratio than (1), plasma of 300 W, 60 °C, bias of 60 Vb). The resulting substrate is shown in Figure 9B . Note that the opening 918b opens, thereby enabling subsequent tungsten deposition into the feature to completely fill the feature. Table 1 below shows the measurement results of the thickness of tungsten deposited in different parts of the substrate and the average thickness of the TiN barrier layer as well as in the trench opening. The measurement results are shown in nanometers.

[0146] Table 1. Measurement Results Before and After ALE

[0147]

[0148] The substrate is further exposed to 5 ALE cycles that involve alternating pulses of (1) Cl2 / BCl3 (900 W of in-situ inductively coupled plasma power, no bias, 60 °C), and (2) argon (at a lower pressure than (1), 300 W of plasma, 60 °C, 60 Vb of bias). The resulting measurements are shown in Table 2 below.

[0149] Table 2. Measurements before and after ALE

[0150]

[0151] These results show that the disclosed embodiments enable precise control of the amount of tungsten film etched based on the number of cycles, parameters, and other factors. For example, to etch more tungsten, multiple cycles can be performed. The results in Table 2 show some tungsten grooves due to the ALE process, but subsequent tungsten deposition cycles can restore the tungsten etched away in the ALE. The TiN barrier layer remains on the substrate, and the etching cycles of the ALE can be adjusted to ensure that enough tungsten remains on the surface of the feature so as not to expose the TiN barrier layer.

[0152] Conclusion

[0153] While the foregoing embodiments have been described in considerable detail for purposes of clear understanding, it will be apparent that certain changes and modifications may be practiced within the scope of the appended claims. It should be noted that there are many alternatives for the processes, systems, and devices having the embodiments of the present invention. Therefore, the embodiments of the present invention should be regarded as illustrative rather than restrictive, and the embodiments are not limited to the details given herein.

Claims

1. A method of manufacturing a semiconductor device, the method comprising: Relative to the interior region of a feature, preferentially etch a first quantity of metal at or near an opening of the feature on a substrate by: (i) exposing the feature to a halogen-containing gas to form a modified surface of the first quantity of metal, wherein relative to the interior region of the feature, the surface of the first quantity of metal at or near the opening of the feature is preferentially modified; (ii) exposing the modified surface to an activation gas to preferentially etch the first quantity of metal at or near the opening of the feature relative to the interior region of the feature to open the feature; (iii) applying a bias to the substrate using a bias power during at least one of (i) and (ii).

2. The method according to claim 1, wherein the metal comprises tungsten or molybdenum.

3. The method according to claim 1, wherein when a bias is applied during (i), the bias power is greater than 0V and less than about 200V.

4. A method of manufacturing a semiconductor device, comprising: Relative to the interior region of a feature, preferentially etch a first quantity of metal at or near an opening of the feature on a substrate by: (i) exposing the feature to a halogen-containing gas to form a modified surface of the first quantity of metal, wherein relative to the interior region of the feature, the surface of the first quantity of metal at or near the opening of the feature is preferentially modified; and (ii) exposing the modified surface to an activation gas at a chamber pressure between about 1 mTorr and about 15 mTorr to preferentially etch the first quantity of metal at or near the opening of the feature relative to the interior region of the feature to open the feature.

5. The method according to claim 4, wherein the metal comprises tungsten or molybdenum.

6. A method of manufacturing a semiconductor device, the method comprising: (a) Form a second metal film on a first metal film formed on a surface of the substrate by alternately supplying a metal-containing gas and a reactive gas to the substrate; (b) Supply a modifying gas to the substrate to modify the second metal film formed on the substrate before providing an etch gas; and (c) Provide an etch gas to the substrate to remove a portion of the second metal film while leaving other portions of the second metal film on the first metal film, wherein the portion of the second metal film is removed by alternately repeating (b) and (c), and wherein the second metal film is grown on the first metal film by alternately repeating (a) and (c), wherein supplying the modifying gas includes directionally modifying the second metal film at or near an opening of the feature relative to the interior of the feature on the substrate, and wherein supplying the etch gas includes directionally etching the second metal film at or near an opening of the feature relative to the interior of the feature to open the feature.

7. The method according to claim 6, wherein the modifying gas comprises a halogen-containing gas.

8. The method according to claim 7, wherein the etching gas comprises an inert gas.

9. The method according to claim 6, wherein the second metal film comprises a tungsten film or a tungsten nitride film.

10. The method according to claim 6, wherein the metal-containing gas comprises tungsten fluoride, the reaction gas comprises a gas selected from the group consisting of diborane, silane, and disilane, and the second metal film comprises a tungsten film.

11. The method according to claim 6, wherein the substrate comprises one or more underlying layers, the underlying layer comprising a dielectric layer and a metal layer.

12. The method according to claim 6, wherein the reaction gas comprises a reducing gas, wherein the reducing gas comprises hydrogen, diborane, silane, disilane, or a combination thereof.

13. The method according to claim 6, wherein the second metal film is deposited on the feature of the substrate, wherein the feature has an aspect ratio of at least about 10:

1.

14. The method according to claim 6, wherein operations (a), (b), and (c) are performed without breaking the vacuum.

15. A substrate processing apparatus, comprising: A processing chamber for accommodating a substrate; A gas supply system configured to supply a metal-containing gas, a reactive gas, a modifying gas, and an etch gas into the processing chamber; and A controller configured to control the gas supply system to perform the following operations: (a) Form a second metal film on a first metal film formed on a surface of the substrate by alternately supplying the metal-containing gas and the reactive gas to the substrate; (b) Supply a modifying gas to the substrate to modify the second metal film formed on the substrate before providing an etch gas; and (c) Providing an etching gas to the substrate to remove a portion of the second metal film while leaving other portions of the second metal film on the first metal film, wherein the portion of the second metal film is removed by alternately repeating (b) and (c), and wherein the second metal film is grown on the first metal film by alternately repeating (a) and (c). Wherein the controller configured to supply the modifying gas is configured to perform directional modification of the second metal film at or near an opening of the feature relative to an interior of the feature on the substrate, and wherein the controller configured to supply the etching gas is configured to perform directional etching of the second metal film at or near the opening of the feature relative to the interior of the feature to open the feature.

16. The substrate processing apparatus according to claim 15, wherein the gas supply system is configured to supply a halogen-containing gas as the modifying gas.

17. The substrate processing apparatus according to claim 16, wherein the gas supply system is configured to supply an inert gas as the etching gas.

18. The substrate processing apparatus according to claim 15, wherein the second metal film comprises a tungsten film or a tungsten nitride film.

19. The substrate processing apparatus according to claim 15, wherein the metal-containing gas comprises tungsten fluoride, the reaction gas comprises a gas selected from the group consisting of diborane, silane, and disilane, and the second metal film comprises a tungsten film.

20. The substrate processing apparatus according to claim 15, wherein the substrate comprises one or more lower layers, and the lower layer comprises a dielectric layer and a metal layer.

21. The substrate processing apparatus according to claim 15, wherein the reaction gas comprises a reducing gas, and the reducing gas comprises hydrogen, diborane, silane, disilane, or a combination thereof.

22. The substrate processing apparatus according to claim 15, wherein the second metal film is deposited on the feature of the substrate, and the feature has an aspect ratio of at least about 10:

1.

23. The substrate processing apparatus according to claim 15, wherein the controller is configured to perform operations (a), (b), and (c) without breaking the vacuum.

Images