Dry etching method or dry cleaning method

11584989 · 2023-02-21

Assignee

Inventors

Cpc classification

International classification

Abstract

Provided are a method of selectively etching a film primarily containing Si, such as polycrystalline silicon (Poly-Si), single crystal silicon (single crystal Si), or amorphous silicon (a-Si) as well as a method for cleaning by removing a Si-based deposited and/or attached matter inside a sample chamber of a film forming apparatus, such as a chemical vapor deposition (CVD) apparatus, without damaging the apparatus interior. By simultaneously introducing a monofluoro interhalogen gas (XF, where X is any of Cl, Br, and I) and nitric oxide (NO) into an etching or a film forming apparatus, followed by thermal excitation, it is possible to selectively and rapidly etch a Si-based film, such as Poly-Si, single crystal Si, or a-Si, while decreasing the etching rate of SiN and/or SiO.sub.2. It is also possible to perform cleaning by removing a Si-based deposited and/or attached matter inside a film forming apparatus, such as a CVD apparatus, without damaging the apparatus interior.

Claims

1. A dry etching method comprising: simultaneously introducing gases consisting of a monofluoro interhalogen gas represented by XF, where X is a halogen element selected from the group consisting of Cl, Br and I, and nitric oxide (NO), and optionally an inert gas as a diluent gas, into a reaction chamber of an etching apparatus; and thermally inducing a reaction of an object to be dry etched with XF and NO to dry etch the object, wherein, during dry etching, a temperature inside the reaction chamber of the etching apparatus or a temperature of the object to be etched is 100° C. to 400° C., and wherein, by adjusting the inside of the reaction chamber of the etching apparatus or the object to be etched to 100° C. to 400° C., a film consisting of Si is selectively etched relative to oxide and/or nitride of Si.

2. The dry etching method according to claim 1, wherein when a mixing ratio of the monofluoro interhalogen gas represented by XF to nitric oxide (NO) is expressed in a volume ratio or a flow rate ratio as XF:NO=1:Y, Y satisfies 0<Y<2.

3. The dry etching method according to claim 1, wherein a monofluoro halogen gas is ClF.

4. The dry etching method according to claim 1, wherein, during dry etching, the temperature inside the reaction chamber of the etching apparatus or a temperature of the object to be etched is 100° C. to 300° C.

5. The dry etching method according to claim 1, wherein by diluting XF and nitric oxide (NO) with at least one inert gas selected from the group consisting of N.sub.2, Ar, He, Kr, and Xe, an etching rate and/or etching selectivity are controlled in dry etching.

6. The dry etching method according to claim 1, wherein a chemical reaction of the following equation 1 is induced by a heating temperature of 100° C. to 400° C. for a mixture of the monofluoro interhalogen gas represented by XF and nitric oxide (NO),
XF+NO.fwdarw.X+FNO  (1) and generated X atoms, where X is a halogen element consisting of Cl, Br and I, and nitrosyl fluoride (FNO) are supplied.

7. The dry etching method according to claim 1, wherein the film consisting of Si is Poly-Si film, and the oxide and/or nitride of Si is SiO.sub.2 and/or SiN.

8. The dry etching method according to claim 1, wherein the object to be etched comprises an underlying film comprising oxide and/or nitride of Si and a film consisting of Si on the underlying film.

9. A dry etching method comprising: mixing gases consisting of a monofluoro interhalogen gas represented by XF, where X is a halogen element selected from the group consisting of Cl, Br and I, and nitric oxide (NO), and optionally an inert gas as a diluent gas, followed by heating to a temperature to induce a chemical reaction between XF and NO; and supplying a generated excited species to the inside of an etching apparatus, wherein the heating temperature for a mixture of the monofluoro interhalogen gas represented by XF and nitric oxide (NO) is 100° C. to 400° C., and wherein, by adjusting the inside of the reaction chamber of the etching apparatus or the object to be etched to 100° C. to 400° C., a film consisting of Si is selectively etched relative to oxide and/or nitride of Si.

10. The dry etching method according to claim 9, wherein the heating temperature for a mixture of the monofluoro interhalogen gas represented by XF and nitric oxide (NO) is 100° C. to 300° C.

11. The dry etching method according to claim 9, wherein a chemical reaction of the following equation 1 is induced by a heating temperature of 100° C. to 400° C. for a mixture of the monofluoro interhalogen gas represented by XF and nitric oxide (NO),
XF+NO.fwdarw.X+FNO  (1) and generated X atoms, where X is a halogen element consisting of Cl, Br and I, and nitrosyl fluoride (FNO) are supplied.

12. The dry etching method according to claim 9, wherein when a mixing ratio of the monofluoro interhalogen gas represented by XF to nitric oxide (NO) is expressed in a volume ratio or a flow rate ratio as XF:NO=1:Y, Y satisfies 0<Y<2.

13. The dry etching method according to claim 9, wherein a monofluoro halogen gas is ClF.

14. The dry etching method according to claim 9, wherein by diluting XF and nitric oxide (NO) with at least one inert gas selected from the group consisting of N.sub.2, Ar, He, Kr, and Xe, an etching rate and/or etching selectivity are controlled in dry etching.

15. The dry etching method according to claim 9, wherein the film consisting of Si is Poly-Si film, and the oxide and/or nitride of Si is SiO.sub.2 and/or SiN.

16. The dry etching method according to claim 9, wherein the object to be etched comprises an underlying film comprising oxide and/or nitride of Si and a film consisting of Si on the underlying film.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 is a schematic view of a thermal CVD apparatus used in the Examples section.

DESCRIPTION OF EMBODIMENTS

(2) The monofluoro interhalogen gas represented by XF used in the dry etching method or the dry cleaning method of the present invention has a purity of desirably 80 vol % or more and particularly preferably 90 vol % or more.

(3) When a mixing ratio of the monofluoro interhalogen gas represented by XF to nitric oxide (NO) is expressed in a volume ratio or a flow rate ratio as XF:NO=1:Y, Y preferably satisfies the range of 0<Y<2. Further, the range of 0.5<Y≤1 is particularly preferable since the following reaction progresses efficiently.
XF+NO.fwdarw.X+FNO
When the ratio of NO is high, the concentration of X atoms is diluted or XNO, which is not quite active in reactions, is generated, thereby lowering etching and cleaning efficiency. Meanwhile, when the ratio of NO is low, excessive reactions progress due to XF, thereby making controlled reactions or selective reactions difficult.

(4) In the present invention, by using an appropriate inert gas, such as N.sub.2, He, Ar, Kr, Xe, or CO.sub.2, as a diluent gas, excessive reactions are suppressed in the reaction between the monofluoro interhalogen gas represented by XF and nitric oxide (NO) as well as in the reaction between X atoms, which are generated by the reaction between XF and NO, and a film or a deposit primarily containing an element, such as Si, W, or Al, that forms a volatile compound by the reaction with X. Such use of a diluent gas is preferable to perform etching or cleaning with better controllability. The diluent gas is preferably mixed to have an XF concentration of 5 to 50 vol % and is particularly preferably mixed to have an XF concentration of 5 to 30 vol %.

(5) In the dry etching method or the dry cleaning method of the present invention, X atoms, which are generated by the reaction between the monofluoro interhalogen gas represented by XF and nitric oxide (NO), has high reaction activity. Accordingly, reactions with a material to be etched or a material to be removed for cleaning progress even at 20° C., thereby making formation and subsequent removal of a volatile substance possible. Meanwhile, to promote formation reactions of a volatile substance and to efficiently volatilize and remove the formed product, the atmosphere within an etching apparatus or an apparatus to be cleaned, a substrate to be etched, a wall surface of an apparatus, or the like is heated to 100° C. or higher in some cases. A low reaction temperature, at which reactions are slow, is preferable when controllability is required. When a short cleaning time or etching rate is required and when the boiling point of a compound to be formed by the reaction is high, etching or cleaning is preferably performed at a temperature of 20° C. or higher and is more preferably performed at a temperature of 100° C. or higher. Meanwhile, when etching or cleaning is performed at 700° C. or higher, controlled reactions become difficult. Accordingly, etching or cleaning is performed preferably at 700° C. or lower, more preferably 400° C. or lower, and particularly preferably 300° C. or lower. At a temperature of 700° C. or lower, the method of the present invention is found to exhibit effects comparable to or higher than conventional methods that use XF, F.sub.2, or ClF.sub.3. In particular, the method of the present invention exhibits, at any temperature, higher performance than conventional methods that use XF and demonstrates performance comparable to or higher than methods that use a highly reactive gas, such as F.sub.2 or ClF.sub.3.

(6) In the dry etching method or the dry cleaning method of the present invention, it is possible to efficiently react X atoms, which are generated by the reaction between XF and NO, with a material to be etched or a material to be removed for cleaning by introducing a monofluoro interhalogen gas represented by XF, nitric oxide (NO), and an inert gas selected from N.sub.2, He, Ar, Kr, Xe, and CO.sub.2 and by adjusting the pressure inside an etching apparatus or a film forming apparatus to 0.001 to 760 Torr. In particular, the pressure is preferably adjusted to 0.001 to 300 Torr since even a reaction product having a high boiling point and thus a low vapor pressure can be efficiently discharged outside the apparatus.

EXAMPLES

(7) Hereinafter, the present invention will be described in further detail with reference to the Examples and Comparative Examples. The present invention, however, is not limited to these Examples.

(8) The examples below were performed by using the thermal CVD apparatus illustrated in FIG. 1. In the apparatus of FIG. 1, a process vessel is provided to ensure space for CVD processing of a sample, and a pipe that can circulate fluids is connected to the process vessel. As illustrated FIG. 1, the pipe originates from an XF supply line and is joined with an NO supply line and further with an inert gas line (N.sub.2 line) on the downstream side of the NO supply line. The pipe is designed to supply a mixture of XF, NO, and an inert gas to the process vessel. A mass flow controller is provided on each line, and a flow rate of a gas is adjustable for each line. Temperature adjustment is possible for the sample mount and the processing space. An exhaust line for discharging a gas after reaction is also provided in the process vessel. Other specifications of the apparatus are as follows.

(9) Materials for the apparatus (chamber wall: quartz, pipe and other parts: SUS 306, susceptor and heating section: Ni)

(10) Reaction chamber size (diameter ø222 mm, height: 200 mm)

(11) Heating mode (resistive heating)

Example 1

(12) A Si wafer sample composed of a 100 nm SiO.sub.2 film formed on a Si substrate and a 300 nm polycrystalline silicon (Poly-Si) film further formed on the SiO.sub.2 film, a Si wafer sample composed of a 300 nm SiN film formed on a Si substrate, and a Si wafer sample composed of a 1,000 nm SiO.sub.2 film formed on a Si substrate were placed inside the process vessels of a vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, ClF was used. CIF and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 100° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the Poly-Si film was over-etched and the etching rate was 600.0 nm/min or more, the etching rate of the SiN film was 1.5 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 400.0 or more, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity.

Example 2

(13) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CIF was used. CIF and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 200° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the Poly-Si film was over-etched and the etching rate was 600.0 nm/min or more, the etching rate of the SiN film was 1.7 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 352.9 or more, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity.

Example 3

(14) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CIF was used. CIF and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 300° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the Poly-Si film was over-etched and the etching rate was 600.0 nm/min or more, the etching rate of the SiN film was 1.2 nm/min, and the etching rate of the SiO.sub.2 film was 12.9 nm/min. The selectivity of Poly-Si relative to SiN is 500.0 or more, and the selectivity of Poly-Si relative to SiO.sub.2 is 46.5 or more.

Example 4

(15) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CIF was used. CIF and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 400° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the Poly-Si film was over-etched and the etching rate was 600.0 nm/min or more, the etching rate of the SiN film was 24.9 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 24.1 or more, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity.

Comparative Example 1

(16) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CIF was used. CIF was supplied to the inside of each process vessel at a flow rate of 100 sccm through amass flow controller from a cylinder. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 100° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the etching rate of the Poly-Si film was 2.0 nm/min, the etching rate of the SiN film was 1.2 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 1.7, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity. Here, the etching rate of Poly-Si is 0.003 times or less the result in Example 1.

Comparative Example 2

(17) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, ClF was used. CIF was supplied to the inside of each process vessel at a flow rate of 100 sccm through a mass flow controller from a cylinder. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 200° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the etching rate of the Poly-Si film was 170.8 nm/min, the etching rate of the SiN film was 3.3 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 51.8, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity. Here, the etching rate of Poly-Si is 0.285 times or less the result in Example 1.

Comparative Example 3

(18) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CF was used. ClF was supplied to the inside of each process vessel at a flow rate of 100 sccm through amass flow controller from a cylinder. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 300° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the etching rate of the Poly-Si film was 375.4 nm/min, the etching rate of the SiN film was 12.5 nm/min, and the etching rate of the SiO.sub.2 film was 4.0 nm/min. The selectivity of Poly-Si relative to SiN is 30.0, and the selectivity of Poly-Si relative to SiO.sub.2 is 93.9. Here, the etching rate of Poly-Si is 0.626 times or less the result in Example 1.

Comparative Example 4

(19) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, ClF was used. CF was supplied to the inside of each process vessel at a flow rate of 100 sccm through amass flow controller from a cylinder. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 400° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the Poly-Si film was over-etched and the etching rate was 600.0 nm/min or more, the etching rate of the SiN film was 141.7 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 4.2 or more, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity.

Example 5

(20) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CIF was used. CIF and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm for CIF and 300 sccm for NO through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 200° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the etching rate of the Poly-Si film was 501.4 nm/min, the etching rate of the SiN film was 2.1 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 238.8, which is lower than that in Example 2, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity.

Example 6

(21) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. As a monofluoro halogen gas, CIF was used. ClF and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm for ClF and 100 sccm for NO through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 20° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, the etching rate of the Poly-Si film was 29.4 nm/min, the etching rate of the SiN film was 0.4 nm/min, and the etching rate of the SiO.sub.2 film was 0.0 nm/min. The selectivity of Poly-Si relative to SiN is 73.5, and the selectivity of Poly-Si relative to SiO.sub.2 is infinity.

Comparative Example 5

(22) Poly-Si, SiN, and SiO.sub.2 samples same as Example 1 were employed and placed inside the process vessels of the vacuum apparatus that can individually heat these samples. In place of a monofluoro halogen gas, F.sub.2 was used. F.sub.2 and NO were each supplied to the inside of each process vessel at a flow rate of 100 sccm through a mass flow controller from the respective cylinders. At the same time, N.sub.2 as a diluent inert gas was supplied at a flow rate of 400 sccm. The temperature and pressure inside the apparatus were adjusted to 200° C. and 100 Torr, respectively, and the samples were processed for 30 seconds. As a result, both the Poly-Si film and the SiN film were over-etched. The etching rate of the Poly-Si film was 600.0 nm/min or more, the etching rate of the SiN film was 600.0 nm/min or more, and the etching rate of the SiO.sub.2 film was 24.2 nm/min. The selectivity of Poly-Si relative to SiN is unknown, and the selectivity of Poly-Si relative to SiO.sub.2 is 24.8 or more.

(23) The etching rate and selectivity of each Example and Comparative Example are shown in Table 1.

(24) TABLE-US-00001 TABLE 1 XF flow NO flow N.sub.2 flow rate rate rate Temperature Etching rate Poly-Si XF [sccm] [sccm] [sccm] [° C.] [nm/min] selectivity Ex. 1 ClF 100 100 400 100 Poly-Si: >600.0 /SiN: >400.0 SiN: 1.5 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Ex. 2 ClF 100 100 400 200 Poly-Si: >600.0 /SiN: >352.9 SiN: 1.7 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Ex. 3 ClF 100 100 400 300 Poly-Si: >600.0 /SiN: >500.0 SiN: 1.2 /SiO.sub.2: > 46.5 SiO.sub.2: 12.9 Ex. 4 ClF 100 100 400 400 Poly-Si: >600.0 /SiN: >24.1 SiN: 24.9 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Comp. ClF 100 0 400 100 Poly-Si: 2.0 /SiN: 1.7 Ex. 1 SiN: 1.2 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Comp. ClF 100 0 400 200 Poly-Si: 170.8 /SiN: 51.8 Ex. 2 SiN: 3.3 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Comp. ClF 100 0 400 300 Poly-Si: 375.4 /SiN: 30.0 Ex. 3 SiN: 12.5 /SiO.sub.2: 93.9 SiO.sub.2: 4.0 Comp ClF 100 0 400 400 Poly-Si: >600.0 /SiN: >4.2 Ex. 4 SiN: 141.7 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Ex. 5 ClF 100 300 400 200 Poly-Si: 501.4 /SiN: 238.8 SiN: 2.1 /SiO.sub.2: ∞ SiO.sub.2: 0.0 Ex. 6 ClF 100 100 400 20 Poly-Si: 29.4 /SiN: 73.5 SiN: 0.4 /SiO.sub.2: ∞ SiO.sub.2 0.0 Comp. F.sub.2 100 100 400 200 Poly-Si: >600.0 /SiN: — Ex. 5 SiN: > 600.0 /SiO.sub.2: > 24.8 SiO.sub.2: 24.2

(25) Examples 1 to 4 reveal that the method of the present invention is excellent in selectivity of Poly-Si to SiN and in selectivity of Poly-Si to SiO.sub.2 at a temperature of 700° C. or lower. In particular, the selectivity of Poly-Si to SiN is found to be stable at a temperature of 100° C. to 300° C.

(26) In comparison between Examples 1 to 4 and Comparative Examples 1 to 4, it is found that in the absence of NO, the etching rate of Poly-Si decreases at all the temperatures as well as the etching rate of Poly-Si is unstable while varying with temperature. As described above, this is because when the ratio of NO is low, excessive reactions progress due to CIF, thereby making controlled reactions difficult.

(27) As in Example 5, when the flow rate of NO is large, the etching rate of Poly-Si decreases, but the etching rate of SiN does not vary considerably. Consequently, the selectivity of Poly-Si to SiN decreases. As described above, this is because when the ratio of NO is high, the concentration of Cl atoms is diluted or ClNO, which is not quite active in reactions, is generated, thereby decreasing etching and cleaning efficiency.

(28) Example 6 reveals that the method of the present invention can be performed even at room temperature of 20° C. while achieving a satisfactorily practical level of the selectivity of Poly-Si to SiN or the selectivity of Poly-Si to SiO.sub.2.

(29) In comparison between Example 2 and Comparative Example 5, it is found that when F.sub.2 is used in place of ClF, not only the etching rate of Poly-Si, but also the etching rate of SiN increases, thereby decreasing the selectivity of Poly-Si to SiN.

(30) As in the foregoing, the method of the present invention are excellent in both selectivity of Poly-Si to SiN as well as selectivity of Poly-Si to SiO.sub.2. In addition, these selectivities are temperature-independent and thus stable in a temperature range of 700° C. or lower. The method of the present invention is found to be particularly excellent in selectivity of Poly-Si to SiN compared with a case in which conventional CIF alone is used as an etchant (Comparative Examples 1 to 4) or a case in which a combination of F.sub.2 and NO is used (Comparative Example 5).