In-situ high power implantation for stress relief in thin films

Abstract

Embodiments of the present disclosure generally relate to techniques for depositing high-density films for patterning applications. In one embodiment, a method for processing a substrate is provided. The method includes: depositing a carbon hard mask over a film stack formed on a substrate, wherein the substrate is positioned on an electrostatic chuck disposed in a processing chamber; implanting ions into the carbon hard mask, wherein depositing the carbon hard mask and implanting the ions into the carbon hard mask are performed in the same processing chamber; and repeatedly depositing the carbon hard mask and implanting the ions into the carbon hard mask in a cyclic manner until a predetermined thickness of the carbon hard mask is reached.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to fabricating integrated circuits. More specifically, embodiments described herein provide techniques for depositing high density films for patterning applications. Background Art

[0002] Hard molds are used to manufacture NAND and dynamic random access memory (DRAM) devices. Hard molds are often used as sacrificial layers during photolithography patterning and are capable of patterning features onto one or more material layers of a semiconductor device via an etching process. For example, the patterned features may form transistors and interconnects that allow NAND and DRAM devices to operate.

[0003] For example, some important properties of hard mask materials are etch resistance and compressive stress. An ideal hard mask has high etch resistance to the etchant used in the etching process when compared to the layer to be etched (hereinafter "underlayer"). As a result, the underlayer can be etched at a much faster rate than the hard mask. An ideal hard mask also has low compressive stress. Lower compressive stress eliminates undesirable substrate bending after hard mask deposition, which can further make device manufacturing difficult.

[0004] To improve the etch selectivity of the hardmask, high-density carbon films and doped carbon films have been developed. One of the challenges of utilizing these new films is that high-density carbon films exhibit high compressive stress, which results in undesirable substrate bending.

[0005] Thus, there is a need in the art for improved methods of forming hardmasks that exhibit increased etch selectivity while maintaining or reducing the compressive stress of the hardmask material. Summary of the invention

[0006] Embodiments of the present disclosure generally relate to techniques for depositing high-density films for patterning applications. In one embodiment, a method for processing a substrate is provided. The method includes: depositing a carbon hard mask over a film stack formed on a substrate, wherein the substrate is positioned on an electrostatic chuck disposed in a processing chamber; implanting ions into the carbon hard mask, wherein depositing the carbon hard mask and implanting the ions into the carbon hard mask are performed in the same processing chamber; and repeatedly depositing the carbon hard mask and implanting the ions into the carbon hard mask in a cyclic manner until a predetermined thickness of the carbon hard mask is reached.

[0007] In another embodiment, a method of processing a substrate is provided. The method includes: depositing a carbon hard mask over a substrate, wherein the carbon hard mask is deposited by applying an RF bias to an electrostatic chuck to generate a plasma, the substrate being positioned on the electrostatic chuck; and while depositing the carbon hard mask over the substrate, implanting ions from the plasma into the carbon hard mask using the RF bias, wherein depositing the carbon hard mask and implanting the ions into the carbon hard mask are performed simultaneously in the same processing chamber.

[0008] In yet another embodiment, a method for processing a substrate is provided. The method includes: depositing a diamond-like carbon hard mask over a film stack formed on a substrate by applying a first RF power having a first power level to an electrostatic chuck via a first electrode, wherein the substrate is positioned on the electrostatic chuck, and the first electrode is disposed in the electrostatic chuck; implanting ions into the diamond-like carbon hard mask, wherein depositing the diamond-like carbon hard mask and implanting ions into the diamond-like carbon hard mask are performed in the same processing chamber; repeatedly depositing the diamond-like carbon hard mask and implanting ions into the diamond-like carbon hard mask in a cyclic manner until a predetermined thickness of the diamond-like carbon hard mask is reached; patterning the diamond-like carbon hard mask; etching the film stack using the patterned diamond-like carbon hard mask; and removing the diamond-like carbon hard mask. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to be able to understand in detail the manner in which the above-mentioned features of the present disclosure are used, a more particular description of the present disclosure briefly summarized above may be made with reference to embodiments, some of which are shown in the accompanying drawings. It will be noted, however, that the accompanying drawings illustrate only typical embodiments of the present disclosure and are not to be considered limiting of its scope, as the present disclosure may admit to other equally effective embodiments.

[0010] Figure 1A and Figure 1B is a flow chart illustrating a method for forming a hard mask on a film stack according to an embodiment of the present disclosure.

[0011] FIG. 2A to FIG. 2F It shows that according to Figure 1A and Figure 1B Schematic cross-sectional view of a stack of hard mold forming sequences of the method.

[0012] Figure 3A and Figure 3B is a flow chart illustrating a method for depositing a hard mask on a film stack according to an embodiment of the present disclosure.

[0013] Figure 4 are graphs depicting (1) stress as a function of bias power and (2) density as a function of bias power for a diamond-like carbon layer formed according to one or more embodiments of the present disclosure.

[0014] Figure 5is a graph depicting stress as a function of implantation dose for a diamond-like carbon layer formed according to one or more embodiments of the present disclosure.

[0015] Figure 6 is a schematic cross-sectional illustration of an exemplary process that can be used to practice the methods described herein.

[0016] To facilitate understanding, identical reference numerals have been used, where possible, to identify identical components that are common to the figures. It is contemplated that components and features of one embodiment may be beneficially incorporated in other embodiments without further recitation. DETAILED DESCRIPTION

[0017] Embodiments described herein include improved methods of manufacturing carbon films having high density (e.g., >1.8 g / cc), high modulus (e.g., >150 GPa), and low stress (e.g., <-500 MPa). Specifically, an in-situ deposition implantation process for forming a high density carbon film with increased hardness and reduced stress in a deposition chamber is disclosed. The in-situ deposition implantation process can be performed in a cyclic or simultaneous manner to allow the formation of a carbon film having any target thickness without being limited to an ion penetration threshold, which would otherwise be encountered if the deposition and implantation processes were performed ex situ. The carbon films manufactured according to the embodiments described herein are amorphous in nature and have higher etch selectivity and very large modulus (e.g., >150 GPa) and lower stress (<-500 MPa) compared to conventional patterned films. The carbon films manufactured according to the embodiments described herein not only have low stress, but also have high sp 3 Carbon content (eg, diamond-like carbon films). In general, the deposition processes described herein are also fully compatible with current integration schemes for hard mask applications.

[0018] Although high-density carbon films are discussed in the present disclosure, it is contemplated that various embodiments of the present disclosure may also be used to improve the film stress, density, or Young's modulus of any film. In addition, it is contemplated that aspects of the present disclosure may be used in any deposition process or patterning scheme utilizing a hard mask or protective sacrificial layer, such as a self-aligned triple patterning (SATP) process, a self-aligned quadruple patterning (SAQP) process, a via / hole shrink process, a back end of line (BEOL), etc., as employed in various semiconductor processes (such as NAND flash applications, DRAM applications, or CMOS applications, etc.).

[0019] The embodiments described herein will be described below with reference to a PECVD process, which can be performed using any suitable thin film deposition system. Examples of suitable systems include those that can be used Processing chamber SYSTEM, PRECISION system, system, GT TM system, XPPrecision TM system, SE TM system, Processing chamber, and Mesa TM processing chambers, all of which are available from Applied Materials, Inc. in Santa Clara, California. Other tools capable of performing PECVD processes may also be adapted to benefit from the embodiments described herein. In addition, any system that implements the PECVD process described herein may be used.

[0020] Figure 1A and Figure 1B is a flow chart illustrating a method 100 for forming a hard mask on a film stack disposed on a substrate according to an embodiment of the present disclosure. FIG. 2A to FIG. 2F is a schematic cross-sectional view of a stack 200 showing a sequence of hard mask formation according to method 100. The hard mask may be a diamond-like carbon layer as described above, and may be used to fabricate a stair-like structure in a film stack of a three-dimensional semiconductor device, or any suitable device manufacturing application. It should also be understood that in Figure 1A and Figure 1B The operations depicted in the examples may be performed simultaneously and / or in parallel. Figure 1A and Figure 1B The order depicted in the figure is different from the order in which it is executed.

[0021] The method 100 is performed by stacking (such as Figure 2AOperation 102 is begun by positioning a stack 200 (depicted) into a processing chamber (such as a PECVD chamber). The stack 200 may be positioned on an electrostatic chuck disposed in the PECVD chamber. However, any suitable substrate support may be used in place of the electrostatic chuck. Once the stack 200 is positioned on the electrostatic chuck, a clamping voltage (constant or pulsed) is applied to the electrostatic chuck to clamp a substrate 202 to the electrostatic chuck. The stack 200 includes a substrate 202 and one or more layers disposed thereon. In the example shown, the substrate 202 has a film stack 204 disposed thereon. The substrate 202 may be a silicon-based material or any suitable insulating or conductive material, as desired. For example, the substrate 202 may be a material such as crystalline silicon (e.g., Si <100> or Si <111> ), silicon oxide, strained silicon, silicon germanium, doped or undoped polysilicon, doped or undoped silicon substrate, and patterned or unpatterned silicon on insulator (SOI), carbon-doped silicon oxide, silicon nitride, doped silicon, germanium, gallium arsenide, glass, sapphire. The substrate 202 may have various sizes, such as 200 mm, 300 mm, and 450 mm or other diameters, and a rectangular or square flat plate shape. Unless otherwise mentioned, the embodiments and examples described herein are performed on a substrate having a 200 mm diameter, a 300 mm diameter, or a 450 mm diameter. In embodiments where an SOI structure is used for the substrate 202, the substrate may include an embedded dielectric layer disposed on a silicon crystalline substrate. In the embodiments depicted herein, the substrate 202 may be a crystalline silicon substrate.

[0022] The film stack 204 may be a single layer or a plurality of vertically stacked layers. For example, the film stack 204 may include a plurality of pairs of first layers (not shown) and second layers (not shown) repeatedly formed in the film stack 204. The pair includes a first layer and a second layer that are alternately formed repeatedly until a desired number of pairs of first layers and second layers are reached. The film stack 204 may be a part of a semiconductor chip (such as a three-dimensional memory chip). In one embodiment, the film stack 204 is used to form a plurality of gate structures of a three-dimensional memory chip. In this case, the first layer formed in the film stack 204 may be a first dielectric layer, and the second layer formed in the film stack 204 may be a second dielectric layer. Suitable dielectric layers for the first and second layers may include silicon oxide, silicon nitride, silicon oxynitride, silicon carbide, silicon oxycarbide, titanium nitride, a composite of oxides and nitrides, at least one or more oxide layers sandwiching a nitride layer, and combinations thereof, and the like. In some embodiments, one or both of the first and second dielectric layers may be a high dielectric constant material having a dielectric constant greater than 4. Suitable examples of high dielectric constant materials include hafnium dioxide (HfO2), zirconium dioxide (ZrO2), hafnium silicon dioxide (HfSiO2), hafnium aluminum oxide (HfAlO), zirconium silicon dioxide (ZrSiO2), tantalum dioxide (TaO2), aluminum oxide, aluminum-doped hafnium dioxide, bismuth strontium titanium (BST), and platinum zirconium titanium (PZT), etc. The film stack 204 may have a dielectric constant of about 100 Å. About In one embodiment, the total thickness of the film stack 204 is about 3 microns to about 10 microns, and will vary as technology advances.

[0023] During operation 102, several process parameters may be adjusted. In one embodiment suitable for processing a 300 mm substrate, a process pressure in the process volume may be maintained at about 0.1 mTorr to about 10 Torr (e.g., about 2 mTorr to about 50 mTorr; or about 5 mTorr to about 20 mTorr). A process temperature and / or a substrate temperature may be maintained at about -50 degrees Celsius to about 350 degrees Celsius (e.g., about 0 degrees Celsius to about 50 degrees Celsius; or about 10 degrees Celsius to about 20 degrees Celsius).

[0024] At operation 104, a hydrocarbon-containing gas is flowed into the processing chamber. The hydrocarbon-containing gas may flow into the processing chamber through a gas distribution assembly (disposed at the top of the processing chamber above the electrostatic chuck) or through a sidewall of the processing chamber. The hydrocarbon-containing gas may include at least one hydrocarbon compound. The hydrocarbon compound may be any liquid or gas. In one embodiment, the hydrocarbon compound is a gaseous hydrocarbon. In another embodiment, the hydrocarbon compound may initially be a liquid and may be delivered to the processing volume via a vaporizer or bubbler, or other liquid precursor delivery system.

[0025] In one embodiment, the hydrocarbon compound has the general formula C x H y , wherein x has a range between 1 and 20, and y has a range between 1 and 20. For example, suitable hydrocarbon compounds may include acetylene (C2H2), ethylene (C2H4), ethane (C2H6), propyne (C3H4), propylene (C3H6), propane (C3H8), butane (C4H 10 ), methane (CH4), butene (C4H8), butane (C4H 10 ), pentane (C5H 12 ), hexane (C6H 14 ), adamantane (C 10 H 16 ), norbornene (C7H 10 ), or a combination thereof. C2H2 may be advantageous due to the formation of a more stable intermediate species that allows for more surface mobility.

[0026] In one embodiment, the hydrocarbon compound is an aromatic hydrocarbon compound such as benzene, styrene, toluene, xylene, ethylbenzene, acetophenone, methyl benzoate, phenyl acetate, phenol, cresol, furan, and the like, alpha-terpinene, cymene, 1,1,3,3-tetramethylbutylbenzene, tert-butyl ether, tert-butylethylene, methyl methacrylate, and tert-butyl furfuryl ether, compounds having the chemical formula C3H2 and C5H4, halogenated aromatic compounds including monofluorobenzene, difluorobenzene, tetrafluorobenzene, hexafluorobenzene, or combinations thereof. In some cases, hydrocarbon compounds containing oxygen and halogenated precursors are not required.

[0027] In some embodiments, an inert gas such as argon (Ar) and / or helium (He) may be supplied into the processing chamber along with the hydrocarbon-containing gas. Other inert gases such as nitrogen (N2) and nitric oxide (NO) may also be used to control the density and deposition rate of the diamond-like carbon layer.

[0028] In some embodiments, the hydrocarbon-containing gas may further include one or more diluent gases. Suitable diluent gases may include, but are not limited to, helium (He), argon (Ar), xenon (Xe), hydrogen (H2), nitrogen (N2), ammonia (NH3), nitric oxide (NO), or a combination thereof. Ar, He, and N2 can be used to control the density and deposition rate of the diamond-like carbon layer. In some cases, the addition of N2 and / or NH3 can be used to control the hydrogen ratio (e.g., carbon to hydrogen ratio) of the diamond-like carbon layer.

[0029] At operation 106, if Figure 2B As shown in , plasma is generated in a processing chamber from a gas mixture to form a hard mask 206 on a film stack 204. The plasma can be generated by applying a first RF power (bias or power) to an electrostatic chuck. The first RF power can be from about 1 kilowatt to about 10 kilowatts at a frequency of about 0.4 MHz to about 300 MHz (e.g., about 11 MHz to about 60 MHz). In one embodiment, the first RF power is provided at a frequency of about 2 kilowatts and about 13.56 MHz. The first RF power can be provided from an RF power generator to the electrostatic chuck via a first electrode disposed in the electrostatic chuck. In some cases, the first electrode can also be electrically connected to a clamping power supply that provides direct current (DC) power to electrostatically fix the substrate 202 to the upper surface of the electrostatic chuck.

[0030] Additionally or alternatively, the first RF power may be applied to an upper electrode, such as a showerhead disposed at the top of the processing chamber opposite the electrostatic chuck. In some embodiments, the first RF power may be applied to at least one of the upper electrode, the bottom electrode (e.g., the first electrode), and an ICP coil surrounding a portion of the processing chamber. The ICP coil may be used to form a plasma or tune the uniformity of the plasma within the processing chamber. Depending on the power supply scheme, the top electrode, the bottom electrode, and the ICP coil may be powered simultaneously, or two of the three may be powered simultaneously. The applied RF frequency may be in the range from hundreds of kHz to tens of MHz. Multiple frequencies may also be applied to the top electrode or the bottom electrode to optimize the ion flux and the energy incident on the substrate.

[0031] In one embodiment, the hard mask 206 is a diamond-like carbon film as discussed above. The hard mask 206 can be deposited by a blanket deposition process over the film stack 204. Note that the hard mask 206 can be formed on any surface or any portion of the substrate 202 with or without the film stack 204 on the substrate 202. In some embodiments, the processing conditions established during operation 102 are maintained during operations 104 and 106. In one embodiment, the pressure of the processing chamber during the formation of the hard mask 206 is maintained at about 2 mTorr to about 20 mTorr.

[0032] At operation 108, Figure 2C As shown in , an ion implantation process is performed in a processing chamber to process the hard mask 206. In one embodiment, the ion implantation process is performed in-situ in the same processing chamber where the deposition process of the hard mask 206 (i.e., operation 106) is performed. During operation 108, the flow of the hydrocarbon-containing gas is interrupted, while the inert gas and / or diluent gas supplied during operation 104 may be maintained. At the same time, the first RF power applied to the electrostatic chuck during operation 106 is reduced to a minimum level sufficient to maintain the plasma (i.e., the plasma is continuous during operations 106 and 108). For example, the first RF power may be reduced from about 2 kilowatts to about 200 watts.

[0033] The ion implantation process is performed by implanting ions, such as ions from a continuous flow of an inert gas, and ions from a hydrocarbon-containing gas still remaining in the plasma. In some embodiments, a direct current (DC) bias is used to implant ions into the hard mold 206. The DC bias is overlaid on the reduced first RF power. In particular, the DC bias drives high monoenergetic ions into the hard mold 206. The DC bias can be provided to the electrostatic chuck via a second electrode. The second electrode can be disposed in the electrostatic chuck and electrically connected to a DC power supply, which provides the bias to the second electrode. Alternatively, the DC bias can be provided to the first electrode from a clamping power supply. In any case, the DC bias can be between about 2 kilovolts and about 15 kilovolts. In one embodiment, the DC bias is between about 5 kilovolts and about 12 kilovolts, for example about 8 kilovolts. Since the DC bias is high, the DC bias can be pulsed during the ion implantation process. In various embodiments, the pulse width of the DC bias can be on the order of about 1 microsecond to about 1 millisecond. In some embodiments, the DC bias is applied at a pulse frequency of 10 Hz to about 10 kHz, with a pulse width of about 5 microseconds to about 30 milliseconds.

[0034] In some embodiments, ions are implanted into the hard mold 206 using a second RF power (bias or power). Again, the second RF power is overlaid on the reduced first RF power. The second RF power may be provided to the electrostatic chuck from an RF power generator via a third electrode disposed in the electrostatic chuck. The second RF power may be from about 1 kilowatt to about 10 kilowatts at a frequency of about 0.4 MHz to about 300 MHz (e.g., about 11 MHz to about 60 MHz). In one embodiment, the first RF power is provided at about 2 kilowatts and a frequency of about 13.56 MHz. The second RF power may be pulsed during the ion implantation process. For example, the second RF power may be pulsed with a duty cycle in the range of about 10% to about 80%, with a frequency of about 1 Hz to about 50 kHz.

[0035] In the case where the second RF power is used and the deposited species is also used as the implanted species (e.g., H2 from a hydrocarbon-containing gas), the reduced first RF power and the second RF power can be temporarily biased so that the reduced first RF power and the pulsed second RF power are synchronous or asynchronous, thereby separating the deposition phase and the processing phase (i.e., ion implantation). For example, when both the second RF power and the reduced first RF power are on, the ions are accelerated, causing the impact of the hard mask 206 to occur, and because most of the film growth comes from the main plasma (e.g., the first RF power) operating at a low level (e.g., 200 watts), the deposition of the hard mask 206 can be minimized. Thus, ion implantation dominates. When the second RF power is pulsed and the reduced first RF power is on, the deposition of the hard mask 206 increases and becomes the dominant process. Thus, film deposition dominates.

[0036] Regardless of whether the ion implantation process uses a DC bias or a second RF power, little or no deposition of the hard mask 206 will occur because the flow of the deposition gas (i.e., the hydrocarbon-containing gas) is turned off. Thus, the pulses of the DC bias or the pulsed second RF power separate the deposition phase and the treatment phase (i.e., the ion implantation process), thereby making the formation and treatment of the hard mask 206 a cyclic deposition-treatment process. During the ion implantation process, ions from the continuous flow of the inert gas (such as argon or helium ions) and ions from the hydrocarbon-containing gas remaining in the plasma are attracted or driven by the DC bias or the second RF power and move forward to the hard mask 206. The DC bias or the second RF power is used to treat the hard mask 206 by bombarding the surface of the hard mask 206 with ions. Therefore, the stress in the deposited hard mask 206 is reduced.

[0037] The ion implantation process may be performed until the implanted ions reach a penetration threshold due to the implanted ions gradually losing energy as they travel through the hard mask 206. The penetration threshold may be determined by the penetration depth of the ions. Alternatively, the ion implantation process may be performed until a predetermined implantation depth is reached. The predetermined implantation depth or ion penetration threshold may be in the range between 10 nanometers and 1 micron, which may vary depending on the type and size of the ions and the bias voltage used to excite the ions 207.

[0038] Depending on the desired implantation depth, the implantation energy may be between about 0.5 keV and about 60 keV, such as about 6 keV to about 45 keV. The ion dose may be between about 1×10 13 cm -2 About 1x10 17 cm -2 In the range of, for example, about 5x10 16 cm -2. Extremely low pressures (e.g., 2 mTorr to about 20 mTorr) enable very high energy ions to treat / implant the surface of the hard mask 206. Without being bound by any particular theory, it is believed that the implanted ions can extract remaining hydrogen atoms from the pendant carbon-hydrogen bonds of the hard mask 206 and form a carbide structure within the hard mask 206. The carbide structure exhibits increased hardness when compared to an untreated hard mask. It is also believed that the implanted ions occupy interstitial voids present within the hard mask 206, which can result in an increased density of the hard mask 206. The increased density further increases the mechanical integrity of the hard mask 206. The increased hardness and density of the hard mask 206 can provide a mechanically robust hard mask 206, which in turn results in improved etching selectivity and reduced internal stress. Thus, undesirable substrate deformation is eliminated.

[0039] Because the implantation process occurs in the same processing chamber where the hard mold deposition occurs, the in-situ deposition-implantation process is advantageous. Thus, the hard mold 206 can be deposited and processed without having to break the vacuum and transfer to an ectopic implantation tool. Therefore, the overall yield is improved and the cost associated with the ectopic implantation tool is reduced. In addition, since the implanted ions have a penetration threshold, the thickness of the hard mold that can be processed ectopically is limited. Using the in-situ deposition / ion implantation process, the deposition and processing of the hard mold can be performed in a cyclic manner to customize the hard mold to any target thickness possible in the same processing chamber.

[0040] At operation 110, a decision is made to determine whether the deposited hardmask 206 that has been processed has reached a target thickness. The hardmask 206 may have a target thickness corresponding to the subsequent etching requirements of the film stack 204. In one example, the target thickness of the hardmask 206 is between about 0.5 μm and about 1.5 μm, such as about 1.0 μm. If the target thickness of the hardmask 206 has not been reached, another cycle of the deposition / ion implantation process (e.g., operations 104, 106, and 108) may be performed before the thickness of the processed hardmask 206 is compared to the target thickness again. In some embodiments where the implanted ions reach a predetermined implantation or penetration depth but the hardmask 206 has not reached a desired thickness, another cycle of the deposition / ion implantation process (e.g., operations 104, 106, and 108) may be performed before the thickness of the processed hardmask 206 is compared to the target thickness again. The cyclic process of in-situ hardmask deposition and ion implantation is repeated until the deposited hardmask 206 reaches the target thickness.

[0041] At operation 112, once the hard mask 206 reaches the target thickness, such as Figure 2DAs shown, a patterned photoresist layer 208 is formed over the processed hard mask 206. An energy source (such as light energy) can be used to transfer features or patterns from the photomask to the photoresist 208. In one embodiment, the photoresist is a polymeric material and the patterning process is performed by a 193nm immersion lithography process or other similar lithography process. Similarly, lasers can also be used to perform the patterning process.

[0042] At operation 114, if Figure 2E As shown in , the processed hard mask 206 is patterned by, for example, photolithography and one or more etching processes to transfer features from the photoresist 208 to the hard mask 206. The etching process can be performed in any suitable etching chamber (such as a plasma etching chamber). Thereafter, the photoresist layer 208 is removed by any suitable process, such as an ashing process or a wet etching process.

[0043] At operation 116, if Figure 2F As shown in , the film stack 204 is etched using the patterned hard mask 206. The etching of the film stack 204 can be performed in any suitable processing chamber, such as a plasma etching chamber. An etchant, such as fluorocarbon, can be used to remove the exposed portion of the film stack 204. The active species of the etchant is selective so that the active species does not substantially react with the implanted ions of the hard mask 206. Therefore, the etchant is selective to the film stack 204. Suitable examples of etchants may include, but are not limited to, CF4, CHF3, HBr, BCl3, or Cl2. The etchant may be provided with an inert carrier gas. The hard mask 206 is subsequently removed using any suitable hard mask removal process. For example, oxygen plasma can be used to remove the hard mask 206. The resulting stack 200 includes the film stack 204 having features, such as high aspect ratio features, formed therein. The resulting stack 200 may then undergo further processing to form a functional semiconductor device.

[0044] Figure 3A and Figure 3B FIG. 3 is a flow chart of a method 300 for depositing a hard mask on a film stack disposed on a substrate according to an embodiment of the present disclosure. FIG. 2A to FIG. 2F As shown in FIG. 3 , method 300 may be used to process stack 200. FIG. 2A to FIG. 2F Method 300 is described. Likewise, the deposited hard mask may be a diamond-like carbon layer as described above, and may be used to fabricate step-like structures in film stacks for three-dimensional semiconductor devices or any suitable device fabrication application. It should also be understood that Figure 3A and Figure 3B The operations depicted in the examples may be performed simultaneously and / or in parallel. Figure 3A and Figure 3B The order depicted in the figure is different from the order in which it is executed.

[0045] Operations 302 and 304 of method 300 are similar to operations 102 and 104 and therefore will not be described again for the sake of brevity. Figure 2A As shown, the stack 200 is formed by depositing a film stack 204 over a substrate 202. At operation 306, as shown in FIG. Figure 2B As shown, a simultaneous process of hard mask deposition and ion implantation treatment is performed in a processing chamber to form a hard mask 206 on the film stack 204. The simultaneous process can be performed by generating a plasma from a gas mixture (e.g., a hydrocarbon-containing gas and an inert gas / diluent gas) in the processing chamber to form a layer of hard mask 206 above the film stack 204. The plasma can be generated at the substrate level by applying a high voltage RF bias to the electrostatic chuck. The high voltage RF bias can be in a range from about 1 kilowatt to about 15 kilowatts at a frequency of about 0.4 MHz to about 300 MHz (e.g., about 11 MHz to about 60 MHz). In one embodiment, the high voltage RF bias is provided at a frequency of about 8 kilowatts and about 13.56 MHz. The high voltage RF bias can be provided to the electrostatic chuck from an RF power generator via a third electrode provided in the electrostatic chuck.

[0046] When depositing the hardmask 206 over the film stack 204, the high voltage RF bias applied to the electrostatic chuck can also be used as implant energy to attract ions from the gas mixture (i.e., hydrocarbon-containing gas, inert gas / diluent gas) moving toward the growing hardmask 206. Figure 2C As shown in FIG. 1 , the ion implantation process of the hard mask 206 is performed simultaneously with the hard mask deposition. The ions are attracted and accelerated by the high voltage RF bias to impact the surface of the hard mask 206 , thereby reducing the stress of the hard mask 206 .

[0047] In some embodiments, a DC bias may be additionally applied to the electrostatic chuck to facilitate the ion implantation process. Ions from the gas mixture (i.e., hydrocarbon-containing gas, inert gas / diluent gas) may be attracted and accelerated by the DC bias to impact the surface of the hard mask 206, thereby reducing the stress of the hard mask 206. Figure 1A and Figure 1B As discussed, a DC bias may be provided to the electrostatic chuck via a second electrode that may be disposed in the electrostatic chuck and in electrical communication with the DC power supply. Alternatively, a DC bias may be provided to the first electrode from the clamping power supply. The bias may be between about 1 kV and about 15 kV. In one embodiment, the DC bias is between about 2 kV and about 6 kV. In another embodiment, the bias is between about 8 kV and about 10 kV. Similar to operation 108, depending on the desired implant depth, the implant energy may be between about 0.5 keV and about 60 keV, for example, about 6 keV to about 45 keV. The ion dose may be between about 1 x 10 13 cm -2 About 1x10 17 cm-2 In the range of, for example, about 5x10 16 cm -2 .

[0048] In any case, the flow of the gas mixture (e.g., hydrocarbon-containing gas and inert / diluent gas) is continuous throughout operation 308. Thus, the deposition species (e.g., ions or neutral species of carbon, hydrogen, etc.) used for hard mask deposition can also be used as implant species for the ion implantation process. The in-situ process of hard mask deposition and ion implantation process enables simultaneous and more efficient deposition and processing of the hard mask 206, as compared to embodiments where the hard mask deposition and ion implantation processes are separated by short duty cycle RF or pulsed DC voltage.

[0049] At operation 308, a decision is made to determine whether the deposited / processed hardmask 206 has reached a target thickness. The hardmask 206 may have a target thickness corresponding to the subsequent etching requirements of the film stack 204. In one example, the target thickness of the hardmask 206 is between about 0.5 μm and about 1.5 μm, such as about 1.0 μm. If the target thickness of the hardmask 206 has not been reached, another cycle of the deposition / ion implantation process (e.g., operations 304 and 306) may be performed before the thickness of the processed hardmask 206 is again compared to the target thickness. The in-situ simultaneous process of hardmask deposition and ion implantation processing is repeated until the deposited / processed hardmask 206 reaches the target thickness.

[0050] Operations 310 , 312 , 314 , and 316 are similar to operations 112 , 114 , 116 , and 118 , and therefore will not be described for the sake of brevity.

[0051] Figure 4 is a description of the above Figure 3A and Figure 3B Graph 400 of (1) stress as a function of bias power (represented by squares) and (2) density as a function of bias power (represented by dots) for a diamond-like carbon layer (e.g., hard mask) formed according to one or more embodiments discussed. The x-axis represents bias power (watts) and the y-axis represents density of the deposited film (g / cc). Figure 4 , in general, as the bias power increases, the density of the deposited film increases. In particular, the stress of the diamond-like carbon layer is lower at higher RF power / ion energy.

[0052] Figure 5 5 is a graph 500 depicting stress versus implantation dose for a diamond-like carbon layer (e.g., hard mask) formed according to one or more embodiments of the present disclosure. Graph 500 depicts stress data for a diamond-like carbon film as deposited and post-ion implanted using an implantation energy of approximately 35 keV. The x-axis represents the stress of the deposited film (MPa), and the y-axis represents the implantation dose (AU). Figure 5 It is shown in Figure 3 that the stress of the deposited film is significantly reduced after the ion implantation process.

[0053] Figure 6 6 is a schematic cross-sectional view of an exemplary process that can be used to practice any one or combination of the methods described herein. The processing chamber 600 includes a chamber lid assembly 601, one or more sidewalls 602, and a chamber bottom 604. The chamber lid assembly 601 includes a chamber lid 606, a showerhead 607 disposed in the chamber lid 606, and an electrically insulating ring 608 disposed between the chamber lid 606 and the one or more sidewalls 602. The showerhead 607, the one or more sidewalls 602, and the chamber bottom 604 together define a processing volume 605. A gas inlet 609 disposed through the chamber lid 606 is fluidly coupled to a gas source 610. The showerhead 607, through which a plurality of openings 611 are disposed, is used to evenly distribute the processing gas from the gas source 610 into the processing volume 605. The showerhead 607 is electrically coupled to a first power supply 612, such as an RF power supply, which supplies power to ignite and maintain a plasma 613 of the processing gas via a capacitive coupling thereto. In other embodiments, the processing chamber 600 includes an inductive plasma generator and the plasma is formed by inductively coupling RF power to the process gas.

[0054] The processing volume 605 is fluidly coupled to a vacuum source, such as one or more dedicated vacuum pumps, via a vacuum outlet 614 that maintains the processing volume 605 at low atmospheric conditions and evacuates processing and other gases therefrom. A substrate support 615 disposed in the processing volume 605 is disposed on a movable support shaft 616 that seals through the chamber bottom 604, such as surrounded by a bellows (not shown), extending in a region below the chamber bottom 604. Herein, the processing chamber 600 is conventionally configured to facilitate transfer of a substrate 617 to and from the substrate support 615 through an opening 618 in one of the one or more sidewalls 602, which opening is conventionally sealed with a door or valve (not shown) during substrate processing.

[0055] Herein, a substrate 617 disposed on the substrate support 615 is maintained at a desired processing temperature using one or both of a heater (such as a resistive heating element 619) and one or more cooling channels 620 disposed in the substrate support 615. Typically, the one or more cooling channels 620 are fluidly coupled to a coolant source (not shown), such as a modified water source or coolant source having a relatively high electrical resistance. In some embodiments, the substrate support 615 or one or more electrodes 621 thereof are electrically coupled to a second power supply 622, such as a continuous wave (CW) RF power supply or a pulsed RF power supply, which supplies a bias voltage thereto.

[0056] The processing chamber 600 further includes a system controller 630 for controlling the operation of the processing chamber 600 and implementing the methods described herein. The system controller 630 includes a programmable central processing unit, herein a central processing unit (CPU 631), which can operate with a memory 632 (e.g., non-volatile memory) and support circuits 633. The support circuits 633 are coupled to the CPU 631 and include caches, clock circuits, input / output subsystems, power supplies, and combinations thereof coupled to various components of the processing chamber 600 to facilitate control thereof. The CPU 631 is one of any form of general-purpose computer processor, such as a programmable logic controller (PLC), for controlling the various components and subprocessors of the processing chamber 600. The memory 632 coupled to the CPU 631 is non-transitory and is typically one or more of readily available memories, such as random access memory (RAM), read only memory (ROM), a floppy disk drive, a hard disk, or any other form of digital storage (local or remote).

[0057] Typically, the memory 632 is in the form of a computer-readable storage medium (e.g., non-volatile memory) containing instructions that, when executed by the CPU 631, facilitate the operation of the processing chamber 600. The instructions in the memory 632 are in the form of a program product, such as a program that implements the methods of the present disclosure. The program code may conform to any of a number of different programming languages. In one example, the present disclosure may be implemented as a program product stored on a computer-readable storage medium for use with a computer system. The program of the program product defines the functionality of the embodiments (including the methods described herein).

[0058] Therefore, a method for forming a diamond-like carbon hard mold layer is provided, and the hard mold layer can be used to form a semiconductor device. By utilizing an in-situ hard mold deposition ion implantation process, the hard mold layer can be deposited and processed without having to break the vacuum and transfer to an ectopic implantation tool. Therefore, the overall yield is improved and the cost associated with the ectopic implantation tool is reduced. In addition, the deposition and processing of the hard mold can be performed in a cyclic or simultaneous manner using an independent plasma source or a high-power (high-voltage) plasma source to customize the hard mold to any possible target thickness in the same processing chamber. In addition, implanted ions are used to increase the mechanical integrity and density of the hard mold while maintaining or reducing the internal stress of the hard mold. The increased mechanical integrity and density of the hard mold reduce linear bending after etching the hard mold, and the maintenance or reduced stress of the hard mold reduces or eliminates undesirable substrate bending or deformation. In short, the etching selectivity of the hard mold is increased.

[0059] While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the disclosure may be devised without departing from the basic scope thereof, and the scope of the same is to be determined by the following claims.

Claims

1. A method for processing a substrate, comprising the following steps: depositing a carbon hardmask over a film stack formed on the substrate by applying a first RF power having a first power level, wherein the substrate is positioned on an electrostatic chuck disposed in a processing chamber; implanting ions into the carbon hardmask, wherein the steps of depositing the carbon hardmask and implanting ions into the carbon hardmask are performed in the same processing chamber; Wherein implanting ions into the carbon hard mask comprises: After the step of depositing the carbon hardmask over the film stack, interrupting the flow of the hydrocarbon-containing gas used to form the carbon hardmask; as well as reducing the first RF power from the first power level to a second power level sufficient to sustain a plasma in the processing chamber; and The steps of depositing the carbon hardmask and implanting ions into the carbon hardmask are repeated in a cyclic manner until a predetermined thickness of the carbon hardmask is reached. 2 . The method of claim 1 , wherein the first RF power having the first power level is applied to an electrostatic chuck via a first electrode disposed in the electrostatic chuck.

3. The method of claim 2, wherein the first RF power is in the range of 1 kW to 10 kW at a frequency of 0.4 MHz to 300 MHz.

4. The method of claim 1, wherein the step of implanting ions into the carbon hard mask is performed by applying a DC bias to the electrostatic chuck via a second electrode disposed in the electrostatic chuck. The method of claim 4 , wherein the DC bias voltage is in the range of 2 kV to 15 kV.

6. The method of claim 4, wherein the DC bias is provided at a pulse frequency of 10 Hz to 10 kHz, with a pulse width of 5 microseconds to 30 milliseconds.

7. The method of claim 1, wherein the step of implanting ions into the carbon hard mask is performed by applying a second RF power to the electrostatic chuck via a third electrode disposed in the electrostatic chuck.

8. The method of claim 7, wherein the second RF power is in the range of 1 kW to 10 kW at a frequency of 0.4 MHz to 300 MHz.

9. The method of claim 8, wherein the second RF power is pulsed with a duty cycle in a range from 10% to 80%.

10. A method for processing a substrate, comprising the following steps: depositing a diamond-like carbon hardmask over a film stack formed on a substrate by applying a first RF power having a first power level to an electrostatic chuck via a first electrode, wherein the substrate is positioned on the electrostatic chuck in which the first electrode is disposed; implanting ions into the diamond-like carbon hardmask, wherein the steps of depositing the diamond-like carbon hardmask and implanting ions into the diamond-like carbon hardmask are performed in the same processing chamber; Wherein implanting ions into the diamond-like carbon hard mask comprises: After depositing the diamond-like carbon hardmask over the film stack, interrupting the flow of the hydrocarbon-containing gas used to form the diamond-like carbon hardmask; and reducing the first RF power from the first power level to a second power level sufficient to sustain a plasma in the processing chamber; and Repeating the steps of depositing the diamond-like carbon hardmask and implanting ions into the diamond-like carbon hardmask in a cyclic manner until a predetermined thickness of the diamond-like carbon hardmask is reached; patterning the diamond-like carbon hardmask; etching the film stack using a patterned diamond-like carbon hardmask; and The diamond-like carbon hardmask is removed.

11. The method of claim 10, wherein the step of implanting ions into the diamond-like carbon hard mask is performed by applying a pulsed DC bias to the electrostatic chuck.

12. The method of claim 10, wherein the step of implanting ions into the diamond-like carbon hardmask is performed by applying a second RF power to the electrostatic chuck.

13. The method of claim 12, wherein the second RF power is pulsed with a duty cycle in a range from 10% to 80%.

Images