JP7812974B2 - Etching Method - Google Patents

Description

The present invention relates to an etching method, and in particular to a process technology for an isotropic dry etching method used in the process of removing silicon nitride films from semiconductor elements such as 3D memory.

Demands for lower power consumption and increased memory capacity are driving further miniaturization and three-dimensional device structures in semiconductor devices. In the manufacture of three-dimensional devices, due to their complex, three-dimensional structures, "isotropic etching," which allows for lateral etching as well as "vertical (anisotropic) etching," is increasingly used in addition to "vertical etching," which involves etching perpendicular to the wafer surface. Isotropic etching has traditionally been performed using wet processing with chemicals. However, with advances in miniaturization, problems such as pattern collapse due to the surface tension of the chemicals and residual etching in minute gaps have become apparent. Furthermore, the need for large amounts of chemical processing poses environmental and cost concerns. Therefore, there is a need to replace wet processing with dry processing for isotropic etching.

Silicon nitride films are widely used in semiconductor devices, and there are known examples of dry etching processes that use hydrogen fluoride (HF) gas but do not use plasma. For example, Patent Document 1 describes a method of supplying hydrogen fluoride gas at a wafer temperature of 60°C to 200°C to etch silicon nitride films without damaging thermal oxide films. Patent Document 2 also describes a method of selectively etching silicon nitride films relative to silicon oxide films by supplying hydrogen fluoride gas at a temperature of 10 to 120°C at a chamber pressure of 1333 Pa or higher.

As a known example of adding another component to HF gas, Patent Document 3 describes a method of selectively etching a silicon nitride film by supplying HF gas with nitric oxide (NO) gas and/or ozone gas. Patent Document 4 also describes a method of etching a silicon nitride film by contacting a mixed gas containing a fluorine-containing carboxylic acid and HF gas at less than 100°C without plasma.

As an example of etching with a fluorine-containing gas other than HF gas, Patent Document 5 discloses a method of selectively etching a silicon nitride film relative to a silicon oxide film using chlorine trifluoride ( _ClF3 ) gas. Patent Document 6 discloses a method of selectively etching a silicon nitride film using a fluorine-containing etching gas selected from the group consisting of nitrosyl fluoride (FNO), nitrogen trifluoride oxide ( _F3NO ), nitroyl fluoride ( _FNO2 ), and combinations thereof. Patent Document 7 discloses etching a silicon nitride film using an etching gas containing a halogen fluoride, which is a compound of bromine or iodine with fluorine, at a pressure of 1 Pa to 80 kPa without using plasma.

As an example of using radicals generated by some kind of plasma, Patent Document 8 describes a method in which a fluorine-containing gas, an alcohol gas, _O2 gas, and an inert gas are supplied in an excited state by external plasma, thereby selectively etching a silicon nitride film relative to a silicon and/or silicon oxide film. Patent Document 9 also describes a method for selectively etching a silicon nitride film, which includes a step of introducing a gas containing H and F and a step of selectively introducing inert gas radicals into a processing space. Furthermore, Patent Document 10 describes a method for selectively etching a silicon nitride film laterally from a stacked structure of silicon nitride and silicon oxide films at temperatures below -20°C using a plasma-generated oxygen-containing precursor and a fluorine-containing precursor.

Furthermore, Patent Documents 6 and 10 describe selective lateral etching of a silicon nitride film from the sidewall of a high aspect ratio opening formed in a multi-layered structure of silicon nitride films and silicon oxide films in a 3D-NAND device, which is a 3D memory.

Furthermore, Patent Document 11 discloses that ammonium silicofluoride [(NH ₄ ) ₂ SiF ₆ ], ammonium hydrogen fluoride [NH ₄ HF ₂ ], and the like formed on a silicon nitride film are removed by heating with a lamp or the like.

Japanese Patent Application Laid-Open No. 2008-187105 Japanese Patent Application Publication No. 2018-207088 Japanese Patent Application Laid-Open No. 2014-197603 Japanese Patent Application Publication No. 2019-091890 Japanese Patent Application Laid-Open No. 2016-58544 Special Publication No. 2021-509538 International Publication No. 2021/079780 Japanese Patent Application Laid-Open No. 2015-228433 Japanese Patent Application Publication No. 2019-012759 U.S. Patent No. 10,319,603 Japanese Patent Application Laid-Open No. 2005-161493

For example, in the processing of stacked films in 3D-NAND flash memory, a three-dimensional semiconductor device, and in the processing of the gate periphery of FinFETs, technology is required to etch silicon nitride films isotropically and selectively with atomic layer control relative to polycrystalline silicon films and silicon oxide films. In particular, in the 3D-NAND structure, silicon oxide films ( _SiO2 films) and silicon nitride films (SiN) are alternately stacked in large numbers, and deep holes and grooves are formed in the layers, so a process is required to selectively and isotropically etch a small amount of silicon nitride film laterally.

As noted in the background art, wet etching using aqueous hydrofluoric acid or buffered hydrofluoric acid solutions poses problems such as incomplete etching of minute gaps and poor etching controllability. Furthermore, dry etching using radicals is rate-limiting due to the supply of highly reactive radicals, which act as active species. Therefore, in the case of deep holes or trenches, the amount of silicon nitride film etched at the top near the surface is large and the amount of silicon nitride film etched at the bottom near the bottom is small, resulting in a top-to-bottom difference in etching amount. Furthermore, gas etching without plasma is difficult to achieve with high precision with a high selectivity relative to silicon oxide film, resulting in degradation of the shape of the silicon oxide film portions that should be retained. Furthermore, in the case of deep holes or trenches, reaction products can re-adhere at the top, resulting in a small amount of silicon nitride film etched at the top, resulting in a top-to-bottom difference in etching amount.

The present invention was made in consideration of the above-mentioned problems, and provides a method that allows highly selective and precise etching of a silicon nitride film relative to a silicon oxide film without deteriorating the shape of the silicon oxide film that is desired to be left, and further reduces the difference between the top and bottom etching amounts of the silicon nitride film.

An etching method according to one aspect of the present invention is a dry etching method for etching a film structure, which is formed in advance on a wafer placed in a processing chamber, and in which a silicon nitride film is sandwiched between silicon oxide films and whose end portions form side walls of a trench or hole, by supplying a processing gas into the processing chamber without using plasma,
(a) reacting hydrogen fluoride gas at a predetermined temperature to form a reaction layer on the silicon nitride film;
(b) treating the reaction layer formed in the step (a) by flowing a polar solvent gas;
(c) a step of heating at a temperature higher than that in the step (a) in a state where hydrogen fluoride gas is not flowed, thereby volatilizing and removing the reaction layer formed in the step (a) and treated in the step (b),
The steps (a), (b) and (c) are repeated multiple times to etch the silicon nitride film laterally from the end portion.

Another aspect of the present invention provides an etching method, which is a dry etching method for etching a film structure, which is formed in advance on a wafer placed in a processing chamber, and in which a silicon nitride film is sandwiched between silicon oxide films and whose end portions form side walls of a trench or hole, by supplying a processing gas into the processing chamber without using plasma,
(a) reacting hydrogen fluoride gas at a predetermined temperature to form a reaction layer on the silicon nitride film;
(d) a step of heating at a temperature higher than that in the step (a) while flowing a polar solvent gas and without flowing hydrogen fluoride gas, thereby volatilizing and removing the reaction layer formed in the step (a),
The steps (a) and (d) are repeated multiple times to etch the silicon nitride film laterally from the end portion.

The present invention provides a method for etching a silicon nitride film with high selectivity and precision relative to a silicon oxide film without deteriorating the shape of the silicon oxide film that is to be retained, and further reduces the difference in the amount of etching of the silicon nitride film between the top and bottom. Other issues, configurations, and effects will become clear from the description of the embodiments below.

1 is a graph showing etching film thicknesses and selectivities of silicon nitride films and silicon oxide films with respect to methanol flow rates when methanol gas is flowed in a process according to the first embodiment of the present invention. 10 is a graph showing the etching film thickness and selectivity of a silicon nitride film and a silicon oxide film versus time when IR lamp irradiation is performed simultaneously with a methanol gas flow in a process according to a third embodiment of the present invention. 1 is a cross-sectional view showing an outline of an etching apparatus according to a first embodiment of the present invention; FIG. 10 is a cross-sectional view showing an outline of an etching apparatus according to a fourth embodiment of the present invention. 1 is a flow chart of a method for etching a silicon nitride film according to an embodiment of the present invention. 1 is a flow chart of a method for etching a silicon nitride film according to an embodiment of the present invention. 1 is a flow chart of a method for etching a silicon nitride film according to an embodiment of the present invention. 1 is a flow chart of a method for etching a silicon nitride film according to an embodiment of the present invention. 4 is a time chart schematically showing the flow of operations over time in an etching process according to a first embodiment of the present invention. 10 is a time chart schematically showing the flow of operations over time in an etching process according to a second embodiment of the present invention. 10 is a time chart schematically showing the flow of operations over time in an etching process according to a third embodiment of the present invention. 10 is a time chart schematically showing the flow of operations over time in an etching process according to a fourth embodiment of the present invention. 10 is a time chart schematically showing the flow of operations over time in an etching process according to a fifth embodiment of the present invention. 1 is a partial cross-sectional view illustrating the progress of an etching process on a stacked film of a silicon nitride film and a silicon oxide film according to an embodiment of the present invention, showing the state before etching. 1 is a partial cross-sectional view for explaining the progress of an etching process of a stacked film of a silicon nitride film and a silicon oxide film according to an embodiment of the present invention, showing an ideal case of etching. 10A and 10B are partial cross-sectional views illustrating the progress of etching of a stacked film of a silicon nitride film and a silicon oxide film when the selectivity is poor according to an embodiment of the present invention. FIG. 10 is a partial cross-sectional view illustrating the progress of an etching process of a stacked film of a silicon nitride film and a silicon oxide film according to an embodiment of the present invention, in which the selectivity is relatively high, and the amount of etching is smaller at the top portion than at the bottom portion while the corners of the silicon oxide film remain rectangular, and the film thickness of the silicon oxide film portion at the bottom portion is thinner. FIG. 1 is a partial cross-sectional view illustrating the progress of an etching process of a stacked film of a silicon nitride film and a silicon oxide film in an embodiment of the present invention, in which the selectivity is relatively high and the rectangular shape of the silicon oxide film is maintained from the top to the bottom.

The following describes in detail an embodiment of the present invention with reference to the drawings. However, in the following description, identical components may be designated by the same reference numerals and repeated explanations may be omitted. Note that the drawings may be more schematic than the actual embodiment to make the description clearer, but they are merely examples and do not limit the interpretation of the present invention.

The present invention relates to a process technology for an isotropic dry etching method used in the removal of silicon nitride films from semiconductor devices such as 3D memory. The inventors investigated plasma-free dry etching using hydrogen fluoride gas (HF) on samples for calibrating slits in single-layer silicon nitride and silicon oxide films formed by plasma CVD (chemical vapor deposition), as well as in laminated films (described below). The following describes a dry etching method according to the present invention, which etches a pre-formed film structure (see Figures 10 and 11 ) on a wafer placed in a processing chamber, in which silicon nitride films are sandwiched between silicon oxide films on top and bottom, with the edges of the film layers forming the sidewalls of a trench or hole, by supplying processing gases (hydrogen fluoride gas, polar solvent gas) into the processing chamber without using plasma.

[Configuration example 1 of etching processing apparatus]
First, an outline of an etching processing apparatus 100 according to a first embodiment of the present invention will be described, including its overall configuration, with reference to Figure 3. The processing chamber 1 of the etching processing apparatus 100 is composed of a base chamber (also referred to as a chamber) 11, which contains a wafer stage (also referred to as a stage or a mounting table) 3 for placing a wafer 2 as a sample thereon. A shower plate 23 is installed in the center of the upper side of the processing chamber 1, and processing gas is supplied to the processing chamber 1 through the shower plate 23.

The supply flow rate of the process gas is adjusted by mass flow controllers 52 installed in the mass flow box 50 for each gas type. A gas distributor 51 is installed downstream of the mass flow controller 52. When viewed from above, the flow rate and composition of the gas supplied to the center of the process chamber 1 and the gas supplied to the periphery can be independently controlled, enabling precise control of the spatial distribution of the process gas partial pressure. FIG. 3 illustrates argon (Ar), nitrogen (N ₂ ), helium (He), hydrogen fluoride (HF), and methanol (CH ₃ OH), an alcohol, as a polar solvent gas. Other process gases can also be supplied. In particular, polar solvents such as water, ethanol, and isopropanol can be used. However, when using water, because it freezes at temperatures below 0°C, the stage temperature of the wafer stage 3 must be higher than 0°C.

To the bottom of the processing chamber 1, an exhaust means 15 is connected via a vacuum exhaust pipe 16 to reduce the pressure inside the processing chamber 1. The exhaust means (exhaust device) 15 may be composed of, for example, a turbomolecular pump, a mechanical booster pump, or a dry pump. In addition, a pressure adjustment means (pressure adjustment device) 14 is installed upstream of the exhaust means 15 to adjust the pressure inside the processing chamber 1.

An IR lamp unit for heating the wafer 2 is installed above the wafer stage 3. The IR lamp unit can be primarily composed of an IR lamp 60, a reflector 61, and an IR light-transmitting window 72. A circular (circular, annular) lamp is used as the IR lamp 60. Note that the light emitted from the IR lamp 60 is assumed to be primarily light ranging from visible light to infrared light (herein referred to as IR light). In this embodiment, three lamps 60-1, 60-2, and 60-3 are installed as the IR lamps 60, but two or four lamps may also be installed. A reflector 61 is installed above the IR lamp 60 to reflect IR light downward (toward the wafer placement direction). The IR light-transmitting window 72 is preferably made of a material that does not contain alkali metal ions, transmits light in the infrared range, and is heat-resistant. Specifically, quartz is a preferred material.

An IR lamp power supply 73 is connected to the IR lamp 60, and a high-frequency cut filter 74 is installed along the way to prevent high-frequency power noise from entering the IR lamp power supply 73. The IR lamp power supply 73 also has a function that allows the power supplied to the IR lamps 60-1, 60-2, and 60-3 to be controlled independently of each other, allowing the radial distribution of the heating amount on the wafer 2 to be adjusted (some of the wiring is not shown). A space is formed in the center of the IR lamp unit to install a shower plate 23 for introducing process gas.

The wafer stage 3 has a coolant flow path 39 formed inside it to cool the wafer stage 3, and the coolant is circulated and supplied by a chiller 38. In this invention, this chiller is capable of temperature control from -50°C to 50°C, for example. Furthermore, a proximity cooling system was used for the wafer stage 3.

Protrusions 56 are provided on the surface of the wafer stage 3, and the wafer 2 is mounted so that it is supported at the points formed by the protrusions 56. The height of the protrusions 56 is preferably, for example, approximately 0.1 mm to 1.0 mm, and the number of supporting points is preferably three or more. Specifically, six 0.25 mm protrusions 56 were used here. The wafer stage 3 can be made of a corrosion-resistant metal or metal compound with high thermal conductivity.

Because there is a gap (gap, space) between the wafer stage 3 and the wafer 2 due to the protrusions 56, by flowing an inert gas such as He, Ar, or _N2 throughout the chamber 11, the inert gas flows into the gap, causing heat conduction and cooling the wafer 2. Note that the electrostatic adsorption method shown in Example 4 can also be used to cool the wafer 2.

In addition, a thermocouple 70 is installed inside the wafer stage 3 to measure the temperature of the wafer stage 3, and this thermocouple 70 is connected to a thermocouple thermometer 71. The temperature of the wafer stage 3 measured by the thermocouple 70 and the thermocouple thermometer 71 was within ±1°C of the set temperature of the chiller 38.

The proximity-cooled wafer stage 3 described above has the advantage of being simple in structure, allowing for low costs. However, when the chamber 11 is idle and in a vacuum state, the wafer 2 is insulated, so it takes a certain amount of time for inert gas to flow and cooling to begin. Furthermore, because the distance between the refrigerant from the chiller 38 and the wafer 2 is relatively long, it was found that the actual wafer temperature tends to be higher than the set temperature of the chiller 38. When the temperature during cooling and processing was measured using a wafer with a thermocouple attached, it was found that the actual wafer temperature was approximately 5°C higher than the set temperature of the chiller 38.

In addition, as a mechanism for cooling the wafer stage 3 used in the etching processing apparatus 100 of the present invention, in addition to one that circulates a refrigerant, a thermoelectric conversion device such as a Peltier element can also be used.

The etching processing apparatus 100 used in the present invention can heat the inside of the chamber 11, excluding the wafer stage 3 that is exposed to hydrogen fluoride (HF) gas, such as the processing chamber 1. For example, a temperature of approximately 40°C to 120°C can be used. This prevents hydrogen fluoride (HF) gas from being adsorbed inside the chamber 11, minimizing corrosion inside the chamber 11.

In this invention, HF gas at 50 Pa to 1000 Pa is used at a stage temperature of 40°C to -50°C. Depending on the stage temperature of the wafer stage 3, it is thought that the HF gas and the polar solvent used may condense on the silicon nitride film and solidify or liquefy. Therefore, when using an electrostatic chuck, if solidification or liquefaction occurs on the backside of the wafer 2, the seal band for the backside cooling gas may break, causing a cooling gas such as He to leak and resulting in an electrostatic chuck error. In contrast, the proximity-cooled stage 3 has an inherent gap, so even if the HF or polar solvent solidifies or liquefies, no wafer stage 3 errors occur, enabling stable processing.

Furthermore, with the electrostatic chuck method, the gap between the wafer 2 and the stage 3 is narrow, so when the HF or polar solvent used solidifies or liquefies, the wafer 2 tends to stick to the stage 3 due to surface tension. As a result, when dechucking the wafer 2, lifting the wafer 2 with the pusher pins can cause the wafer 2 to crack. To address this issue, we adopted a proximity cooling method with a 0.25 mm gap between the wafer 2 and the stage 3, which has mitigated the problem of the wafer 2 sticking to the stage when the HF liquefies.

When applying processes that use low temperatures, such as the present invention, condensation can form on components that come into contact with the atmosphere inside the electrostatic chuck electrode, which serves as the cooling source, and this can cause short circuits in electrical circuits such as the power supply. In this respect, the structure of the stage 3, which uses proximity cooling and simplifies the internal electrode components, is advantageous.

[Etching process flow 1]
Next, the flow of the proposed plasma-free etching process (etching step) using hydrogen fluoride gas will be described with reference to Figures 3, 5-1, and 7-1. This flow is a dry etching method in which a film structure (see Figures 10 and 11) formed in advance on a wafer 2 placed in a processing chamber 1, in which a silicon nitride film sandwiched between silicon oxide films and whose edges form the sidewalls of a groove or hole, is etched by supplying a processing gas into the processing chamber 1 without using plasma. First, the wafer 2 is transferred into the processing chamber 1 through a transfer port (not shown) provided in the processing chamber 1, and then the wafer 2 is placed on the protrusion 56 on the wafer stage 3.

Thereafter, wafer cooling in step S101 of FIG. 5-1 is performed by supplying Ar gas for wafer cooling to the wafer 2 via the mass flow controller 52, the gas distributor 51, and the shower plate 23. Because Ar gas serves as both a heat transfer gas to the wafer 2 and a dilution gas for diluting the HF gas, steps S101 and S102 of FIG. 5-1 are performed simultaneously. The flow rate of Ar gas can be changed between when cooling the wafer and when used as a dilution gas. The flow of Ar gas for dilution can be continued or stopped until the etching process is completed. Instead of Ar gas, _N2 gas can be used as an inert gas.

Next, in step S103 of Figure 5-1, HF gas was supplied to the process chamber 1 at a predetermined amount for a predetermined time, and simultaneously, heating was performed to form a reaction layer. Here, heating was performed using IR (infrared) lamps 60. The wafer temperature obtained as a result of cooling by the stage 3 and heating by the IR lamps 60 is preferably 30°C to 55°C, and more preferably 35°C to 50°C. The reaction layer thickness can be controlled by the total pressure or HF partial pressure, heating temperature, IR lamp output, time, number of repetitions, etc. Furthermore, if the wafer temperature is below 30°C, etching is difficult because an insufficient reaction layer is formed. Conversely, if the wafer temperature is above 55°C, an excessive reaction layer is formed, and when it is decomposed and volatilized, it tends to etch the adjacent silicon oxide film, which is undesirable, resulting in reduced selectivity.

In the present invention, the pressure used is preferably approximately 10 to 1000 Pa, more preferably 50 to 1000 Pa, and especially 100 to 1000 Pa. The higher the pressure, the easier it is to form a reaction layer on the silicon nitride film, and the lower the temperature required for formation. Even when the pressure is high, by controlling the output of the IR lamp 60, it is possible to form a reaction layer on the silicon nitride film without affecting the silicon oxide film.

After the reaction layer has been formed for a predetermined time, a process (not shown in Figure 5-1) is performed in which the supply of HF gas is stopped and the HF gas remaining in the gas phase is evacuated. When evacuating to a vacuum, it is desirable to set the pressure to 5 Pa or less. In other words, here, the process of evacuating HF gas while flowing an inert gas is inserted between S103 and the subsequent S104.

Next, in S104 of Figure 5-1, a polar solvent gas is flowed to modify the reaction layer and remove a portion of it. Here, the polar solvent gas can be an alcohol such as methanol, ethanol, or isopropanol, or water. The flow rate of the polar solvent gas is preferably 0.1 to 3.0 L/min, and the flow time is preferably approximately 30 to 300 seconds. The pressure at this time is also preferably approximately 10 to 1000 Pa. When flowing the polar solvent gas, an inert gas such as Ar or nitrogen gas can also be flowed simultaneously as a dilution gas.

After the reaction layer has been formed for a predetermined time, the supply of polar solvent gas is stopped and the polar solvent gas remaining in the gas phase is evacuated, although this is not shown in Figure 5-1. If evacuating to a vacuum, it is desirable to set the pressure to 5 Pa or less. In other words, here, a process of evacuating the polar solvent gas while flowing an inert gas is inserted between S104 and the subsequent S105.

Next, heating is performed without flowing HF gas or polar solvent gas to remove the reaction layer (S105 in Figure 5-1). The heating temperature here is preferably 70°C to 110°C (70°C or higher and 110°C or lower), and more preferably 70°C to 100°C. If the heating temperature is lower than 70°C, the reaction layer will not be sufficiently removed, and etching will not proceed. Furthermore, if a heating temperature higher than 110°C is used, undesired etching of adjacent silicon oxide films is likely to occur when the reaction layer decomposes and volatilizes, reducing selectivity and tending to deteriorate the vapor shape of the oxide film that should be left.

Here, an IR lamp 60 was used as the heating method for removal. Using an IR lamp 60 has the advantage of being able to heat the wafer in a short time. However, the heating method is not limited to this. For example, the wafer stage 3 may be heated, or the wafer 2 may be transported to a separate heating-only device and heated. Ar gas or nitrogen gas may be introduced during irradiation by the IR lamp 60. Heating may also be performed multiple times as needed. After heating, the process returns to S101 to cool the wafer. Steps S101 through S105 constitute one cycle, and this cycle is repeated N times. The cycle is repeated multiple times until the required etching depth is achieved, and then the process ends. Here, steps S103 and S105 are performed by heating the stage to a low temperature of -50°C or higher and 0°C or lower, followed by lamp heating. This allows the stage to reach a temperature of 30°C or higher and 55°C or lower in step S103, and then a temperature of 70°C or higher and 110°C or lower in step S105.

Figure 7-1 shows a time chart for the flow shown in Figure 5-1 when using the etching processing apparatus 100 shown in Figure 3. One cycle includes a process of IR lamp heating while flowing HF gas, a process of flowing polar solvent gas, and a process of IR lamp heating without flowing HF gas or polar solvent gas. By repeating this process N times, the silicon nitride film is etched.

[Etching result 1]
The results of etching using hydrogen fluoride (HF) gas without using plasma according to the present invention are shown below. The temperature of stage 3 was set to -30°C, and the etching rates of single-layer silicon nitride (PE-SiN) and silicon oxide (PE-SiO2) films formed by plasma CVD were measured.

Here, the base wafer 2 was a high-resistivity substrate (31 Ωcm) with a diameter of 300 mm, to which 2 cm square coupon samples of silicon nitride film and silicon oxide film were attached with vacuum grease.

After placing the wafer 2 in the etching processing device 100 shown in Figure 3, etching was performed using the process flow shown in Figure 5-1. First, to cool the wafer, Ar was flowed at a flow rate of 1.4 L/min and a pressure of 900 Pa for 60 seconds. After that, while maintaining the pressure at 900 Pa, HF was introduced at 0.40 L/min and Ar as a dilution gas at 0.20 L/min, while the IR lamp 60 was simultaneously irradiated at 60% output for 60 seconds. This resulted in the formation of a reaction layer on the silicon nitride film.

Then, the exhaust valve (not shown) was opened 100% and the mixture was evacuated for 120 seconds. This exhaust operation evacuated fluorine gas and some of the reaction products.

Next, while the set temperature of stage 3 remained the same, methanol was flowed as a polar solvent gas at a pressure of 300 Pa for 60 seconds. By flowing methanol gas, the reaction layer was modified and partially removed. Here, the flow rates of methanol gas were set to 0.40, 0.80, and 1.2 L/min. For comparison, a flow without this polar solvent gas was also performed.

Next, with the set temperature of stage 3 unchanged, Ar was flowing at 0.50 L/min, and the exhaust valve was fully open, heating was performed for 40 seconds at 70% lamp intensity. This process removes the reaction layer modified by the polar solvent gas flow. After that, the process returned to the beginning and cooled with Ar flowing at 1.4 L/min for 60 seconds at a pressure of 900 Pa. This series of processes was repeated 10 times, following the flow shown in Figure 5-1.

Table 1-1 shows the results of actual measurements of the process temperature during lamp irradiation using a high-resistance substrate equipped with a thermocouple, replacing HF gas with Ar. The temperature measured here is the temperature reached when the IR lamp 60 used in the previous process was irradiated at 60% output for 60 seconds, resulting in a temperature of 50°C. Furthermore, when the IR lamp used in the heating process to remove the reaction layer was irradiated at 70% output for 40 seconds, the temperature measured was found to be 80°C.

The etching thickness of the single-layer silicon nitride film (PE-SiN) and silicon oxide film (PE- _SiO2 : PE-SiO2 in the table) obtained after 10 cycles when the flow rate of methanol gas was changed, and the selectivity of the silicon nitride film to the silicon oxide film are shown in Figure 1. Figure 1 also shows the results when there was no methanol gas flow, with the methanol gas flow rate set to 0 L/min.

By using a methanol gas flow, the etching thickness of the silicon nitride monolayer film decreased, and the selectivity also decreased. However, increasing the flow rate of the methanol gas flow increased the etching thickness of the silicon nitride film, and at 1.2 L/min, the etching rate was similar to that without methanol gas flow. The selectivity also improved slightly. The effect of the methanol gas flow is evident in that it contributes to some removal of the reaction layer.

The film structure targeted by this invention is the structure required for 3D-NAND, as shown in Figure 10, in which multiple silicon nitride films 103 and silicon oxide films 102 are alternately stacked on a substrate 101, with deep hole or trench-shaped openings 104 formed therein. The thickness of the silicon nitride film 103 used here is several nanometers to 100 nanometers, and the thickness of the silicon oxide film 102 is several nanometers to 100 nanometers. Furthermore, these layers are stacked in tens to hundreds of layers. The total thickness 105 of these stacks is several micrometers to tens of micrometers. The width 104W of the opening 104 is several tens to hundreds of nanometers. Using the process of this invention, the silicon nitride film 103 is etched laterally with high selectivity relative to the silicon oxide film 102, as shown in the ideal cross-sectional view of Figure 11. The dimension 106 of this lateral etching is several nanometers to tens of nanometers.

When etching the silicon nitride film 103 laterally, it is desirable for the selectivity ratio to the silicon oxide film 102 to be 10 or greater, and more desirably 20 or greater. If this selectivity ratio is low, etching of the silicon oxide film 102, which should not be etched, occurs simultaneously, and the shape of the edge of the silicon oxide film 102 after etching becomes rounded rather than rectangular, as shown by 111 in Figure 12, which adversely affects device performance.

Empirically, when the selectivity is 10 or more, and more preferably 20 or more, a shape closer to a rectangle, as shown in Figure 11, is obtained. Furthermore, when the selectivity is less than 5, the edge shape of the silicon nitride film 103 becomes rounded, as shown by 111 in Figure 12, which is undesirable.

As an example of the present invention, a sample was formed with a total of 40 layers of silicon nitride film 103 (30 nm thick) and silicon oxide film 102 (30 nm thick) alternately deposited, and a 200 nm slit-shaped space was formed in the sample to evaluate the etching characteristics in a fine pattern. Ten cycles of etching were performed using the experimental conditions discussed in Figure 1. The results are shown in Table 1-2.

As a result, even when etching proceeds at a relatively high selectivity and the silicon oxide film 102 to be left is shaped like a rectangle, as shown in Figure 13, the silicon nitride film 103 is efficiently etched at the bottom, as shown in 112, but the thickness of the silicon oxide film 102 is reduced. Furthermore, at the top, as shown in 111, there is no tendency for the thickness of the silicon oxide film 102 to be reduced, but the amount of etching at the top 106 tends to be smaller than the amount of etching at the bottom 109.

Table 1-2 shows the etching amount (the etching amount of the silicon nitride film 103 minus the etching amount of the silicon oxide film 102), the selectivity ratio from the slit pattern results (the etching amount from the initial dimensions of the silicon nitride film 103 divided by the etching amount of the silicon oxide film 102), and the remaining _SiO2 thickness (the thickness 108 of the tip 112 of the silicon oxide film 102 after etching shown in FIG. 12 divided by the initial thickness 107 of the silicon oxide film 102). Here, good etching conditions are those in which the difference in etching amount between the top and bottom is small, the selectivity ratio is large, and the remaining _SiO2 thickness is close to 1.

To make the evaluation results easier to understand, symbols such as ◎, 〇, △, and × are also listed in Table 1-2. The criteria are shown in Table 1-3.

As shown in Table 1-2, the etching amount of the silicon nitride film 103 in the slit sample also increased as the flow rate of the methanol gas flow increased. When the methanol gas flow was 0 L/min, that is, when the methanol gas flow process was not applied, the selectivity was relatively good, but the remaining _SiO2 thickness at the bottom was thin at 0.70, which was found to be insufficient. It was also found that the difference in etching amount between the top and bottom was large at -8.9 nm. The shape of the slit sample in this case was as shown in Figure 13.

When no methanol gas flow was used, the reaction products increased the etching amount at the bottom and decreased the etching amount at the top. In contrast, when the methanol gas flow process was applied, the residual _SiO2 thickness at the bottom improved to 0.90 or more, and the difference in etching amount between the top and bottom was reduced to 2.0 nm or less, resulting in good etching results as shown in Figure 14.

Next, under the experimental conditions that yielded the results in Table 1-2, an experiment was conducted in which the methanol gas flow was fixed at 0.4 L/min and 300 Pa, and the time was changed to 60 s, 120 s, and 180 s. The experimental results for the slit sample at that time are shown in Table 1-4.

As shown in Table 1-4, the etching amount of the silicon nitride film 103 in the slit sample increased as the methanol gas flow time increased, and it was found that the etching amount was saturated at 120 seconds or more. Even when the methanol gas flow time was increased, the residual _SiO2 thickness at the bottom was 0.90 or more, and the difference in etching amount between the top and bottom was small, at 2.0 nm or less, indicating good etching results as shown in FIG.

[Effect of reaction layer composition and methanol gas flow]
To confirm the effect of the polar solvent gas flow of the present invention, a study was conducted to analyze the reaction layer. For a single layer of silicon nitride film 103, an experiment was conducted under the conditions shown in Table 1-2 of Example 1, with and without a methanol gas flow rate of 0.40 L/min, in which the reaction layer was not removed by IR heating but was instead intentionally left intact.

Using the etching processing apparatus 100 shown in Figure 3, with a stage temperature of -30°C and a single-layer silicon nitride film, Ar was first flowed at a flow rate of 1.4 L/min at 900 Pa for 60 seconds to cool the wafer. Then, while maintaining the pressure at 900 Pa, HF was introduced at 0.40 L/min and Ar as a dilution gas at 0.20 L/min, while the IR lamp 60 was simultaneously irradiated at 60% output for 60 seconds. This resulted in the formation of a reaction layer on the silicon nitride film 103. The exhaust valve was then left fully open for 120 seconds, allowing the chamber to be evacuated. This exhaust operation evacuates fluorine gas and some of the reaction products.

Next, while the set temperature of stage 3 remained the same, methanol gas was passed as a polar solvent gas at a flow rate of 0.40 L/min at a pressure of 300 Pa for 120 seconds. By passing methanol gas, the reaction layer was modified and partially removed.

Then, without removing the reaction layer by heating, the wafer was cooled and this series of processes was repeated eight times. Regarding the flow, steps S101 to S104 were repeated eight times, without removing the reaction layer by heating (S105) as shown in Figure 5-1. The reaction layer formed by this process is called reaction layer B (with methanol gas flow).

For comparison, a reaction layer was also formed without the flow of methanol gas as a polar solvent gas and without the removal of the reaction layer by heating. The flow shown in Figure 5-1 was repeated eight times from S101 to S103, omitting the modification and removal of the reaction layer by polar solvent (S104) and the removal of the reaction layer by heating (S105). The reaction layer formed by this process is referred to as reaction layer A (without methanol gas flow).

These two reaction layers were subjected to cross-sectional and surface observation using a scanning electron microscope (SEM), composition analysis using X-ray photoelectron spectroscopy (XPS), and desorption gas analysis using thermal desorption spectroscopy (TDS).

SEM observation of the cross section and surface of reaction layer A (without methanol gas flow) revealed that the cross-sectional reaction layer thickness was 61 nm, with unreacted silicon nitride film 103 present below, measuring 281 nm. In contrast, reaction layer B (with methanol gas flow) showed a cross-sectional reaction layer thickness of 65 nm, with the unreacted silicon nitride film 103 present below that measuring 285 nm. While the reaction layer thickness was the same with and without methanol gas flow, reaction layer B (with methanol gas flow) exhibited greater surface roughness than reaction layer A (without methanol gas flow), with partial aggregation of the reaction layer and missing portions. These findings demonstrate that methanol gas flow is effective in partially removing the reaction layer.

Next, compositional analysis was performed using X-ray photoelectron spectroscopy (XPS) on reaction layer A (without methanol gas flow) and reaction layer B (with methanol gas flow). The compositional analysis compared the outermost surface (a) in its original state, the interior after 10 seconds of Ar sputtering (b), and the interior after an additional 10 seconds of Ar sputtering (c).

The outermost surface (a) of reaction layer A (without methanol gas flow) showed only a 401 eV peak for nitrogen (N1s), with no 395 eV peak for silicon nitride. This 401 eV peak was determined to be attributable to ammonium salt. In the interior (b) sputtered with Ar for 10 seconds, a 395 eV peak for silicon nitride appeared, with the ratio of the 401 eV peak for ammonium salt to the 395 eV peak for silicon nitride being approximately 1:1. In the interior (c) sputtered with Ar for a further 10 seconds, the 401 eV peak for ammonium salt was not observed, with only the 395 eV peak for silicon nitride.

With regard to silicon (Si2P), in the outermost surface (a) of reaction layer A (without methanol gas flow), peaks at 99 eV for silicon nitride and 102 eV for silicate, i.e., peaks attributable to hexafluorosilicate _SiF62- ^, were observed in a ratio of approximately 1:1. In the interior (b) sputtered with Ar for 10 seconds, the 99 eV peak for silicon nitride and the 102 eV peak for silicate were also observed in a ratio of approximately 1:1. In the interior (c) sputtered with Ar for a further 10 seconds, the 102 eV peak for silicate disappeared, and only the 99 eV peak for silicon nitride remained.

In contrast, the outermost surface (a) of reaction layer B (with methanol gas flow) exhibits a 401 eV ammonium salt peak for nitrogen (N1s) and a 395 eV silicon nitride peak in a ratio of approximately 2:1. Since the underlying silicon nitride film is visible, it is believed to be thinner than reaction layer A. This 401 eV peak was attributed to ammonium salt. In the interior (b) sputtered with Ar for 10 seconds, a significant 395 eV silicon nitride peak appears, with the 401 eV ammonium salt peak and the 395 eV silicon nitride peak being approximately 1:4. Furthermore, in the interior (c) sputtered with Ar for 10 seconds, the 401 eV ammonium salt peak was not observed, and only the 395 eV silicon nitride peak was present.

With regard to silicon (Si2P), in the outermost surface (a) of reaction layer B (with methanol gas flow), peaks at 99 eV silicon nitride and 102 eV silicate, i.e., peaks attributable to hexafluorosilicate _SiF62- ^, were observed in a ratio of approximately 1:1. In the interior (b) sputtered with Ar for 10 seconds, silicon nitride at 99 eV and silicate at 102 eV were observed in a ratio of approximately 1:4, and this result also showed that reaction layer B was thinner than reaction layer A. In the interior (c) sputtered with Ar for a further 10 seconds, the silicate at 102 eV disappeared, and only the 99 eV silicon nitride peak remained.

From the above, analysis of the reaction layer using XPS showed that no significant differences in composition were observed from the peaks of reaction layer A (without methanol gas flow) and reaction layer B (with methanol gas flow), but it was confirmed that the thickness of the reaction layer was clearly thinner when methanol gas flow was present.

In the case of ammonium silicofluoride [( _NH4 ) _2SiF6 ], the elemental ratios are Si = 1, F = 6, _and N = 2. Here, the elemental ratios of the outermost surface (a) of reaction layer A (without methanol gas flow) measured by XPS were Si = 1, F = 1.25, and N = 0.58, and the elemental ratios of the outermost surface (a) of reaction layer B (with methanol gas flow) measured by XPS were Si = 1, F = 1.46, and N = 0.88, indicating that both were deficient in fluorine and nitrogen. It was found that reaction layer B (with methanol gas flow) had slightly more fluorine and nitrogen.

The reaction layer is formed when ammonium fluorosilicate [(NH ₄ ) ₂ SiF ₆ ] and similar compounds decompose and volatilize, producing HF and NH ₃ , which, depending on the conditions, etch the adjacent silicon oxide film, which is thought to be related to the top-bottom difference and the narrowing of the bottom oxide film. The partial removal of the reaction layer by the methanol gas flow is thought to be the reason for the improvement in shape.

Next, desorbed gas analysis was performed on reaction layer A (without methanol gas flow) and reaction layer B (with methanol gas flow) by thermal desorption spectroscopy (TDS). In TDS analysis, a 1 cm square sample on which a reaction layer was formed was heated from a vacuum state of 5.0 × ¹⁰ Pa in the temperature range of 80 to 500 °C at a rate of 10 °C/min, and the pressure increase and desorbed gas were analyzed using a mass spectrometer (Q-mass).

In reaction layer A (no methanol gas flow), the pressure began to rise around 150°C, with one peak, and the temperature at which the maximum pressure was reached was 182°C. The masses at this time included a strong F ⁺ peak at m/z = 19, followed by _NH2 ⁺ and _NH3 ⁺ at m/z = 16 and 17 at approximately 1/50th the intensity, and then SiF+ at m/z = 47 at approximately 1/5 the intensity. Furthermore, no peaks attributable to methanol were found. The TDS analysis results also suggested that the reaction layer contained ammonium silicofluoride.

In reaction layer B (with methanol gas flow), two pressure peaks were observed, with the temperatures at which they reached their maximums being 176°C and 208°C. This indicates that with methanol gas flow, components with high decomposition and volatilization temperatures are generated. The mass results were similar to those of reaction layer A (without methanol gas flow), except for the fact that there were two peaks. The mass peaks were F ⁺ at m/z = 19, with a large intensity, followed by _NH2 ⁺ and _NH3 ⁺ at m/z = 16 and 17 at approximately 1/50th the intensity, and then SiF+ at m/z = 47 at approximately 1/5th the intensity. Furthermore, no peaks attributable to methanol were observed, and the results appear to be unchanged from the case without methanol gas flow. While the exact reason for the two degassing peaks is unknown, methanol gas flow resulted in the generation of components with high decomposition and volatilization temperatures. Considering this together with the previous XPS results, it is estimated that the excess HF is removed, resulting in a composition close to ammonium silicofluoride. However, the peak on the high-temperature side shifted by the methanol gas flow does not appear to have changed in its constituent mass, so the details are currently unknown.

Comparing the peak areas between 150°C and 250°C in pressure, it was found that in reaction layer A (without methanol gas flow) it was 4.72e-14 (Pa·°C), while in reaction layer B (with methanol gas flow) it was 4.15e-14 (Pa·°C), a reduction of approximately 15%. This also shows that the methanol gas flow removes part of the reaction layer.

(Comparative Example)
In order to confirm the effect of the polar solvent gas flow of the present invention in more detail, an experiment was conducted under the condition of a methanol gas flow rate of 0.40 L/min shown in Table 1-2 of Example 1, without removing the reaction product by IR heating.

Using the etching processing apparatus 100 shown in Figure 3, with a stage temperature of -30°C, Ar was first flowed at a flow rate of 1.4 L/min at 900 Pa for 60 seconds to cool the wafer. Then, while maintaining the pressure at 900 Pa, HF was introduced at 0.40 L/min and Ar as a dilution gas at 0.20 L/min, while the IR lamp 60 was simultaneously irradiated at 60% output for 60 seconds. This resulted in the formation of a reaction layer on the silicon nitride film 103. The exhaust valve was then left fully open for 120 seconds, allowing the chamber to be evacuated. This exhaust operation evacuates fluorine gas and some of the reaction products.

Next, while the stage temperature was maintained, methanol gas was passed as a polar solvent gas at a flow rate of 0.40 L/min at a pressure of 300 Pa for 60 seconds. By passing methanol gas, the reaction layer was modified and partially removed.

Then, without removing the reaction layer by heating, the wafer was cooled and this series of processes was repeated 10 times. Regarding the flow, steps S101 to S104 were repeated without removing the reaction layer by heating (S105) as shown in Figure 5-1.

Using the above etching process, the etching characteristics of fine patterns were evaluated using a sample in which a total of 40 layers of silicon nitride film 103 (thickness 30 nm) and silicon oxide film 102 (thickness 30 nm) were alternately formed, as used in Example 1, and a 200 nm slit-shaped space was formed.

As a result, etching of the silicon nitride portion was beginning to occur, but it was found that a large amount of deposits existed in that area. As shown in the reaction layer composition and the effect of methanol gas flow in Example 1, this deposit is believed to be a compound similar to ammonium silicofluoride, and scanning electron microscope (SEM) observation revealed that it volatilized when irradiated with an electron beam. Table 1-5 shows the amount of etching of the silicon nitride film when the deposits believed to be ammonium silicofluoride were ignored, and it was found to be smaller than the amount obtained when the reaction was removed by IR heating as shown in Table 1-2. From the above, it was found that the silicon nitride film 103 cannot be sufficiently etched by simply modifying and removing the reaction layer with polar solvent gas, and that a process of removing the reaction layer by heating is necessary.

Another embodiment of the present invention's etching process using hydrogen fluoride gas without plasma will be described. Here, the flow shown in Figures 5-2 and 7-2 will be explained using the etching processing apparatus 100 shown in Figure 3 in Example 1.

[Etching process flow 2]
First, the wafer 2 is transferred into the processing chamber 1 through a transfer port (not shown) provided in the processing chamber 1, and then the wafer 2 is placed on the protrusion 56 on the wafer stage 3. In this case, the stage temperature was set to 40°C.

Then, Ar gas for thermal conduction to the wafer 2 is supplied via the mass flow controller 52, gas distributor 51, and shower plate 23, thereby heating the wafer at the stage temperature in S101 of Figure 5-2. Here, S101 is described as heating/cooling because it means that the wafer temperature is adjusted to the stage temperature. After S105, described below, the wafer temperature becomes higher than the stage temperature, so cooling occurs.

Since Ar gas serves to conduct heat to the wafer 2 and as a dilution gas for diluting the HF gas, steps S101 and S102 in FIG. 5-2 are performed simultaneously. The flow rate of Ar gas can be changed when conducting heat to the wafer 2 and when used as a dilution gas. Furthermore, the dilution Ar gas can be either continued to flow or not until the etching process is completed. Furthermore, _N2 gas can be used as an inert gas instead of Ar gas.

Next, in S103 of Figure 5-2, a predetermined amount of HF gas was supplied to the processing chamber 1 for a predetermined time to form a reaction layer. Here, heating by the IR (infrared) lamps 60 as shown in the flow of Figure 5-2 was not used, but only the temperature transferred by the heat of the stage 3 was used. Here, a stage temperature of 40°C was used, but the stage temperature, i.e., the wafer temperature, is preferably between 30°C and 55°C, and more preferably between 35°C and 50°C. The film thickness of the reaction layer can be controlled by the stage temperature, total pressure or HF partial pressure, time, number of repetitions, etc.

In the present invention, the pressure used is preferably approximately 10 Pa to 1000 Pa, and particularly preferably 300 Pa to 1000 Pa. The higher the pressure, the easier it is to form a reaction layer on the silicon nitride film 103, and the lower the temperature required for formation. After the reaction layer has been formed for a predetermined time, the supply of HF gas is stopped and the HF gas remaining in the gas phase is evacuated, although this is not shown in Figure 5-2. When evacuating to a vacuum, it is desirable to keep the pressure below 5 Pa. In other words, here, a process of evacuating HF gas while flowing an inert gas is inserted between S103 and the subsequent S104.

Next, in S104 of Figure 5-2, a polar solvent gas is flowed to modify the reaction layer and remove a portion of it. Here, the polar solvent gas can be an alcohol such as methanol, ethanol, or isopropanol, or water. The flow rate of the polar solvent gas is preferably 0.1 to 3.0 L/min, and the flow time is preferably approximately 30 to 300 seconds. The pressure at this time is also preferably approximately 10 to 1000 Pa. When flowing the polar solvent gas, an inert gas such as Ar or nitrogen gas can also be flowed simultaneously as a dilution gas.

After the reaction layer has been formed for a predetermined time, the supply of polar solvent gas is stopped and the polar solvent gas remaining in the gas phase is evacuated, although this is not shown in Figure 5-2. When evacuating to a vacuum, it is desirable to set the pressure to 5 Pa or less. In other words, here, a process of evacuating the polar solvent gas while flowing an inert gas is inserted between S104 and the subsequent S105.

Next, heating is performed without flowing HF gas or polar solvent gas to remove the reaction layer (S105 in Figure 5-2). The heating temperature here is preferably 70°C to 110°C (70°C or higher and 110°C or lower), more preferably 70°C to 100°C. Here, an IR lamp 60 was used as the heating method. The heating method is not limited to this; for example, heating the wafer stage 3 or transporting the wafer 2 separately to a heating-only device for heating treatment may also be used. Ar gas or nitrogen gas can also be introduced during IR lamp irradiation. Heating treatment can also be performed multiple times as needed. After heating, the process returns to S101, where the wafer is cooled by the stage. Steps S101 to S105 constitute one cycle, and this is repeated N times. The cycle is repeated until the required etching depth is achieved, and then the process ends. Here, in steps S103 and S105, the stage is set to a low temperature of between -50°C and 0°C and then heated with a lamp, thereby achieving a stage temperature of between 30°C and 55°C in step S103, and further achieving a stage temperature of between 70°C and 110°C in step S105.

Figure 7-2 shows a time chart for the flow shown in Figure 5-2. One cycle includes a process of flowing HF gas and Ar (the process of forming a reaction layer), a process of flowing polar solvent gas, and a process of IR lamp heating without flowing HF gas. By repeating this process N times, the silicon nitride film is etched.

[Etching result 2]
Using the etching processing apparatus 100 of FIG. 3 and etching process flow 2 (FIGS. 5-2 and 7-2), a process was investigated in which the stage temperature was set to 40° C. and IR heating was not performed in the HF/Ar flow step. First, to conduct heat to the wafer 2, Ar was flowed at a flow rate of 1.4 L/min and 900 Pa for 60 seconds. Then, while controlling the pressure at 900 Pa, HF was introduced at 0.40 L/min and Ar as a diluent gas at 0.20 L/min for 60 seconds. As a result, a reaction layer was formed on the silicon nitride film 103.

Then, with the exhaust valve fully open, the reactor was evacuated for 120 seconds. This exhaust operation evacuated fluorine gas and some of the reaction products. Next, while the set temperature of stage 3 remained the same, methanol gas was passed as the polar solvent gas at a flow rate of 0.40 L/min at a pressure of 300 Pa for 120 seconds. By passing methanol gas, the reaction layer was modified and partially removed.

Next, with the set temperature of stage 3 unchanged, Ar was flowing at 0.50 L/min, and the exhaust valve was fully open, heating was performed for 40 seconds at 70% lamp intensity. This process removes the reaction layer modified by the polar solvent gas flow. After that, the process returned to the beginning and cooled with Ar flowing at 1.4 L/min for 60 seconds at a pressure of 900 Pa. This series of processes was repeated 10 times, following the flow shown in Figure 5-2.

As in Example 1, a sample was formed with a total of 40 layers of silicon nitride film 103 (30 nm thick) and silicon oxide film 102 (30 nm thick) alternately deposited, and a sample with 200 nm slit-shaped spaces formed was used to evaluate the etching characteristics in a fine pattern. The results are shown in Table 2.

Even when IR heating is not performed in the HF/Ar flow step, as in this example, the methanol gas flow is effective, and as shown in Table 2, the residual _SiO2 thickness at both the top and bottom is 0.90 or more, and the difference in etching amount between the top and bottom is small, at 2.0 nm or less, and it was found that good etching results were obtained as shown in FIG. 14.

Another embodiment of the present invention's etching process using hydrogen fluoride gas without plasma will be described. Here, the flow shown in Figures 6-2 and 8 will be explained using the etching processing apparatus 100 shown in Figure 3 in Example 1.

[Etching process flow 3]
First, the wafer 2 is transferred into the processing chamber 1 through a transfer port (not shown) provided in the processing chamber 1, and then the wafer 2 is placed on the protrusion 56 on the wafer stage 3. In this case, the stage temperature was set to 30°C.

Then, Ar gas for thermal conduction to the wafer 2 is supplied via the mass flow controller 52, gas distributor 51, and shower plate 23, thereby heating/cooling the wafer by stage 3 in S111 of Figure 6-2. The term "heating/cooling" here refers to matching the wafer temperature to the stage temperature; after S114, described below, the wafer temperature becomes higher than the stage temperature, so cooling occurs.

Since Ar gas serves both as a heat transfer gas to the wafer and as a dilution gas for diluting the HF gas, steps S111 and S112 in FIG. 6-2 are performed simultaneously. The flow rate of Ar gas can be changed when conducting heat to the wafer 2 and when used as a dilution gas. Furthermore, the dilution Ar gas can be either continued to flow or not until the etching process is completed. Furthermore, _N2 gas can be used as an inert gas instead of Ar gas.

Next, in S113 of Figure 6-2, a predetermined amount of HF gas was supplied to the processing chamber 1 for a predetermined time to form a reaction layer. Here, heating by the IR (infrared) lamps 60 as shown in the flow of Figure 6-2 was not used, but only the temperature transferred by the heat of the stage 3 was used. Here, a stage temperature of 30°C was used, but the stage temperature, i.e., the wafer temperature, is preferably between 30°C and 55°C, and more preferably between 35°C and 50°C. The film thickness of the reaction layer can be controlled by the stage temperature, total pressure or HF partial pressure, time, number of repetitions, etc.

In the present invention, the pressure used is preferably approximately 10 Pa to 1000 Pa, and particularly preferably 300 Pa to 1000 Pa. The higher the pressure, the easier it is to form a reaction layer on the silicon nitride film 103, and the lower the temperature required for formation. After the reaction layer has been formed for a predetermined time, the supply of HF gas is stopped and the HF gas remaining in the gas phase is evacuated, although this is not shown in Figure 6-2. When evacuating to a vacuum, it is desirable to keep the pressure below 5 Pa. In other words, here, a process of evacuating HF gas while flowing an inert gas is inserted between S113 and the subsequent S114.

Next, in step S114 of FIG. 6-2, the reaction layer is removed while being modified by heating while introducing a polar solvent gas. The polar solvent gas can be water or alcohols such as methanol, ethanol, or isopropanol. The flow rate of the polar solvent gas is preferably 0.1 to 3.0 L/min. The heating temperature is preferably 70 to 110°C, more preferably 70 to 100°C. Here, an IR lamp 60 was used as the heating method. The heating method is not limited to this; for example, the wafer stage 3 can be heated, or the wafer 2 can be transported separately to a heating-only device and subjected to heating treatment. Furthermore, inert gases such as Ar gas or nitrogen gas can be introduced in the process of removing the reaction layer while being modified by heating while introducing a polar solvent gas. Furthermore, this process can be performed multiple times as needed.

Next, although not shown in Figure 6-2, the heating and supply of polar solvent gas are stopped, and the polar solvent gas remaining in the gas phase is evacuated. When evacuating to a vacuum, it is desirable to set the pressure to 5 Pa or less. In other words, here, a process of evacuating the polar solvent gas while flowing an inert gas is inserted after S114.

After this, the process returns to wafer heating/cooling by stage 3 in S111. S111 is a process for matching the stage temperature with the wafer temperature, and since wafer 2 has been heated in S114, the next step is cooling in S111. After this, the steps from S111 to S114 constitute one cycle, and this is repeated N times. The cycle is repeated until the required etching amount is achieved, and then the process ends. In steps S113 and S114, the stage is cooled to a low temperature of between -50°C and 0°C, and then lamp heating is performed, thereby achieving a stage temperature of between 30°C and 55°C in step S113, and then a stage temperature of between 70°C and 110°C in step S114.

Figure 8 shows a time chart for the flow shown in Figure 6-2. One cycle includes a process of flowing HF gas and Ar (the process of forming a reaction layer), a process of flowing polar solvent gas, and a process of IR lamp heating without flowing HF gas. By repeating this process N times, the silicon nitride film is etched.

[Etching result 3]
Using the etching processing apparatus 100 of FIG. 3 and etching process flow 3 (FIGS. 6-2 and 8) shown in Example 1, a process was investigated in which the stage temperature was set to 30°C and IR heating was not performed in the HF/Ar flow step. First, to conduct heat to the wafer 2, Ar was flowed at a flow rate of 1.4 L/min and 900 Pa for 60 seconds. Then, while controlling the pressure at 900 Pa, HF was introduced at 0.40 L/min and Ar as a diluent gas at 0.20 L/min for 60 seconds. As a result, a reaction layer was formed on the silicon nitride film 103.

Then, with the exhaust valve fully open, the wafer was evacuated for 120 seconds. This exhaust operation evacuates fluorine gas and some of the reaction products. Next, while the set temperature of stage 3 remained the same, the pressure was controlled to 300 Pa with methanol gas flowing at 0.40 L/min, and heating was performed for 30 to 50 seconds with the IR lamp output at 70%. This removes the reaction layer as the methanol gas flows. After that, returning to the beginning, wafer 2 was cooled with Ar flowing at 1.4 L/min at a pressure of 900 Pa for 60 seconds, until it reached the same temperature as the stage temperature. This series of processes was performed in 10 cycles according to the flow in Figure 6-2.

The etching thickness of the single layer silicon nitride film (PE-SiN) 103 and silicon oxide film (PE-SiO ₂ in FIG. 2 ) 102 obtained after 10 cycles when the time of IR lamp irradiation simultaneously with the methanol gas flow was changed, and the selectivity ratio of the silicon nitride film 103 to the silicon oxide film 102 are shown in FIG. 2 .

As shown in Figure 2, etching of the silicon nitride film 103 also occurred in this process, and it was found that the etching amount of the silicon nitride film 103 saturated when the IR lamp irradiation time was approximately 40 seconds with the methanol gas flow. The etching amount of the silicon oxide film 102 hardly changed, but the selectivity ratio for the single layer film was higher when the IR lamp irradiation time with the methanol gas flow was shorter.

Here, similar to Example 1, a sample was formed with a total of 40 layers of silicon nitride film 103 (30 nm thick) and silicon oxide film 102 (30 nm thick) alternately deposited, and a sample with 200 nm slit-shaped spaces formed was used to evaluate the etching characteristics in a fine pattern. The results are shown in Table 3.

As shown in Table 3, the etching amount of the silicon nitride film 103 in the slit sample also increased as the time of IR irradiation simultaneously with the methanol gas flow increased. In order to make the evaluation results easier to understand, Table 3 also lists the symbols ◎, ◯, △, and × shown in Tables 1-3 above.

Simultaneous IR irradiation with methanol gas flow, as in this example, was also effective, and except for the short time of 30 seconds in Table 3, the residual _SiO2 thickness at the bottom improved to 0.90 or more, and the difference in etching amount between the top and bottom was reduced to 2.0 nm or less, resulting in good etching results as shown in Figure 14. It was found that simultaneous methanol gas flow with heat treatment suppressed the reaction products from increasing the etching amount at the bottom and, conversely, decreasing the etching amount at the top.

Another embodiment of the present invention's etching process using hydrogen fluoride gas without plasma will be described. Here, the flow shown in Figures 5-1 and 9-1 will be explained using the etching processing apparatus 100A shown in Figure 4.

[Configuration example 2 of etching processing apparatus]
Next, referring to FIG. 4 , an outline of an etching processing apparatus 100A according to a fourth embodiment of the present invention will be described, including its overall configuration. The processing chamber 1 of the etching processing apparatus 100A is composed of a base chamber 11, which is equipped with a wafer stage 3 for supporting a wafer 2. A plasma source is installed above the processing chamber 1 and employs an ICP (Inductively Coupled Plasma) discharge method. The ICP plasma source can be used to clean the inner walls of the chamber 11 using plasma or to generate reactive gases using plasma. A cylindrical quartz chamber 12 constituting the ICP plasma source is installed above the processing chamber 1, and an ICP coil 20 is installed outside the quartz chamber 12. A high-frequency power source 21 for plasma generation is connected to the ICP coil 20 via a matching device 22. The high-frequency power frequency is assumed to be in the tens of megahertz range, such as 13.56 MHz. A top plate 25 is installed above the quartz chamber 12. A gas dispersion plate 24 and a shower plate 23 are installed below the top plate 25 , and the processing gas is introduced into the quartz chamber 12 via the gas dispersion plate 24 and the shower plate 23 .

The supply flow rate of the process gas is adjusted by mass flow controllers 52 installed in mass flow boxes 50 for each gas type. A gas distributor 51 is installed downstream of the mass flow controller 52, allowing the flow rate and composition of the gas supplied near the center of the quartz chamber 12 and the gas supplied near the periphery to be independently controlled, thereby enabling detailed control of the spatial distribution of the process gas partial pressure. Note that FIG. 4 illustrates Ar, N ₂ , HF, O ₂ , and methanol (CH ₃ OH), an alcohol, as a polar solvent gas. Other process gases can also be supplied. In particular, water, ethanol, isopropanol, and other polar solvents can be used. However, when using water, because it freezes below 0°C, a stage temperature higher than 0°C is required.

The lower part of the processing chamber 1 is connected to an exhaust means (exhaust device) 15 via a vacuum exhaust pipe 16 in order to reduce the pressure inside the processing chamber 1. The exhaust means may be composed of, for example, a turbomolecular pump, a mechanical booster pump, or a dry pump. In addition, a pressure adjustment means (pressure adjustment device) 14 is installed upstream of the exhaust means 15 to adjust the pressure inside the processing chamber 1.

An IR lamp unit for heating the wafer 2 is installed above the wafer stage 3. The IR lamp unit mainly consists of an IR lamp 60, a reflector 61, and an IR light-transmitting window 72. A circular (circular, annular) lamp is used for the IR lamp 60. The light emitted from the IR lamp 60 is primarily light ranging from visible light to infrared light (herein referred to as IR light). In this embodiment, three lamps 60-1, 60-2, and 60-3 are installed as the IR lamp 60, but two or four lamps may be installed. A reflector 61 is installed above the IR lamp 60 to reflect IR light downward (toward the wafer placement direction). The IR light-transmitting window 72 is preferably made of a material that does not contain alkali metal ions, transmits light in the infrared range, and is heat-resistant. Quartz is a preferred material.

An IR lamp power supply 73 is connected to the IR lamp 60, and a high-frequency cut filter 74 is installed along the way to prevent high-frequency power noise from entering the IR lamp power supply. The IR lamp power supply 73 also has a function that allows the power supplied to IR lamps 60-1, 60-2, and 60-3 to be controlled independently of each other, allowing the radial distribution of the heating amount on the wafer 2 to be adjusted (some of the wiring is not shown).

A flow path 27 is formed in the center of the IR lamp unit. This flow path 27 is equipped with a slit plate 26 with multiple holes that blocks ions and electrons generated in the plasma and allows only neutral gases and neutral radicals to pass through and irradiate the wafer 2. The slit plate 26 is preferably heat-resistant and does not contain alkali metal ions, and specific materials that can be used are alumina and quartz.

The wafer stage 3 has a coolant flow path 39 formed inside it for cooling the stage 3, and the coolant is circulated and supplied by a chiller 38. In this invention, a chiller 38 capable of temperature control between -50°C and 50°C is used. To fix the wafer 2 by electrostatic adsorption, plate-shaped electrodes 30 are embedded in the stage 3, each connected to a DC (direct current) power supply 31. To efficiently cool the wafer 2, He gas can be supplied between the backside of the wafer 2 and the wafer stage 3. To prevent damage to the backside of the wafer 2 even when heating and cooling are performed while the wafer 2 is still adsorbed, the surface of the wafer stage 3 (the surface on which the wafer 2 is placed) is coated with a resin such as polyimide. A thermocouple 70 is installed inside the wafer stage 3 to measure the temperature of the stage 3, and the thermocouple 70 is connected to a thermocouple thermometer 71.

The temperature of stage 3 measured by thermocouple thermometer 71 using thermocouple 70 was within ±1°C of the set temperature of chiller 38, and the wafer temperature measured separately by thermocouple 70 was within ±3°C of the set temperature (within ±2°C of the stage temperature).

In addition, as a mechanism for cooling the stage 3 used in the etching processing apparatus 100A of the present invention, in addition to one that circulates a refrigerant, a thermoelectric conversion device such as a Peltier element can also be used.

The etching processing apparatus 100A used in the present invention can also heat the inside of the chamber other than the wafer stage, which is exposed to hydrogen fluoride gas, such as the processing chamber 1. For example, a temperature of approximately 40°C to 120°C can be used. This prevents hydrogen fluoride gas from being adsorbed inside the chamber 11, minimizing corrosion inside the chamber 11.

[Etching process flow 4]
Next, the etching process using hydrogen fluoride gas without using plasma proposed in the present invention will be described. Here, the flow is basically the same as that of Example 1, but since the etching processing apparatus 100A shown in Figure 4 is used, there are some differences from Flow 1 of the etching process of Example 1. The flow will be described using Figures 5-1, 9-1, and 4 (apparatus diagram).

First, the wafer 2 is transported into the processing chamber 1 through a transport port (not shown) provided in the processing chamber 1, and then the wafer 2 is fixed to the wafer stage 3 using a DC power supply 31 for electrostatic adsorption, and He gas 55 for wafer cooling is supplied to the back surface of the wafer 2, thereby performing wafer cooling in S101 of Figure 5-1.

After this, etching is performed according to the flow diagram in Figure 5-1 described in Example 1. The explanation of the flow in Figure 5-1 is omitted here to avoid redundancy.

Figure 9-1 shows a time chart for the flow shown in Figure 5-1 when using the etching processing apparatus 100A shown in Figure 4. It is an electrostatic adsorption stage 3, and a section has been added that supplies He gas 55 for wafer cooling to the backside of the wafer 2. One cycle includes a process of IR lamp heating while flowing HF gas, a process of flowing polar solvent gas, and a process of IR lamp heating without flowing HF gas or polar solvent gas. This process is repeated N times to etch the silicon nitride film.

[Etching result 4]
Etching was performed using the etching processing apparatus 100A shown in Figure 4 and the process flows shown in Figures 5-1 and 9-1. In the etching processing apparatus 100A shown in Figure 4, a voltage of ±1200 V was applied during etching to electrostatically attract the wafer 2. In addition, to improve the thermal conductivity of the stage 3, He was flowed from the rear surface at a pressure of 1.0 kPa.

The stage temperature was set to -30°C, and the pressure was increased to 900 Pa with Ar at 1.4 L/min. After that, the pressure was maintained at 900 Pa, and HF was introduced at 0.40 L/min and Ar as a dilution gas at 0.20 L/min. At the same time, the IR lamp 60 was irradiated at 60% output for 60 seconds. This resulted in the formation of a reaction layer on the silicon nitride film 103.

Then, with the exhaust valve fully open, the reactor was evacuated for 60 seconds. This exhaust operation evacuated fluorine gas and some of the reaction products. Next, while the set temperature of stage 3 remained the same, methanol was flowed as a polar solvent gas for 60 seconds at a flow rate of 0.4 L/min and pressures of 300 Pa, 450 Pa, and 600 Pa. By flowing this polar solvent gas, the reaction layer was modified and partially removed.

Next, with the set temperature of stage 3 unchanged, Ar was flowing at 0.50 L/min, and the exhaust valve was fully open, heating was performed for 40 seconds at 70% lamp intensity. This removes the reaction layer modified by the polar solvent gas flow. After that, the process returned to the beginning and cooled with Ar flowing at 1.4 L/min for 60 seconds at a pressure of 900 Pa. This series of processes was repeated 10 times, following the flow shown in Figure 5-1.

The experimental results for the slit samples evaluated in Example 1 in the same manner as shown in Tables 1-2 and 1-4 are shown in Table 4.

As shown in Table 4, looking at the etching amount of the silicon nitride film 103 in the slit sample, it was found that as the pressure of the methanol gas flow was increased, the etching amount increased and was saturating at 450 Pa or more. To make the evaluation results easier to understand, Table 4 also lists the symbols ◎, ◯, △, and × shown in Tables 1-3 above.

Even when the pressure of the methanol gas flow was increased, the residual _SiO2 thickness at the bottom was 0.90 or more, and the difference in etching amount between the top and bottom was small, at 2.0 nm or less, demonstrating favorable etching results as shown in FIG. 14.

4 used in this example is equipped with an ICP plasma generation mechanism, and therefore prior to the etching process, the chamber was cleaned with oxygen plasma for 300 seconds under conditions of _O2 = 1.0 L/min, 50 Pa, and 1500 W. This enabled the reduction of foreign matter in the chamber.

Another embodiment of the present invention's etching process using hydrogen fluoride gas without plasma will be described. Here, the flow shown in Figures 6-1 and 9-2 will be explained using the etching processing apparatus 100A shown in Figure 4.

[Etching process flow 5]
Next, a plasma-free hydrogen fluoride gas etching process proposed in the present invention will be described. The process is basically similar to the flow shown in the previous embodiment, but there are some differences from Flow 1 of the etching process in Example 1 because the etching processing apparatus 100A shown in Figure 4 is used.

First, the wafer 2 is transported into the processing chamber 1 through a transport port (not shown) provided in the processing chamber 1, and then the wafer 2 is fixed to the wafer stage 3 using a DC power supply 31 for electrostatic adsorption, and He gas 55 for wafer cooling is supplied to the back surface of the wafer 2, thereby performing wafer cooling in S101 of Figure 6-1.

Etching was performed using the etching processing apparatus 100A shown in Figure 4 and the process flows shown in Figures 6-1 and 9-2. In the etching processing apparatus 100A shown in Figure 4, a voltage of ±1200 V was applied during etching to electrostatically adsorb the wafer 2. Here, the stage temperature was set to -30°C. In addition, to improve the thermal conductivity of the stage 3, He was flowed from the backside at a pressure of 1.0 kPa. This allowed for stage cooling in S111.

Next, Ar gas was introduced in step S112. This Ar gas is a dilution gas and is used here to increase the pressure. Since we are using electrostatic adsorption stage 3 and cooling from the backside with He, this step can be omitted.

Next, as shown in step S113 of FIG. 6-1, HF gas and Ar gas were supplied to the process chamber 1 in predetermined amounts for a predetermined time, and simultaneously, heating was performed to form a reaction layer. Here, heating was performed using IR (infrared) lamps 60. The wafer temperature obtained as a result of cooling by the stage 3 and heating by the IR lamps 60 is preferably 30°C to 55°C, more preferably 35°C to 50°C. The reaction layer thickness can be controlled by adjusting the total pressure or HF partial pressure, heating temperature, IR lamp output, time, and number of repetitions. Furthermore, if the wafer temperature is below 30°C, etching is difficult because an insufficient reaction layer is formed. Conversely, if the wafer temperature is above 55°C, an excessive reaction layer is formed, which, when decomposed and volatilized, tends to etch the adjacent silicon oxide film 102, resulting in reduced selectivity.

In the present invention, the pressure used is preferably approximately 10 Pa to 1000 Pa, and particularly preferably 100 Pa to 1000 Pa. The higher the pressure, the easier it is to form a reaction layer on the silicon nitride film 103, and the lower the temperature required for formation. Even when the pressure is high, by controlling the output of the IR lamp 60, it is possible to form a reaction layer on the silicon nitride film 103 without affecting the silicon oxide film 102.

After the reaction layer is formed, the supply of HF gas is stopped and any HF gas remaining in the gas phase is evacuated, although this is not shown in Figure 6-1. If evacuating to a vacuum, it is desirable to set the pressure to 5 Pa or less. In other words, here, a process of evacuating HF gas while flowing an inert gas is inserted between S113 and the subsequent S114.

Next, in step S114 of FIG. 6-1, the reaction layer is removed while being modified by heating while introducing a polar solvent gas. The polar solvent gas can be water or alcohols such as methanol, ethanol, or isopropanol. The flow rate of the polar solvent gas is preferably 0.1 to 3.0 L/min. The heating temperature is preferably 70 to 110°C, more preferably 70 to 100°C. Here, an IR lamp 60 was used as the heating method. The heating method is not limited to this; for example, the wafer stage 3 can be heated, or the wafer 2 can be transported separately to a heating-only device and subjected to heating treatment. Furthermore, inert gases such as Ar gas or nitrogen gas can be introduced in the process of removing the reaction layer while being modified by heating while introducing a polar solvent gas. Furthermore, this process can be performed multiple times as needed.

Next, although not shown in Figure 6-1, the heating and supply of polar solvent gas are stopped, and the polar solvent gas remaining in the gas phase is evacuated. When evacuating to a vacuum, it is desirable to set the pressure to 5 Pa or less. In other words, here, a process of evacuating the polar solvent gas while flowing an inert gas is inserted after S114.

After this, the process returns to cooling the wafer using stage 3 in S111. After this, the steps from S111 to S115 constitute one cycle, and this is repeated N times. The cycle is repeated until the required amount of etching is achieved, and then the process ends.

Figure 9-2 shows a time chart for the flow shown in Figure 6-1. One cycle includes a process of flowing HF gas and Ar (the process of forming a reaction layer), a process of flowing polar solvent gas, and a process of IR lamp heating without flowing HF gas. By repeating this process N times, the silicon nitride film is etched.

[Etching result 5]
Etching was performed using the etching processing apparatus 100A shown in Figure 4 and the process flows shown in Figures 6-1 and 9-2. In the etching processing apparatus 100A shown in Figure 4, a voltage of ±1200 V was applied during etching to electrostatically attract the wafer 2. In addition, to improve the thermal conductivity of the stage 2, He was flowed from the rear surface at a pressure of 1.0 kPa.

The stage temperature was set to -30°C, and the pressure was increased to 900 Pa with Ar at 1.4 L/min. After that, the pressure was maintained at 900 Pa, and HF was introduced at 0.40 L/min and Ar as a dilution gas at 0.20 L/min. At the same time, the IR lamp 60 was irradiated at 60% output for 60 seconds. This resulted in the formation of a reaction layer on the silicon nitride film 103.

Then, the exhaust valve was opened 100% and the mixture was evacuated for 120 seconds. This exhaust operation allowed the fluorine gas and some of the reaction products to be exhausted.

Next, with the stage temperature set unchanged, the pressure was controlled to 300 Pa with methanol gas flowing at 0.40 L/min, and heating was performed for 30 to 50 seconds at 70% IR lamp output. This removes the reaction layer as the methanol gas flows. After that, returning to the beginning, the wafer was cooled by flowing Ar at 1.4 L/min for 60 seconds at a pressure of 900 Pa until it reached the same temperature as the stage. This series of processes was repeated 10 times, following the flow in Figure 6-1.

Here, similar to Example 1, a sample was formed with a total of 40 layers of silicon nitride film 103 (30 nm thick) and silicon oxide film 102 (30 nm thick) alternately deposited, and a sample with 200 nm slit-shaped spaces formed was used to evaluate the etching characteristics in a fine pattern. The results are shown in Table 5.

As shown in Table 5, the etching amount of the silicon nitride film 103 in the slit sample also increased as the time of IR irradiation with the methanol gas flow increased. In order to make the evaluation results easier to understand, Table 5 also lists the symbols ◎, ◯, △, and × shown in Tables 1-3 above.

Simultaneous IR irradiation with methanol gas flow, as in this example, was also effective, and except for the short time of 30 seconds in Table 5, the difference in etching amount between the top and bottom was small, at 2.5 nm or less, and the residual _SiO2 thickness was 0.95 in all cases, demonstrating favorable etching results as shown in Figure 14. It was found that simultaneous methanol gas flow with heat treatment suppresses the reaction products from increasing the etching amount at the bottom and, conversely, decreasing the etching amount at the top.

The invention made by the inventor has been specifically described above based on examples, but it goes without saying that the present invention is not limited to the above embodiments and examples and can be modified in various ways.

1: Processing chamber, 2: Wafer, 3: Wafer stage, 11: Base chamber, 12: Quartz chamber, 13: Discharge area, 14: Pressure adjustment means, 15: Exhaust means, 16: Vacuum exhaust piping, 20: ICP coil, 21: High frequency power supply, 22: Matching machine, 23: Shower plate, 24: High gas dispersion plate, 25: Top plate, 26: Slit plate, 27: Flow path, 30: Electrostatic adsorption electrode, 31: DC power supply for electrostatic adsorption, 38: Chiller, 39: Refrigerant flow path, 50: Mass flow box, 51: Gas distributor, 52: Mass flow controller, 54: Valve, 55: He gas, 56: Proximity cooling protrusion, 60, 60-1, 60-2, 60-3: IR lamp, 61: Reflector, 64: IR lamp power supply, 70: Thermocouple, 71: Thermocouple thermometer , 72: IR light transmitting window, 73: IR lamp power supply, 74: high frequency cut filter, 100, 100A: etching processing apparatus, 101: substrate, 102: silicon nitride film, 103: silicon oxide film, 104: opening, 105: laminated film, 106: etching amount of silicon oxide film relative to silicon nitride film, 109: etching amount of silicon oxide film relative to silicon nitride film, 111: edge of silicon oxide film after etching when selectivity is low, 112: diagram showing an example of the edge of silicon oxide film after etching, where the corners of the silicon oxide film remain rectangular while the film thickness of the silicon oxide film portion has become thinner, 113: diagram showing an example of the edge of silicon oxide film after etching when selectivity is high and the film thickness is also maintained.

Claims (16)

A dry etching method for etching a film structure in advance formed on a wafer placed in a processing chamber, the film structure including a silicon nitride film sandwiched between silicon oxide films from above and below, and an end portion of a film layer constituting a side wall of a groove or hole, by supplying a processing gas into the processing chamber without using plasma, comprising:
(a) reacting hydrogen fluoride gas at a predetermined temperature to form a reaction layer on the silicon nitride film;
(b) treating the reaction layer formed in the step (a) by flowing a polar solvent gas;
(c) a step of heating at a temperature higher than that in the step (a) in a state where the hydrogen fluoride gas is not flowed, thereby volatilizing and removing the reaction layer formed in the step (a) and treated in the step (b),
an etching method, characterized in that the silicon nitride film is etched laterally from the end by repeating the steps (a), (b), and (c) multiple times. A dry etching method for etching a film structure in advance formed on a wafer placed in a processing chamber, the film structure including a silicon nitride film sandwiched between silicon oxide films from above and below, and an end portion of a film layer constituting a side wall of a groove or hole, by supplying a processing gas into the processing chamber without using plasma, comprising:
(a) reacting hydrogen fluoride gas at a predetermined temperature to form a reaction layer on the silicon nitride film;
(d) a step of heating at a temperature higher than that in the step (a) while flowing a polar solvent gas and without flowing the hydrogen fluoride gas, thereby volatilizing and removing the reaction layer formed in the step (a),
an etching method, characterized in that the steps (a) and (d) are repeated a plurality of times to etch the silicon nitride film laterally from the end portion. 2. The etching method according to claim 1,
An etching method characterized in that the temperature in the step (a) is 30°C or higher and 55°C or lower. 2. The etching method according to claim 1,
The etching method is characterized in that the temperature in the step (c) is 70°C or higher and 110°C or lower. 2. The etching method according to claim 1,
An etching method characterized in that the polar solvent in the step (b) is an alcohol. 2. The etching method according to claim 1,
An etching method, wherein the heating in the step (c) is lamp heating. 2. The etching method according to claim 1,
The etching method is characterized in that, in the steps (a) and (c), the stage is set to a low temperature of -50°C or more and 0°C or less, and lamp heating is performed thereon, thereby obtaining a stage temperature of 30°C or more and 55°C or less in the step (a), and further obtaining a stage temperature of 70°C or more and 110°C or less in the step (c). 2. The etching method according to claim 1,
An etching method characterized in that the pressure in the step (a) is 50 Pa or more and 1000 Pa or less. 2. The etching method according to claim 1,
An etching method characterized by inserting a step of exhausting the gas while flowing an inert gas between the steps (a), (b) and (c). 3. The etching method according to claim 2,
An etching method characterized in that the temperature in the step (a) is 30°C or higher and 55°C or lower. 3. The etching method according to claim 2,
The etching method is characterized in that the temperature in the step (d) is 70°C or higher and 110°C or lower. 3. The etching method according to claim 2,
An etching method characterized in that the polar solvent in step (d) is an alcohol. 3. The etching method according to claim 2,
An etching method, wherein the heating in the step (d) is lamp heating. 3. The etching method according to claim 2,
The etching method is characterized in that, in the steps (a) and (d), the stage is set to a low temperature of -50°C or more and 0°C or less, and then lamp heating is performed to obtain a stage temperature of 30°C or more and 55°C or less in the step (a), and further, in the step (d), the stage is set to a temperature of 70°C or more and 110°C or less. 3. The etching method according to claim 2,
An etching method characterized in that the pressure in the step (a) is 50 Pa or more and 1000 Pa or less. 3. The etching method according to claim 2,
An etching method characterized by inserting a step of exhausting the gas while flowing an inert gas between the steps (a) and (d).