JP4194521B2 - Manufacturing method of semiconductor device - Google Patents

Description

  The present invention relates to a method for manufacturing a semiconductor device using a fluorine-added carbon film as an insulating film, for example, an interlayer insulating film.

  One of the techniques for achieving high integration of semiconductor devices is a technique for multi-layered wiring. In order to take a multi-layered wiring structure, the nth wiring layer and the (n + 1) th wiring layer are connected by a conductive layer. In addition, a thin film called an interlayer insulating film is formed in a region other than the conductive layer. A typical example of this interlayer insulating film is a silicon dioxide (SiO2) film, but in recent years, it has been required to lower the relative dielectric constant of the interlayer insulating film in order to further increase the operation speed of the device. . Due to such a demand, a fluorine-added carbon film (fluorocarbon film) which is a compound of carbon (C) and fluorine (F) has attracted attention. The relative dielectric constant of silicon dioxide film is around 4, whereas the fluorine-added carbon film is extremely effective as an interlayer insulating film because the relative dielectric constant becomes 2.5 or less, for example, if the type of source gas is selected. It is a simple film.

  For this reason, in Patent Document 1, C5F8 gas is used as a source gas, and electron cyclotron resonance is caused by the interaction between a 2.45 GHz microwave and an 875 gauss magnetic field, such as argon (Ar) gas. A technique is described in which a gas for generating plasma is turned into plasma, and a raw material gas is turned into plasma by this plasma to form a fluorine-added carbon film on a semiconductor wafer (hereinafter referred to as a wafer). By the way, as described in Patent Document 1, since the fluorine-added carbon film is an organic film, the gas for etching the fluorine-added carbon film is an organic material at the same time in the etching process. A certain resist film is also etched. For this reason, when a resist film is laminated on a fluorine-added carbon film as in general etching, the selectivity of the two films approximates, so the thickness of the resist film must be greater than that of the fluorine-added carbon film. In addition, when the resist film is removed by ashing with oxygen plasma, the fluorine-added carbon film is also ashed.

  For this reason, when a fluorine-added carbon film is used, it is necessary to laminate a hard mask thin film that functions as a mask during etching on the fluorine-added carbon film. As a material for the thin film for the hard mask, an SiO2 film or a silicon nitride film (Si3N4 film) is known (Patent Document 1). In order to suppress an increase in the relative dielectric constant of the entire interlayer insulating film. The present inventors pay attention to an oxygen-added silicon carbide (SiCO) film having a low relative dielectric constant as a material for a hard mask. This SiCO film is, for example, a silicon carbide film containing about 20 atomic% of oxygen. For example, a silicon nitride film or a silicon carbide film which is a barrier layer for preventing copper diffusion when copper serving as a wiring is embedded in an interlayer insulating film In addition, there is an advantage such as high resistance to CMP when removing copper on the interlayer insulating film by polishing copper called CMP (Chemical Mechanical Polishing) after embedding copper, and effective. It is thought that it is a perfect film.

  However, when the SiCO film is formed on the fluorine-added carbon film, for example, the vapor (gas) of an organic source such as trimethylsilane and oxygen gas are turned into plasma, and at this time, active species of oxygen are contained in the fluorine-added carbon film. It reacts with carbon and is released as carbon dioxide (CO2). With regard to oxygen in particular, it is known that the fluorine-added carbon film has a large amount of degassing from the fluorine-added carbon film when heated in an atmosphere in which oxygen is present at a very low level of 1 ppm, for example. As a result, the denseness of the part deteriorates, resulting in a problem of poor adhesion to the SiCO film.

  Further, when a hard mask SiO2 film is formed on the fluorine-added carbon film instead of the SiCO film, for example, vapor (gas) of organic source such as tetraethylorthosilicate (Si (OC2H5) 4) and oxygen Since the gas is converted into plasma, the oxygenated gas causes a problem similar to that described above.

Japanese Patent Laid-Open No. 10-144676 (paragraphs 0008, 0029, 0031)

  The present invention has been made in view of such circumstances, and an object thereof is to include an active species of oxygen on a fluorine-added carbon film in a method of manufacturing a semiconductor device including a fluorine-added carbon film as an insulating film. An object of the present invention is to provide a method for obtaining high adhesion between a thin film and a fluorine-added carbon film when forming a thin film for a hard mask using plasma.

  The method of manufacturing a semiconductor of the present invention includes a step of forming an insulating film made of a fluorine-added carbon film on a substrate, and then exposing the surface of the substrate to plasma obtained by converting nitrogen gas into plasma, so that the fluorine Nitriding the surface portion of the additive carbon film, and then exposing the surface of the substrate to plasma containing active species of silicon and oxygen to form a thin film for a hard mask containing silicon and oxygen, It is characterized by including. The plasma containing active species of silicon and oxygen is, for example, plasma obtained by converting an organic compound gas containing silicon and oxygen gas into plasma. The thin film for the hard mask is, for example, an oxygen-added silicon carbide film or a silicon dioxide film.

  A more specific method of the present invention includes a step of forming a resist film in a predetermined pattern on a surface of the thin film for the hard mask, and then etching the thin film for the hard mask with plasma, and the pattern is formed on the thin film. The method further includes a step of obtaining a hard mask by forming a pattern corresponding to the above, and a step of etching the fluorine-added carbon film by plasma using the hard mask.

  According to the present invention, when a thin film containing silicon and oxygen, for example, a SiCO film or a SiO2 film is used as a hard mask, the surface portion of the fluorine-added carbon film is nitrided and then a SiCO film or a SiO2 film is formed thereon. Since the film is formed, the penetration of the active species of oxygen used when forming the thin film for the hard mask into the film is suppressed by the nitriding region on the surface portion of the fluorine-added carbon film. Accordingly, the amount of degassing from the fluorine-added carbon film is reduced in the subsequent process, and as a result, the adhesion between the fluorine-added carbon film and the SiCO film or SiO2 film as the hard mask is increased.

  An embodiment of a method for manufacturing a semiconductor device according to the present invention is a substrate for manufacturing a multilayer wiring structure, wherein an nth (n is an integer of 1 or more) wiring layer made of copper wiring in an insulating film The case where the (n + 1) th wiring layer is formed on the upper side will be described as an example. FIG. 1A schematically shows a surface structure of a substrate 1 which is a semiconductor wafer, for example, in which a copper wiring 11 which is an nth wiring layer is formed in an insulating film 10. In this embodiment, a film forming gas of a compound containing carbon and fluorine, for example, C5F8 gas is converted into plasma, and the atmosphere in which the substrate 1 is placed is changed to a plasma atmosphere. As shown in FIG. 1B, an interlayer insulating film made of a fluorine-added carbon film 20 is formed to a thickness of 200 nm, for example, on the surface of the substrate 1.

  Next, N2 (nitrogen) gas is converted into plasma, and the atmosphere in which the substrate 1 is placed is changed to a plasma atmosphere, so that active species generated from the N2 gas enter the fluorine-added carbon film 20 of the substrate 1. . In this case, the process pressure is set to, for example, 10 to 100 Pa, and the wafer temperature is set to, for example, 0 to 100 ° C., preferably 25 to 80 ° C. The time required for the nitriding treatment is, for example, several seconds to several tens of seconds. By performing such nitriding treatment, as will be apparent from the examples described later, nitrogen penetrates into the fluorine-added carbon film 20, but in a region deeper than about 20 nm from the film surface, the amount of nitrogen penetration Accordingly, as shown in FIG. 1C, only the surface portion from the film surface to about 20 nm is substantially nitrided, and a nitrogen-added fluorine-added carbon film (hereinafter abbreviated as “CFN film”) 21 is formed. Will be formed.

  After nitriding the surface of the fluorine-added carbon film 20 in this way, a first hard mask SiCO film 22 to be used as a hard mask in a subsequent process is formed. As a raw material gas for forming the SiCO film 22, a gas of an organic compound containing silicon, for example, trimethylsilane (SiH (CH3) 3) gas and oxygen (O2) gas is used, and the trimethylsilane gas and oxygen gas are used as plasma. As a result, the SiCO film 22 is formed on the surface of the CFN 21 film by the active species of silicon, carbon, and oxygen contained in the plasma, as shown in FIG.

  Next, as shown in FIG. 2B, for example, a SiO 2 film 23 for a second hard mask made of a material different from the SiCO is formed on the surface of the SiCO film 22. As the source gas for forming the SiO2 film 23, for example, vapor (gas) of an organic source such as tetraethylorthosilicate (Si (OC2H5) 4) and oxygen gas are used. By converting the tetraethylorthosilicate gas and the oxygen gas into plasma, the SiO2 film 23 is formed on the surface of the SiCO film 22 by the active species of silicon and oxygen contained in the plasma. Thereafter, a resist film is formed on the SiO2 film 23, a pattern is formed, and the SiO2 film 23 is etched using the resist mask to obtain a second hard mask having a pattern corresponding to the pattern ( FIG. 2 (c)).

  Thereafter, a resist film 24 is formed on the surface of the wafer, and a pattern narrower than the above pattern is formed (FIG. 2D), and the SiCO film 22 is made of, for example, a halide using the resist mask 24. The first hard mask is obtained by etching with plasma containing active species, and the CFN film 21 and the fluorine-added carbon film 20 (specifically, the CFN film 21 on the surface portion is included using the first hard mask). Therefore, in the description relating to etching, the CFN film 21 and the fluorine-added carbon film 20 will be described below) by, for example, oxygen plasma (FIG. 2E). In addition, although a barrier layer and a hard mask layer actually exist on the surface of the fluorine-added carbon film 10 and the copper wiring 11 as the underlying layer, they are omitted for convenience in the drawing. Further, when the CFN film 21 and the fluorine carbon film 20 exposed by the first hard mask are etched by oxygen plasma, the selection ratio between the resist film 24 and the fluorine-added carbon film 20 is approximated as described above. At the same time, the resist film 24 is also removed.

  Further, using the second hard mask made of the SiO2 film 23, the SiCO film 22 is etched, and further the CFN film 21 and the fluorine-added carbon film 20 are etched, so that the width is smaller than the recess formed by the previous etching. A wide concave portion is formed (FIG. 3A). A narrow concave portion corresponds to a via hole, and a wide concave portion corresponds to a wiring embedding region of a circuit in the layer. Thereafter, as shown in FIG. 3B, for example, copper 25 which is a wiring metal is embedded, and copper 25 other than the recesses is removed by polishing called, for example, CMP (Chemical Mechanical Polishing) to form the copper wiring 25. (FIG. 3C). Thereafter, although not shown, a barrier layer such as a SiC layer is formed on the surface.

  According to the above-described embodiment, the surface portion of the fluorine-added carbon film 20 is subjected to nitriding treatment by exposing it to a plasma atmosphere obtained by converting nitrogen gas into plasma, and the CFN film 21 is formed on the film surface. Invasion of active species of oxygen gas used when forming the film 22 into the film is suppressed by the nitride region (CFN film 21) on the surface of the fluorine-added carbon film 20. Therefore, in the subsequent step, degassing from the fluorine-added carbon film 20 that is considered to be the release of carbon dioxide due to the reaction between oxygen and carbon in the fluorine-added carbon film 20 is suppressed, and the surface portion of the fluorine-added carbon film 20 is dense. As a result, the adhesion between the fluorine-added carbon film 20 as an interlayer insulating film and the SiCO film 22 as a hard mask thin film is enhanced. Further, since degassing of the fluorine-added carbon film 20 is suppressed, deterioration of the film quality of the fluorine-added carbon 20 itself can be prevented. From this, it can be understood that the CFN film 21 plays two roles of a protective layer and an adhesion layer that protects against active species of oxygen gas.

  In FIG. 2, the SiCO film 22 is formed on the CFN film 21, but the SiO2 film 23 is formed on the CFN film 21, and the SiCO film 22 is formed thereon. You may make it replace the material of a hard mask and a 2nd hard mask.

  Next, an example of a semiconductor manufacturing apparatus for carrying out the above-described semiconductor device manufacturing method will be described. FIG. 4 is a view showing a semiconductor manufacturing apparatus for performing the steps up to the above-described FIG. 2B, that is, the steps of forming the fluorine-added carbon film 20, the CFN film 21, the SiCO film 22, and the SiO 2 film 23. is there. In FIG. 4, 31 and 32 are carrier chambers into which a carrier C, which is a wafer transfer container, is carried from the atmosphere side through the gate door GT, 33 is a first transfer chamber, and 34 and 35 are preliminary vacuums. 36 is a second transfer chamber, which has an airtight structure and is partitioned from the atmosphere side. The second transfer chamber 36 and the preliminary vacuum chambers 34 and 35 are in a vacuum atmosphere, but the carrier chambers 31 and 32 and the first transfer chamber 33 may be in an inert gas atmosphere. Reference numeral 37 denotes a first conveying means, and 38 denotes a second conveying means. Further, in the second transfer chamber 36, a film forming apparatus 40 for forming the fluorinated carbon film 20 as an interlayer insulating film, and an annealing apparatus 41 for nitriding the fluorinated carbon film 20 with nitrogen plasma. The film forming apparatus 50 for forming the SiCO film 22 and the SiO2 film 23 which are thin films for hard masks are connected in an airtight manner.

  In the semiconductor manufacturing apparatus of FIG. 4, the substrate 1 in the carrier C is transported by a path of, for example, the first transport unit 37 → the preliminary vacuum chamber 34 (or 35) → the second transport unit 38 → the film forming apparatus 40. The film formation apparatus 40 forms the fluorine-added carbon film 20. Then, the substrate 1 is carried into the annealing apparatus 41 via the second conveying means 38, the surface portion of the fluorine-added carbon film 20 is nitrided to form the CFN film 21, and then carried into the film forming apparatus 50. Then, the SiCO film 22 and the SiO 2 film 23 are formed on the CFN film 21. Thereafter, the substrate 1 is returned into the carrier C through a path reverse to that described above.

  Here, the film forming apparatus 40 for forming the fluorinated carbon film 20 will be briefly described with reference to FIGS. In the figure, reference numeral 61 denotes a processing container (vacuum chamber), and 62 denotes a mounting table provided with temperature control means. The mounting table 62 is connected to a high frequency power source 63 for bias of 13.56 MHz, for example.

  A first gas supply unit 64 made of, for example, alumina having a substantially circular planar shape is provided on the upper portion of the processing container 61 so as to face the mounting table 62. A number of first gas supply holes 65 are formed on the surface of the gas supply unit 64 facing the mounting table 62. The first gas supply hole 65 communicates with the first gas supply path 67 through the gas flow path 66. The first gas supply path 67 is connected to a plasma gas supply source such as argon (Ar) gas or krypton (Kr) gas which is a plasma gas.

  Further, a second gas supply unit 68 made of a conductor having a substantially circular planar shape is provided between the mounting table 62 and the first gas supply unit 64, for example. A number of second gas supply holes 69 are formed on the surface of the portion 68 facing the mounting table 62. As shown in FIG. 6, for example, a lattice-like gas flow channel 71 communicating with one end side of the second gas supply hole 69 is formed inside the second gas supply unit 68. One end side of the second gas supply path 72 is connected to the path 71. The second gas supply unit 68 has a large number of openings 73 so as to penetrate the second gas supply unit 68. The opening 73 is for allowing plasma to pass through the space below the second gas supply unit 68 and is formed, for example, between adjacent gas flow paths 71.

  Here, the second gas supply unit 68 is connected to a supply source (not shown) of the C5F8 gas, which is the above-described source gas, via the second gas supply path 72, and the second gas supply path 72 is connected to the second gas supply path 68. Through the gas flow passage 71 and is uniformly supplied to the space below the second gas supply portion 68 through the gas supply hole 69. Reference numeral 74 denotes an exhaust pipe, which is connected to the vacuum exhaust means 75.

  A cover plate 76 made of a dielectric material such as alumina is provided on the upper side of the second gas supply unit 68, and the upper side of the cover plate 76 is in close contact with the cover plate 76. An antenna portion 77 is provided. As shown in FIG. 7, the antenna portion 77 is provided with a flat antenna body 78 having a circular planar shape that opens on the lower surface side, and the lower surface side of the antenna body 78 so as to close the opening portion. A disk-shaped planar antenna member (slit plate) 79 having slits is provided. The antenna main body 78 and the planar antenna member 79 are made of a conductor, and form a flat hollow circular waveguide. ing.

  Further, between the planar antenna member 79 and the antenna body 78, there is provided a slow phase plate 81 made of a low-loss dielectric material such as alumina, silicon oxide, silicon nitride or the like. This retardation plate 81 is for shortening the wavelength of the microwave to shorten the guide wavelength in the waveguide. In this embodiment, the antenna main body 78, the planar antenna member 79, and the slow phase plate 81 constitute a radial line slit antenna.

  The antenna section 77 thus configured is connected to an external microwave generating means 83 via a sealing member (not shown) so that the planar antenna member 79 is in close contact with the cover plate 76, and has a frequency of 2.45 GHz, for example. An 8.4 GHz microwave is supplied. At this time, the waveguide 82 A outside the coaxial waveguide 82 is connected to the antenna body 78, and the center conductor 82 B is connected to the planar antenna member 79 through an opening formed in the slow phase plate 81. .

  The planar antenna member 79 is made of, for example, a copper plate having a thickness of about 1 mm, and a plurality of slits 84 for generating, for example, circular flat waves are formed as shown in FIG. The slit 84 is formed, for example, concentrically or spirally along the circumferential direction, with a pair of slits 84A and 84B arranged in a substantially T-shape and slightly spaced apart. Note that the slits 84 may be arranged so as to be slightly separated from each other in an approximately eight-character shape. Thus, since the slits 84A and 84B are arranged so as to be substantially orthogonal to each other, circularly polarized waves including two orthogonal polarization components are radiated. At this time, by arranging the slit pairs 84 </ b> A and 84 </ b> B at an interval corresponding to the wavelength of the microwave compressed by the retardation plate 81, the microwave is radiated by the planar antenna member 79 as a substantially planar wave.

  Subsequently, an example of a film forming process of the fluorine-added carbon film 20 performed by the film forming apparatus 40 will be described. First, the substrate 1 is carried into the processing container 61 and placed on the mounting table 62. Subsequently, the inside of the processing vessel 61 is evacuated to a predetermined pressure, and a plasma gas, for example, Ar gas is supplied from the first gas supply unit 64 into the processing vessel 61 at a predetermined flow rate, for example, 200 sccm. The C5F8 gas is supplied into the processing container 61 from the second gas supply unit 68 at a predetermined flow rate, for example, 100 sccm. Then, the inside of the processing container 61 is maintained at a process pressure of 10.6 Pa (80 mTorr), for example, and the wafer temperature is set to 380 ° C.

  On the other hand, when a high frequency (microwave) of 2.45 GHz and 3000 W is supplied from the microwave generation means 83, the microwave propagates in the coaxial waveguide 82 in the TM mode or the TE mode, and the planar antenna of the antenna unit 77. While reaching the member 79 and propagating radially from the central portion of the planar antenna member 79 toward the peripheral region via the central conductor 82B of the coaxial waveguide, microwaves from the slit pairs 84A and 84B are covered by the cover plate. 76, the gas is discharged toward the processing space below the first gas supply unit 64 via the first gas supply unit 64.

  At this time, since the slit pairs 84A and 84B are arranged as described above, the circularly polarized wave is uniformly emitted over the plane of the planar antenna member 79, and the electric field density in the space on the lower side is made uniform. . The microwave energy excites high density and uniform plasma of, for example, argon gas in the space between the first gas supply unit 64 and the second gas supply unit 68. On the other hand, the C5F8 gas blown out from the second gas supply unit 68 is activated in contact with the plasma flowing from the upper side through the opening 73, and the active species generated from the C5F8 gas are activated on the surface of the substrate 1. Then, an interlayer insulating film made of the fluorine-added carbon film 20 is formed.

  Here, as the rare gas for generating plasma, Ar gas is used in the above example. However, other rare gases such as helium (He) gas, neon (Ne) gas, krypton (Kr) gas, and xenon (Xe) are used. Gas or the like can be used. Further, the use of the fluorine-added carbon film is not limited to the interlayer insulating film, but may be other insulating films. The raw material gas for the fluorinated carbon film is not limited to C5F8 gas, but CF4 gas, C2F6 gas, C3F8 gas, C3F9 gas, C4F8 gas, or the like may be used.

  In this example, the annealing apparatus 41 has the same configuration as the film forming apparatus 40 described above, and an N 2 gas supply source is connected to the first gas supply path 67. In the annealing apparatus 41, the substrate 1 on which the fluorine-added carbon film 20 has already been formed is carried into a processing vessel 61 (described using the same reference numerals as those of the film forming apparatus 40 for convenience), and the first gas The N2 gas is supplied from the supply unit 64 into the processing vessel 61 at a predetermined flow rate, for example, 10 to 100 sccm, and the inside of the processing vessel 61 is maintained at, for example, a process pressure of 33.3 to 66.5 Pa (250 to 500 mTorr). Is set to 25-80 ° C.

  On the other hand, by supplying a high frequency (microwave) of 2.45 GHz and 500 W from the microwave generating means 83, N2 gas is excited by the energy of the microwave to generate plasma. The excited species of nitrogen (active species) in the plasma penetrate into the fluorine-added carbon film 20, but as can be seen from the examples described later, most of the active species of nitrogen is the surface from the film surface to about 20 nm, for example. Therefore, only the substantial surface portion is nitrided and the CFN film 21 is formed. At this time, the time for exposing the surface of the fluorine-added carbon film 20 to a nitrogen plasma atmosphere is, for example, about 5 to 10 seconds. An apparatus for performing such nitriding treatment is not limited to the same structure as the film forming apparatus 40 shown in FIG. 5, and may be a parallel plate type plasma processing apparatus, for example. In this example, since the process temperature of the nitriding process is lower than the film formation temperature of the fluorine-added carbon film 20, the apparatus is provided separately from the viewpoint of throughput in consideration of the time required for raising and lowering the temperature. Of course, both processes may be performed by a common apparatus.

  As the film forming apparatus 50 for forming the SiCO film 22 and the SiO2 film 23, an apparatus having the same configuration as the film forming apparatus 40 described above is used in this example, and a plasma gas such as Ar gas is provided in the first gas supply path 67. And a supply source of oxygen gas are connected to each other, and a supply source of trimethylsilane gas and a supply source of tetraethylorthosilicate gas are connected to the second gas supply path 72. In the film forming apparatus 50, the substrate 1 on which the CFN film 21 has already been formed is carried into the processing container 61, and then the inside of the processing container 61 is evacuated to a predetermined pressure. Then, the SiCO film 22 that is the first hard mask is formed. First, a plasma gas such as Ar gas and oxygen gas is supplied from the first gas supply unit 64 into the processing container 61. The trimethylsilane gas is supplied into the processing vessel 61 from the second gas supply unit 68 as a raw material gas supply unit at a predetermined flow rate, for example, 200 sccm, while being supplied at a predetermined flow rate, for example, 1000 sccm and 200 sccm. Then, the inside of the processing container 61 is maintained at a process pressure of 33.3 Pa (250 mTorr), for example, and the wafer temperature of the mounting table 62 is set to 380 ° C.

  On the other hand, by supplying a high frequency (microwave) of 2.45 GHz and 1500 W from the microwave generating means 83, Ar gas is converted into plasma by the energy of the microwave, and oxygen gas and trimethylsilane gas are excited by this plasma. A SiCO film 22 as a first hard mask is formed on the CFN film 21. Subsequently, the SiO2 film 23, which is the second hard mask, is formed. The film formation process is performed, for example, by switching from trimethylsilane gas to tetraethylorthosilicate gas and processing the gas from the second gas supply unit 68. The film forming process is the same as described above except that the container 61 is supplied at, for example, 100 sccm. Therefore, Ar gas is converted into plasma by microwave energy, and oxygen gas and tetraethylorthosilicate gas are excited by this plasma, and a SiO 2 film 23 as a second hard mask is formed on the SiCO film 22. The apparatus for forming the SiCO film 22 and the SiO2 film 23 may be shared with the apparatus 40 for forming the fluorine-added carbon film 20, or may be shared with the annealing apparatus 41 for performing nitriding treatment. All these film forming processes may be performed by one apparatus, for example, the film forming apparatus 40.

A. Deposition of fluorinated carbon film and thin film for hard mask (Example 1)
In the semiconductor manufacturing apparatus, a fluorine-added carbon film having a thickness of 120 nm was formed on a silicon bare wafer as a substrate using the film forming apparatus 40 shown in FIG. Regarding the process conditions, the microwave power is 3000 W, the process pressure is 10.6 Pa (80 mTorr), the wafer temperature is 380 ° C., the Ar gas and the second gas supply path 72 are supplied to the first gas supply path 67. The flow rates of the C5F8 gas to be supplied were set to 200 sccm and 100 sccm, respectively.

  Subsequently, the fluorine-added carbon film was subjected to nitriding treatment (annealing) using plasma obtained by converting N2 gas into plasma using an annealing device 41. Regarding the process conditions, the microwave power was set to 500 W, the process pressure was set to 33.3 Pa (250 mTorr), the wafer temperature was set to 80 ° C., the N 2 gas flow rate was set to 50 sccm, and the processing time was set to 5 seconds.

  Next, a SiCO film having a thickness of 50 nm was formed on the wafer using the film forming apparatus 50. Regarding the process conditions, the microwave power is set to 1500 W, the process pressure is set to 33.3 Pa (250 mTorr), the wafer temperature is set to 380 ° C., and trimethylsilane gas, Ar gas, and oxygen gas are flowed at 200 sccm, 1000 sccc, and 200 sccm, respectively. Each supplied.

(Comparative Example 1)
The film formation was performed in the same manner as in Example 1 except that the surface of the fluorine-added carbon film was not nitrided (the CFN film was not formed) and the SiCO film was directly formed on the fluorine-added carbon film. went.

B. Consideration of thin film adhesion The substrates of Example 1 and Comparative Example 1 were heated to 400 ° C. in a vacuum atmosphere and allowed to stand for 30 minutes. When the surface of these substrates was visually observed and the state of film peeling was examined by applying a tape, Comparative Example 1 showed many discoloration areas based on the occurrence of bubbles from the film, and most of them The film was peeled off. On the other hand, in Example 1, no discoloration region as in Comparative Example 1 was observed, and no film peeling occurred in the tape test. Therefore, it is understood that the adhesion of the SiCO film to the fluorine-added carbon film is increased by interposing the CFN film between the fluorine-added carbon film and the SiCO film.
C. Measurement of nitrogen concentration in fluorine-added carbon film (including CFN film) Secondary ion mass spectrometry (SIMS) of a fluorine-added carbon film (laminated body of fluorine-added carbon film 20 and CFN film 21) whose surface is nitrided : Secondary Ion Mass Spectroscopy) conducted mass analysis of secondary ions released when the surface of the stack was irradiated with an ion beam, and investigated the nitrogen concentration profile in the stack using the secondary ion intensity as an index. . The results are shown in FIG. 8, and FIG. 8 also shows the concentration profiles of silicon, carbon, and fluorine. In FIG. 8, the vertical axis represents the secondary ion intensity (count / sec), and the horizontal axis represents the film depth (nm).

  As can be seen from FIG. 8, nitrogen penetrates deeply into the fluorine-added carbon film, in this example, to the lower surface of the fluorine-added carbon film having a thickness of 100 nm, but in a region deeper than about 20 nm from the film surface, nitrogen penetrates. The amount is extremely small. Therefore, it is understood that most of the active species of nitrogen are almost trapped between the film surface and about 20 nm, and only the surface portion is nitrided. Since only the surface portion is nitrided in this way, the intrusion of the active species of oxygen gas used when forming the SiCO film into the film is caused by the nitride region (CFN film) on the surface portion of the fluorine-added carbon film. It seems to be suppressed.

By the way, if the thermal energy is too large when nitriding the fluorine-added carbon film, there is a risk that many of the active species of nitrogen will penetrate deep into the film, and for this purpose, the process temperature during the nitriding process is not so high. It is desirable. When nitrogen penetrates deep into the film, the C—F bond of the fluorine-added carbon film is broken, and C or CF is desorbed from the film in a high-temperature process in the subsequent process. Therefore, a hard mask such as a SiCO film There is a concern that the adhesiveness will deteriorate. Therefore, one way to reduce thermal energy is to lower the process temperature. If the process pressure is 33.3 to 66.5 Pa (250 to 500 mTorr), the process temperature is set to 100 ° C. or lower. By doing so, it is understood that a satisfactory profile can be obtained for the nitrogen concentration. However, in the present invention, such process conditions are not important, and as a result, any profile that does not cause the influence of adhesion due to C or CF desorption may be used. When an ADEPT1010 is used as an ionic strength measuring device; when the ADEPT1010 is used, the secondary ionic strength is 2 × 10 ² or less at a depth of the fluorinated carbon film of 50 nm. The nitrogen concentration at the surface is sufficient to prevent the active species of oxygen from entering, and the nitrogen concentration up to, for example, 20 nm near the surface is reduced so as to suppress adverse effects caused by the penetration of nitrogen into the film. It is considered that a profile that rapidly decreases in the depth direction may be used. Specifically, if the ratio of (secondary ion intensity at a depth of 20 nm) / (secondary ion intensity at the surface) is 30% or less, preferably 20 or less, desorption of C or CF is not a problem. I believe.

It is explanatory drawing which shows the mode of the film-forming of the fluorine addition carbon film | membrane and the nitriding process of the surface part in embodiment of this invention. It is process drawing which shows a mode that the semiconductor device is manufactured in steps in embodiment of this invention. It is process drawing which shows a mode that the semiconductor device is manufactured in steps in embodiment of this invention. It is a top view which shows an example of the vacuum processing system used for embodiment of this invention. It is a vertical side view which shows an example of the plasma film-forming apparatus used for embodiment of this invention. It is a top view which shows the 2nd gas supply part used for said plasma film-forming apparatus. It is a perspective view which shows the antenna part used for said plasma film-forming apparatus in a partial cross section. It is a characteristic view which shows each density | concentration of nitrogen, silicon, carbon, and fluorine in a fluorine addition carbon film.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Substrate 10 Fluorine-added carbon film 11 Copper 20 Fluorine-added carbon film 21 CFN film 22 SiCO film 23 SiO2 film 24 Resist film 25 Copper 40 Film formation apparatus 41 Annealing apparatus 50 Film formation apparatus 61 Processing vessel 64 First gas supply section 67 First gas supply path 68 Second gas supply section 72 Second gas supply path 77 Antenna section 78 Antenna body

Claims (4)

Forming an insulating film made of a fluorine-added carbon film on the substrate;
Next, the surface of the substrate is exposed to plasma obtained by converting nitrogen gas into plasma, and the surface portion of the fluorine-added carbon film is nitrided;
And thereafter exposing the surface of the substrate to a plasma containing active species of silicon and oxygen to form a thin film for a hard mask containing silicon and oxygen. .   2. The method of manufacturing a semiconductor device according to claim 1, wherein the plasma containing active species of silicon and oxygen is plasma obtained by converting an organic compound gas containing silicon and oxygen gas into plasma.   3. The method of manufacturing a semiconductor device according to claim 1, wherein the thin film for the hard mask is an oxygen-added silicon carbide film or a silicon dioxide film. Forming a resist film in a predetermined pattern on the surface of the hard mask thin film;
Next, etching the hard mask thin film with plasma, forming a pattern corresponding to the pattern on the thin film to obtain a hard mask;
4. The method of manufacturing a semiconductor device according to claim 1, further comprising: etching the fluorine-added carbon film with plasma using the hard mask. 5.

Images