Patent Application
20060068104
The present invention relates to thin-film formation on a semiconductor substrate, and more particularly, to a technique for fabricating a barrier film for preventing diffusion of metal species, without damaging underlying layers.
Along with recent and continuing demand for high-performance semiconductor devices, highly integrated semiconductor devices are being developed with further miniaturization scales. The design rule of metal processes is shifting from 0.13 μm to 0.10 μm or even smaller. Concerning the material of interconnects, conventionally used aluminum (Al) is being replaced with copper (Cu), which material has a lower electric resistance and less influence of wiring delay.
For this reason, the combination of copper (Cu) film formation and fine-pitch wiring technology becomes more important in the recent technology for fabricating high-performance semiconductor devices.
When employing copper (Cu) interconnects, it is necessary to form a Cu-diffusion barrier film to prevent copper species from diffusing into the surrounding dielectric (or insulating) film. Such a diffusion barrier film requires high film quality with less impurity content and satisfactory crystal orientation. It is also required for the diffusion barrier film to achieve high coverage on the minute patterns.
Atomic layer deposition (ALD) is one of film formation techniques satisfying the above-described requirements. With atomic layer deposition, one of multiple types of source gases is supplied alternately onto the substrate to form an atom layer or a molecular layer one by one through adsorption of the source gas onto the substrate surface. By repeating the layer-by-layer film formation (atomic layers or molecular layers), a thin film with a predetermined thickness can be fabricated.
To be more precise, the first source gas is supplied onto the substrate to form the adsorbed layer of the first material. Then the second source gas is supplied onto the substrate to cause the second gas to react with the first material. Since the second source gas reacts with the first source gas after adsorption onto the substrate, the temperature in film formation can be lowered. The amount of impurities in the film is smaller, and a high-quality thin film can be obtained. In addition, high coverage can be achieved over minute patterns, while preventing undesirable voids from being generated. Such voids are generated in the conventional CVD method when the source gas is reactively consumed over the holes.
High refractory metals or nitrides thereof are typically used as the copper (Cu) diffusion barrier film. It is currently known that titanium nitride (TiN) film, tantalum (Ta) film. tantalum nitride (TaN) film, Ta/TaN layered film, tungsten (W) film, tungsten nitride (WN) film, and W/WN layered film can be employed as the copper diffusion barrier film.
For example, when forming a titanium nitride (TiN) film, the first source gas is a chemical compound containing titanium (Ti), such as TiCl4, and the second source gas is a reducing gas containing nitrogen, such as plasma-activated NH3. The reason why plasma-activated NH3 is used is to reduce the impurity density in the TiN film.
With the layer-by-layer deposition, by supplying the first source gas to form the adsorbed layer on the substrate and then supplying the second source gas, a high-quality TiN film with less impurities and lower resistance can be formed.
The above-described background technologies are disclosed in, for example, patent laid-open publications JP 6-89873A and JP 7-252660A. They are also disclosed in non-patent publications, such as
However, a problem arises when employing atom-layer level or molecular-layer level film deposition to form a copper (Cu) diffusion barrier film. That is, the underlayers existing beneath the Cu diffusion barrier layer are damaged.
For example, when forming copper (Cu) interconnects by dual damascene, lower-level copper (Cu) interconnects or tungsten (W) interconnects exist under the copper (Cu) diffusion barrier film. There is also interlayer dielectric film existing around the upper-level Cu interconnects.
If a TiN film is formed as the Cu diffusion barrier film, the interlayer dielectric film is damaged by the plasma process. Since the second source gas NH3 is plasma-activated, the ions and radicals dissociated from NH3 damage the dielectric film. In particular, low dielectric constant films are often used as the dielectric film in these years. If such low dielectric constant films are damaged by ions and/or radicals, permittivity of the dielectric film becomes high.
The lower-level Cu interconnects are also damaged by the fabrication process of the Cu diffusion barrier film. Again, if a TiN film is formed as the Cu diffusion barrier film, the lower-level Cu interconnects are corroded by halogen and the copper surface is roughened because metal halide TiCl4 gas is used as the first source gas.
Therefore, it is an object of the present invention to solve the above-described problems in the prior art, and to provide a novel and useful film formation technique that does not damage underlayers when forming a Cu-diffusion barrier film, while achieving high film quality.
In particular, it is an object of the invention to provide a film fabrication method for fabricating a high-quality Cu-diffusion barrier film with less impurities contained therein, without damaging the underlying dielectric film.
It is also an object of the invention to provide a film fabrication method for fabricating a high-quality Cu-diffusion barrier film with less impurities contained therein, without damaging the underlying copper film.
To achieve the objects, in the first aspect of the invention, a film fabrication method for forming a film over a substrate in a processing chamber is provided. The film fabrication method includes a first film formation process and a second film formation process. In the first film formation process, (a) a first step of supplying a first source gas containing a metal into a chamber and then removing the first source gas from the chamber, and (b) a second step of supplying a second source gas containing hydrogen or a hydrogen compound into the chamber and then removing the second source gas from the chamber, are repeated a predetermined number of times. In the second film formation process,
(c) a third step of supplying the first source gas into the chamber and removing the first gas from the chamber, and (d) a fourth step of supplying a plasma-activated third source gas containing hydrogen or a hydrogen compound into the chamber and removing the third source gas from the chamber, are repeated a predetermined number of times.
In the second aspect of the invention, a film fabrication method for forming a film over a substrate in a processing chamber includes a first film formation process and a second film formation process. In the first film formation process, (a) a first step of supplying a first source gas containing an organic-metal compound and without containing a halogen element into the chamber and then removing the first source gas from the chamber, and (b) a second step of supplying a second source gas containing hydrogen or a hydrogen compound into the chamber and then removing the second source gas from the chamber, are repeated a predetermined number of times. In the second film formation process, (c) a third step of supplying a third source gas containing a metal halide compound into the chamber and then removing the third gas from the chamber, and (d) a fourth step of supplying a plasma-activated fourth source gas containing hydrogen or a hydrogen compound into the chamber and then removing the fourth source gas from the chamber, are repeated a predetermined number of times.
In the third aspect of the invention, a film fabrication method for forming a film over a substrate in a processing chamber includes a first film formation process and a second film formation process. In the first film formation process, (a) a first step of supplying a first source gas containing an organic-metal compound and without containing a halogen element into the chamber and then removing the first source gas from the chamber, and (b) a second step of supplying a second source gas containing hydrogen or a hydrogen compound into the chamber and then removing the second source gas from the chamber, are repeated a predetermined number of times. In the second film formation process, (c) a third step of supplying a third source gas containing a metal halide compound into the chamber and then removing the third gas from the chamber, and (d) a fourth step of supplying a plasma-activated fourth source gas containing hydrogen or a hydrogen compound into the chamber and then removing the fourth source gas from the chamber, are repeated a predetermined number of times.
In these methods, the first film formation process is performed to form a first copper-diffusion barrier film, and the second film formation process is performed to form a second copper-diffusion barrier film. Thus, a layered barrier film for preventing copper diffusion is obtained.
With the above-described methods, the Cu-diffusion barrier film can be formed without damaging underlying layers.
The resultant Cu-diffusion barrier film has a satisfactory film quality with less impurity content and good crystal orientation. In addition, high coverage over a minute pattern can be achieved.
In the fourth aspect of the invention, a film deposition apparatus is provided. The apparatus 42 includes:
(a) a processing chamber;
(b) a stage configured to hold a substrate to be processed in the processing chamber;
(c) a first gas supply system configured to supply a first source gas or a third source gas into the processing chamber;
(d) a second gas supply system configured to supply a second source gas or a fourth source gas into the processing chamber, independently from the first gas supply system; and
(e) plasma excitation means configured to excite the second source gas or the fourth source gas into plasma.
With this apparatus, a layered film, such as a layered copper-diffusion barrier film, can be formed without damaging underlying layers.
Other objects, features, and advantages of the invention will become more apparent from the following detailed description when read in conjunction with the accompanying drawings, in which
With the present invention, a high-quality Cu-diffusion barrier film can be formed on a semiconductor substrate layer by layer (on the atomic layer basis or the molecular layer basic) the following steps. The first source gas is supplied onto the substrate held in a chamber to form a adsorbed layer on the substrate, and unreacted first source gas is removed from the chamber. Then, the second source gas is supplied onto the substrate in the chamber to cause reaction, and unreacted second source gas is removed from the chamber.
In this manner, film formation is carried out layer by layer (on the atomic or molecular layer basis), and a high-quality film with less impurities and low electric resistance can be obtained. When forming a Cu-diffusion barrier film over minute patterns, good coverage can be achieved with few voids generated due to reactive consumption of the source gas over the holes. In addition, the film fabrication method of the invention is superior in film quality and uniformity of film thickness on the processed substrate. In addition, the film fabrication method of the invention is advantageous because of the lowered process temperature, especially when a film likely to deteriorate at a high temperature (at or above 400° C.), such as a low dielectric constant film, is used in underlayers. The film fabrication method of the invention may be referred to as atomic layer deposition (ALD).
Actual examples of fabrication of a Cu-diffusion barrier film according to the preferred embodiment are now described below in conjunction with the attached drawings.
In
Then, as shown in
Then, as shown in
Because in the step shown in
In contrast, plasma-activated NH3 gas contains radicals, such as N*, H*, or NH*, which radicals are likely to etch the dielectric film 1. In addition, ions existing in the plasma-activated NH3 gas give physical sputtering damage to the dielectric film. The first step shown in
Silicon dioxide film is conventionally used as the dielectric film 1; however, using low dielectric constant films with permittivity at or below 4 (which permittivity is lower as compared with ordinary silicon dioxide) has become more popular in the semiconductor industry in these years. Such low dielectric constant films are easy to be etched chemically or physically. The film quality is also likely to change, causing the permittivity to increase. When a porous film with a number of pores formed in the film to lower the permittivity is used as the dielectric film, the film is more likely to be damaged because of insufficient strength.
For the above-described reason, the film fabrication method of the present invention is more advantageous when forming a Cu-diffusion barrier film over a low dielectric constant film more likely to be damaged as compared with silicon dioxide film. Low dielectric constant film is roughly grouped into inorganic film and organic film. Examples of inorganic film include alkyl siloxane polymer and HSQ (hydrogenated silsesquioxane polymer), which are known as inorganic spin-on dielectrics (SOD) film formed by spin coat. Low dielectric constant film can also be formed by chemical vapor deposition (CVD), and an example of inorganic low dielectric constant film formed by CVD is fluoridated silicon dioxide film.
The above-described inorganic films and silicon dioxide films can be made porous to further decrease the permittivity.
Examples of organic low dielectric constant film include organic polymer films, such as films of PTFE group, polyamide group, fluoridated polyamide, BCB (benzocyclobutene), parylene-N, parylent-F, MSQ (alkyl silsesquioxane polymer), HOSP (hydrogenated alkyl silsesquioxane polymer). Examples of organic low dielectric constant film formed by CVD include fluoridated carbon films, diamond-like carbon (DLC) films, SiCO films, and SiCO(H) films.
These organic films can also be formed as porous films to further decrease the permittivity.
To realize the advantageous applicability to the low dielectric constant film, non-plasma-activated source gas without containing reactive species (ions and radicals) is used in the step of forming the first diffusion barrier film 2 shown in
In the subsequent step shown in
Since the dielectric film 1 is covered with the first Cu-diffusion barrier film 2, the dielectric film 1 is not subjected to damage due to ions or radicals existing in the plasma-activated gas.
In this manner, in Example 1, by forming a layered Cu-diffusion barrier film consisting of the first Cu-diffusion barrier film 2 and the second Cu-diffusion barrier film 3, a high-quality TiN film (Cu-diffusion barrier film) with less impurities can be formed without damaging the underlying dielectric film 1.
In Example 1, gases other than TiCl4 may be used as the first source gas. Similarly, gases other than NH3 and plasma-activated NH3 may be used as the second source gas.
Using the same process, other types of Cu-diffusion barrier film can be fabricated. For example, TaN film, Ta/TaN layered film, WN film, W/WN layered film, Ti(C)N film, Ta(C)N film, W(C)N film, or W/W(C)N layered film can be formed, achieving the same effect as fabrication of TiN film, which effect is described in detail below. Ti(C)N film is a film containing carbon (C) as an impurity in a TiN film, and is fabricated when forming a film containing titanium nitride (TiN) using a metal-organic gas. Ta(C)N film is a film containing carbon as an impurity in a TaN film and is fabricated when forming a film containing tantalum nitride (TaN) using a metal-organic gas. W(C)N film is a film containing carbon (C) as an impurity in a WN film and is fabricated when forming a film containing tungsten nitride (WN) using a metal-organic gas.
Next, Example 2 is explained based on fabrication of a high-quality Cu-diffusion barrier film over a copper (Cu) film, without damaging the surface of the underlying copper film.
Then, in
Then, in
In Example 2, a metal-organic gas TEMAT is used in place of a halogen compound gas in the first step shown in
It is preferable to use a metal-organic compound not containing halogen, such as metal polyamide compounds or metal carboxyl compounds, which compounds prevent corrosion of the underlying copper (Cu) film 5. The underlying film is not limited to copper, and the same anti-corrosion effect can be achieved with respect to tungsten (W) film and aluminum (Al) film.
In the second step shown in
Thus, in Example 2, a high-quality Cu-diffusion barrier film with a TiN/Ti(C)N layered structure can be formed, without damaging the underlying copper (Cu) film 5, while preventing impurities from mixing into the diffusion barrier film.
In Example 2, gases other than TEMAT and TiCl4 may be used as the first source gas and the third source gas, respectively. Similarly, gases other than NH3 may be used as the second and fourth source gases. Using the same process, other types of Cu-diffusion barrier film can be fabricated. For example, TaN/Ta(C)N film, Ta/Ta(C)N layered film, WN/W(C)N layered film, or W/W(C)N layered film can be formed, achieving the same effect as fabrication of TiN/Ti(C)N film, which effect is described in detail below.
The second source gas and the fourth source gas used in steps shown in
In addition, in the first step of fabricating the first Cu-diffusion barrier film 6 shown in
The underlying film is not limited to copper, and the same damage preventing effect can be achieved with respect to tungsten (W) film and aluminum (Al) film.
In Example 3, fabrication of a high-quality Cu-diffusion barrier film under the situation where both a Cu film and a dielectric film exist in the underlying layer is explained. In this case, a high-quality Cu-diffusion barrier film is formed without damaging the underlying dielectric film or the underlying Cu layer.
In
Then, in
Then, in
In Example 3, non-plasma-activated NH3 gas is used as the second source gas in the first step shown in
In contrast, a plasma-activated NH3 gas is used to form the second Cu-diffusion barrier film in the second step shown in
In the first step for forming the first Cu-diffusion barrier film shown in
In contrast, TiCl4, which is a halogen compound gas, is used as the first source gas in the second step for forming the second Cu-diffusion barrier film shown in
In conclusion, a high-quality Cu-diffusion barrier film (TiN/Ti(C)N) with less impurity content can be formed without damaging the underlying dielectric film 1 or Cu film 5.
Any suitable gas, other than TEMAT and TiCl4, can be used as the first source gas. Similarly, any suitable gas, other than NH3, can be used as the second source gas. The Cu-diffusion barrier film is not limited to the TiN/Ti(C)N film, but any suitable combination, such as TaN/Ta(C)N film, Ta/Ta(C)N layered film, WN/W(C)N film, or W/W(C)N layered film, may be formed. These films have the same advantages as the TiN/Ti(C)N film in this embodiment.
Next, explanation is made of a film deposition apparatus used to form the Cu-diffusion films illustrated in Examples 1 through 3, in conjunction with
The processing chamber 11 is evacuated by an evacuation system (not shown) connected to the evacuation port 18 so as to maintain the chamber under reduced pressure. The wafer W to be processed is transported into or out of the processing chamber 11 through a gate valve (not shown).
A lifter pin 13 is provided to the wafer stage 12 in order to allow the wafer W to be mounted on or removed from the wafer stage 12 when the wafer W is transferred into or out of the processing chamber 11. The lifter pin 13 is coupled, via a coupling rod 14 vacuum-sealed with a bellows 16, to the up/down driving mechanism 17. By bringing up or down the lifter pin 13, the wafer W is mounted on or lifted from the wafer stage 12.
The processing chamber 11 is furnished with a gas supply port 11A, through which source gases or diluting gases required for film formation are introduced into the chamber 11.
A gas supply line 24 extends from the gas supply port 11A for supplying the first source gas and the first diluting gas into the processing chamber 11. The gas supply line 24 is connected to a halogen compound gas supply line 25 and a metal-organic gas supply line 26, through which the first source gases are to be supplied, respectively, as well as to a diluting gas supply line 27.
The halogen compound gas supply line 25 is connected, via a mass flow controller 25A and a valve 25B, to the first source gas supply 25C for supplying halogen compound gas. The first source gas supply 25C has a halogen compound gas supply source containing titanium (Ti), tantalum (Ta), or tungsten (W) in order to supply the halogen compound gas containing Ti, Ta, or W, as the first source gas, to the processing chamber 11.
The metal-organic gas supply line 26 is connected, via a mass flow controller 26A and a valve 26B, to another first source gas supply 26C for supplying a metal-organic gas. The first source gas supply 26C has a metal-organic gas supply source containing titanium (Ti), tantalum (Ta), or tungsten (W) in order to supply the metal-organic gas containing Ti, Ta, or W, as the first source gas, to the processing chamber 11.
The diluting gas supply line 27 is connected to a diluting gas supply 27C via a mass flow controller 27A and a valve 27B. The diluting gas supply 27C has a diluting gas supply source for supplying a diluting gas, such as nitrogen (N2), argon (Ar), or helium (He), via the gas supply line 24 to the processing chamber 11 to dilute the first source gas. Supplying the diluting gas through the gas supply line 24 is advantageous in preventing back-flow of the gases from the processing chamber 11 back to the gas supply line 24.
A second gas supply line 20 also extends from the gas supply port 11A via a remote plasma source 19, which is explained below. The second gas supply line 20 is connected to a nitride gas supply line 21 and a hydrogen gas supply line 22, through which the second sources gases are to be supplied, as well as to a diluting gas supply line 23. The nitrogen gas supply line 21 is connected, via a mass flow controller 21A and a valve 21B, to the second source gas supply 21C for supplying nitride gas. The second source gas supply 21C has a nitride gas supply source for supplying a nitrogen compound, such as NH3, N2H4, NH(CH3)2, N2H3CH3, to the processing chamber 11.
The hydrogen gas supply line 22 is connected, via a mass flow controller 26A and a valve 26B, to another second source gas supply 26C for supplying a metal-organic gas. The second source gas supply 22C has a reducing gas supply source, such as hydrogen (H2) gas supply source, to supply the hydrogen gas, for example, to the processing chamber 11.
The diluting gas supply line 23 is connected to a diluting gas supply 23C via a mass flow controller 23A and a valve 23B. The diluting gas supply 23C supplies a diluting gas, such as nitrogen (N2), argon (Ar), or helium (He), via the second gas supply line 20 to the processing chamber 11 to dilute the second source gas. Supplying the diluting gas through the gas supply line 20 is advantageous in preventing back-flow of the gases from the processing chamber 11 back to the remote plasma source 19 or to the gas supply line 20.
The remote plasma source 19 has a plasma generating apparatus, to which apparatus RF power is applied to excite the gases into plasma. For example, the remote plasma source 19 excites the nitrogen source gas or the hydrogen source gas supplied to the remote plasma source 19, into the plasma, as necessary. If plasma excitation is not performed, the gas passes through the remote plasma source 19, as it is, and is supplied to the processing chamber 11. Under plasma excitation, reacting species, such as ions radicals, are generated by gas dissociation, which species are supplied to the processing chamber 11 through the gas supply port 11A. For example, if the second source gas is plasma-excited, NHx* (radicals), H* (radicals), or N* (radicals) are supplied into the processing chamber 11.
In this embodiment, the plasma excitation is performed using an ICP (induced coupling plasma) source of high-frequency waves at 2 MHz. However, the invention is not limited to this method, and, for example, parallel plate plasma excitation or ECR plasma excitation may be used. In addition, another exciting frequency, such as 13.56 MHz high-frequency waves or microwaves (at 2.45 GHz) may be employed. As long as the supplied gas is dissociated through plasma excitation, any suitable frequency and excitation method can be employed.
The operations of the film deposition apparatus 10, including opening and closing of the valves 21B through 27B, the motion of the lifter pin 13, and the plasma excitation in the remote plasma source 19, are comprehensively controlled by the controller 10A.
Next, detailed operations of the film deposition apparatus 10 for forming the films as illustrated in Examples 1 through 3 are explained below.
In step S101, the wafer W to be processed is transferred into the film deposition apparatus 10.
Then, in step S102, the wafer W is placed onto the wafer stage 12.
Then, in step S103, the wafer W is heated by the heater set inside the wafer stage 12, and is maintained at about 400° C. in this step and the subsequent steps.
Then, in step S104, the valve 25B is opened to supply TiCl4 (the first source gas) into the processing chamber 11 at gas flow rate of 30 sccm under the control of the mass flow controller 25A. Simultaneously, the valves 27B and 23B are also opened to supply N2 gas as the diluting gas to the processing chamber 11, through diluting gas supply lines 27 and 23, at 100 sccm each under the control of mass flow controllers 27A and 23A, respectively. Accordingly, the total of 200 sccm N2 gas is supplied to the processing chamber 11. The TiCl4 gas is supplied onto the wafer W, and is adsorbed onto the dielectric (oxide) film 1.
Then, in step S105, the valves 23B, 25B and 27B are closed to stop the gas supply of TiCl4 and N2 into the processing chamber 11. The residual TiCl4 remaining in the processing chamber 11, without being adsorbed onto the dielectric film 1, is purged from the evacuation port 18.
Then, in step S106, the valve 21B is opened to introduce NH3 into the processing chamber 11 at 800 sccm under the control of mass flow controller 21A. Simultaneously, the valves 27B and 23B are opened to supply N2 gas as the diluting gas to the processing chamber 11, through diluting gas supply lines 27 and 23, at 100 sccm each under the control of mass flow controllers 27A and 23A, respectively. Accordingly, the total of 200 sccm N2 gas is supplied to the processing chamber 11. The NH3 gas is supplied onto the wafer W, and is reacted with the TiCl4 adsorbed onto the dielectric (oxide) film 1 to form TiN.
Then, in step S107, the valves 21B, 23B and 27B are closed to stop gas supply of NH3 and N2 to the processing chamber 11. The unreacted NH3 remaining in the processing chamber 11 is purged from the evacuation port 18.
Then, in step S108, the process returns to step S104 to repeat steps S104 through S107 until a desired thickness of TiN film is obtained. After the necessary number of repetitions, the process proceeds to step S109. Since non-plasma-excited NH3 gas is used as the second source gas, damaging species, such as ions or radicals, does not exist in the source gas, and therefore, the underlying dielectric film 1 is not damaged.
The subsequent steps S109 and S110 are the same as steps S104 and S105.
Then, in step S111, the valve 21B is opened to supply NH3 at 400 sccm under the control of mass flow controller 21A. Simultaneously, the valves 27B and 23B are opened to supply N2 gas as the diluting gas to the processing chamber 11, through diluting gas supply lines 27 and 23, at 100 sccm each under the control of mass flow controllers 27A and 23A, respectively. Accordingly, the total of 200 sccm N2 gas is supplied to the processing chamber 11. In this step, high-frequency power of 400 W is applied to the remote plasma source 19 to perform plasma excitation. The supplied NH3 is dissociated into NHx*, which is then supplied to the processing chamber 11. This NHx* reacts with TiCl4 adsorbed onto the previously formed TiN film 2., and an additional TiN film 3 is formed. In this case, NHx*, is used for the reaction in place of NH3, and therefore, the reaction with TiCl4 is promoted to form the TiN film promptly. Consequently, impurities, such as residual chlorine, contained in the resultant TiN film can be reduced, and high film quality is realized.
Then, in step S112, application of high-frequency power to the remote plasma source is stopped, and the valves 21B, 23B and 27B are closed to stop gas supply of NH3 and N2 to the processing chamber 11. The unreacted NH3 remaining in the processing chamber 11 is purged from the evacuation port 18.
Then, in step S113, the process returns to step S109 to repeat steps S109 through S112 until a desired thickness of the second Cu-diffusion barrier film 3 is obtained. After the necessary number of repetitions, the process proceeds to step S114.
In step S114, the lifter pin 13 is elevated to remove the wafer W from the wafer stage 12.
In step S115, the wafer W is transported out of the processing chamber 11.
In step S116, the wafer W is transported to a Cu-film deposition apparatus, such as a PVD apparatus, a CVD apparatus, or a plating apparatus, to form copper (Cu) film 4 over the second Cu-diffusion barrier film 3.
In this flow, TiCl4 is used as the first source gas, which gas is introduced in steps S104 through S109. When forming the first Cu-diffusion barrier film 2, NH3 is introduced as the second source gas in step S106, while plasma-activated NH3 is introduced as the second source gas when forming the second Cu-diffusion barrier film 4, to form the double layered TiN film. However, the invention is not limited to this example.
Table 1 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when the TiN barrier film shown in
Table 2 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when tantalum nitride (TaN) barrier films are formed using a halogen compound gas as the first source gas. By using the source gases listed in Table 2, TaN films can be formed in a process similar to the process described above. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, a Ta/TaN film is formed, which film can achieve the same effect as that illustrated above.
Table 3 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when tungsten nitride (WN) barrier films are formed using a halogen compound gas as the first source gas. By using the source gases listed in Table 3, WN film can be formed in a process similar to the process described above. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, a W/WN film is formed, which film can achieve the same effect as that illustrated above.
Table 4 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when Ti(C)N barrier films are formed using a metal-organic gas as the first source gas. By using the source gases listed in Table 4, a Ti(C)N film can be formed in a process similar to the process described above.
Table 5 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when Ta(C)N barrier films are formed using a metal-organic gas as the first source gas. By using the source gases listed in Table 5, a Ta(C)N film can be formed in a process similar to the process described above.
Table 6 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as example of the first and second source gases used to form the second Cu-diffusion barrier film, when W(C)N barrier films are formed using a metal-organic gas as the first source gas. By using the source gases listed in Table 6, a W(C)N film can be formed in a process similar to the process described above. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, a W(C)/W(C)N film is formed, which film can achieve the same effect as that illustrated above.
Steps S201 through S203 and steps S214 through S216 are the same as steps S101 through S103 and steps S114 through S116, respectively, illustrated in
In step S204, the valve 26B is opened to supply TEMAT (which is the first source gas) into the processing chamber 11 at gas flow rate of 30 sccm under the control of the mass flow controller 26A. Simultaneously, the valves 27B and 23B are also opened to supply N2 gas as the diluting gas to the processing chamber 11, through diluting gas supply lines 27 and 23, at 100 sccm each under the control of mass flow controllers 27A and 23A, respectively. Accordingly, the total of 200 sccm N2 gas is supplied to the processing chamber 11. The first source gas TEMAT is supplied onto the wafer W, and is adsorbed onto the copper (Cu) film 5 formed over the wafer W.
Then, in step S205, the valves 23B, 26B and 27B are closed to stop the gas supply of TEMAT and N2 into the processing chamber 11. The residual TEMAT remaining in the processing chamber 11, without being adsorbed onto the copper (Cu) film 5, is purged from the evacuation port 18.
Then, in step S206, the valve 21B is opened to introduce NH3 into the processing chamber 11 at 800 sccm under the control of mass flow controller 21A. Simultaneously, the valves 27B and 23B are opened to supply N2 gas as the diluting gas to the processing chamber 11, through diluting gas supply lines 27 and 23, at 100 sccm each under the control of mass flow controllers 27A and 23A, respectively. Accordingly, the total of 200 sccm N2 gas is supplied to the processing chamber 11. The NH3 gas is supplied onto the wafer W, which has been heated up to about 400° C., and is reacted with the TEMAT adsorbed onto the wafer W.
Then, in step S207, the valves 21B, 23B and 27B are closed to stop gas supply of NH3 and N2 to the processing chamber 11. The unreacted NH3 remaining in the processing chamber 11 is purged from the evacuation port 18.
Then, in step S208, the process returns to step S204 to repeat steps S204 through S207 until a desired thickness of Ti(C)N film 6 is obtained. After the necessary number of repetitions, the process proceeds to step S209.
Then, in steps S209 through S212, the second Cu-diffusion barrier film 7 is formed. As the second Cu-diffusion barrier film 7, TiN film is formed using TiCl4 as the first source gas. Steps S209 through S212 are the same steps S104 through S107 shown in
In step S213, the process returns to step S209 to repeat steps S209 through S212 until a desired thickness of the second Cu-diffusion barrier film 7 is obtained. After the necessary number of repetitions, the process proceeds to step S214.
In Example 6, a metal-organic gas is used as the first source gas (step S204) to form the first Cu-diffusion barrier film Ti(C)N, while a halogen compound gas is used as the first source gas (step S209) to form the second Cu-diffusion barrier film (TiN). With this arrangement, the underlying copper (Cu) film 5 is prevented from corroding due to the halogen. Consequently, a Cu-diffusion barrier film with less impurity content and lower electric resistance can be obtained.
In Example, 6, the first source gas used in step S204 is TEMAT as an example of the metal-organic gas, the first source gas used in step S209 is TiCl4 as an example of the halogen compound gas, and the second source gas used in steps S206 and S211 is NH3. However, the invention is not limited to this example.
Table 7 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film 6, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film 7, when fabricating the TiN/Ti(C)N layered film. Any combination of these gases can achieve the same effect as that illustrated in the above-described example.
Table 8 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when a TaN/Ta(C)N layered film is formed. By using the source gases listed in Table 8, the above-described effect can be achieved.
Table 9 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when a WN/W(C)N layered film is formed. By using the source gases listed in Table 9, the above-described effect can be achieved.
In Example 6, the second source gas may be plasma-activated in steps S206 and S211, as in Example 2. In this case, dissociation of the second source gas is promoted, and the reaction for forming the Cu-diffusion barrier film is advanced. As a result, the amount of impurities contained in the Cu-diffusion barrier film can be reduced, and the electric resistance is lowered.
Table 10 illustrates examples of the first source gases used to form the first and second Cu-diffusion barrier films, in combination with examples of the plasma-activated second source gases used to form the first and second Cu-diffusion barrier films. By using the source gases listed in Table 10, TiN/Ti(C)N films can be formed, achieving the same effect.
Table 11 illustrates examples of the combinations of the first and second source gases to form the first and second Cu-diffusion barrier films, respectively, when TaN/Ta(C)N barrier films are formed using a plasma-activated gas as the second source gas. By using the source gases listed in Table 11, TaN/Ta(C)N films can be formed in a process similar to the process described above, achieving the effect of reduced impurity content. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, a Ta/Ta(C)N film is formed, in place of the TaN/Ta(C)N film, which film can achieve the same effect as that illustrated above.
Table 12 illustrates examples of the combinations of the first and second source gases to form the first and second Cu-diffusion barrier films, respectively, when WN/W(C)N barrier films are formed using a plasma-activated gas as the second source gas. By using the source gases listed in Table 12, WN/W(C)N films can be formed in a process similar to the process described above. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, W/W(C)N film is formed, in place of a the WN/W(C)N film, which film can achieve the same effect.
Steps S301 through S310 and steps S313 through S316 are the same as steps S201 through S210 and steps S213 through S216, respectively, illustrated in
The first and second source gases used to form each of the first and second Cu-diffusion barrier films may be changed from those illustrated in the flowchart of
Table 13 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film 8, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film 9, when fabricating the TiN/Ti(C)N barrier films. By using these gases, the same effect can be achieved.
Table 14 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when a TaN/Ta(C)N layered film is formed. By using the source gases listed in Table 14, the above-described effect can be achieved. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, a Ta/Ta(C)N film is formed, in place of the TaN/Ta(C)N film, which film can achieve the same effect.
Table 15 illustrates examples of the first and second source gases used to form the first Cu-diffusion barrier film, as well as examples of the first and second source gases used to form the second Cu-diffusion barrier film, when a WN/W(C)N layered film is formed. By using any one of the source gases listed in Table 15, the above-described effect can be achieved. It should be noted that if plasma-activated H2 gas (H+/H*) is used as the second source gas to form the second Cu-diffusion barrier film, a W/W(C)N film is formed, in place of the WN/Ta(C)N film, which film can achieve the same effect.
Any combination of the first and second source gases illustrated in Example 7 allows a high-quality Cu-diffusion barrier film, without damaging underlying layers located beneath the Cu-diffusion film.
In
In
The steps shown in
Then, in
Since non-plasma-activated NH3 is used as the second source gas to form the TiN film 39, as described in Example 1, damaging species, such as ions or radicals, are not contained in the second source gas. This arrangement can prevent the first dielectric film 33 or the second dielectric film 35 from being damaged.
Then, in
In this step, plasma-activated NH3 is used as the second source gas to promote dissociation of the gas and advance the reaction with TiCl4 supplied as the first source gas. Consequently, the impurity content, such as chlorine (Cl) content, in the TiN film can be reduced, and a high-quality TiN film with less electric resistance is obtained.
The first dielectric film 33 and the second dielectric film 35 are covered with the first Cu-diffusion barrier film 39 during the formation of the second Cu-diffusion barrier film 40 using plasma-activated gas. Accordingly, the first dielectric film 33 and the second dielectric film 35 are protected from damage by ions or radicals existing in the plasma-activated gas. In other words, by employing the double-layered structure, a high-quality Cu-diffusion film with less impurity content can be realized, while preventing damage to the underlying first dielectric film 33 and second dielectric film 35.
Then, in
The film formation method can also be applied to fabrication of a semiconductor device with multilevel interconnect lines, as illustrated in
The semiconductor device of
For example, after the CMP performed in
When a TiN film is formed as the Cu-diffusion barrier film, the first source gas and the second source gas may be altered, as described in Example 5.
In addition, by changing the first source gas and the second source gas, another type of Cu-diffusion barrier film, such as TaN film, Ta/TaN layered film, WN film, W/WN layered film, Ti(C)N film, Ta(C)N film, W(C)N film, or W(C)/W(C)N layered, may be formed.
In either case, a high-quality Cu-diffusion film with less impurity content can be obtained, without damaging the underlying first dielectric film 33 or second dielectric film 35, by employing the double layered structure of the first and second Cu-diffusion films.
Dielectric films (such as the first dielectric film 33 and the second dielectric film 35) may be roughly grouped into inorganic films and organic films.
Examples of inorganic dielectric film include alkyl siloxane polymer and HSQ (hydrogenated silsesquioxane polymer), which are known as inorganic spin-on dielectric (SOD) film formed by spin coat. Low dielectric constant films are formed by chemical vapor deposition (CVD), and an example of inorganic low dielectric constant film formed by CVD includes fluoridated silicon dioxide film.
The above-described inorganic films and silicon dioxide films can be made porous to further decrease the permittivity.
Examples of organic dielectric film include organic polymer films, such as films of PTFE group, polyamide group, fluoridated polyamide, BCB (benzocyclobutene), parylene-N, parylent-F, MSQ (alkyl silsesquioxane polymer), and HOSP (hydrogenated alkyl silsesquioxane polymer). Examples of organic low dielectric constant film formed by CVD include fluoridated carbon films, diamond-like carbon (DLC) films, SiCO films, and SiCO(H) films.
These organic films can also be formed as porous films to further decrease the permittivity.
In Example 9, the film formation method shown in Example 6 is applied to another fabrication process of a semiconductor device. In this case, the steps of forming the first Cu-diffusion barrier film 39 and the second Cu-diffusion barrier film 40 illustrated in Example 8 in conjunction with
To be more precise, steps S204 to S208 illustrated in
Then, steps S209 through S213 illustrated in
During the formation of the second Cu-diffusion barrier film 40, the underlying copper film 31 is covered with the first Cu-diffusion barrier film 39, and therefore is not damaged by the halogen contained in the first source gas. The same applies to other metals, such as tungsten (W) or aluminum (Al), used in the underlying film.
As the first Cu-diffusion barrier film, a Ti(C)N film of satisfactory quality, with less impurity content, can be formed without damaging the underlying copper film 31.
By appropriately selecting the first source gas and the second source gas, various types of layered film, including TaN/Ta(C)N film, Ta/Ta(C)N film, WN/W(C)N film, and W/W(C)N film, can be formed. By forming a double-layered Cu-diffusion barrier film with the first and second Cu-diffusion barrier films, the impurity content is reduced as a whole, while preventing damage to the underlying metal layer.
In Example 10, the film formation method shown in Example 7 is applied to another fabrication process of a semiconductor device. In this case, the steps of forming the first Cu-diffusion barrier film 39 and the second Cu-diffusion barrier film 40 illustrated in Example 8 in conjunction with
To be more precise, steps S304 to S308 illustrated in
In addition, a metal-organic gas TEMAT is used as the first source gas, in place of the halogen compound gas, to prevent the underlying copper film 31 from corroding due to halogen. Consequently, all the underlying films, including the first and second dielectric films 33 and 35 and the metal film (Cu film) 31, are protected from the damage.
Then, steps S309 through S313 illustrated in
Since the underlying first dielectric film 33 and second dielectric film 35 are covered with the first Cu-diffusion barrier film 39 during the formation of the second Cu-diffusion barrier film 40, they are protected from damage by ions or radicals existing in the plasma-activated gas.
As the first source gas, TiCl4, which is a halogen compound gas, is used to prevent organic compounds including carbon (C) and CHx from being taken into the film and to reduce the electric resistance of the resultant TiN film.
Since the underlying copper film 31 is covered with the first Cu-diffusion barrier film 39 during the formation of the second Cu-diffusion barrier film 40, the copper film 31 is protected from damage by halogen contained in the first source gas. The same applies to other metals, such as tungsten (W) or aluminum (Al), used in the underlying film.
By forming double-layered Cu-diffusion film with the first and second Cu-diffusion films 39 and 40, the impurity content is reduced as a whole, while preventing damage to all the underlying films, including the first and second dielectric films 33 and 35 and the copper film 31.
As described in Example 7, the first source gas and the second source gas may be selected appropriately to form the TiN/Ti(C)N diffusion barrier film. In this case, various types of layered film, including TaN/Ta(C)N film, Ta/Ta(C)N film, WN/W(C)N film, and W/W(C)N film, can be formed. In any one, the impurity content in the barrier film is reduced as a whole, while preventing damage to the dielectric film and the metal film.
Regardless of the types and materials of the underlying dielectric films 33 and 35, the film formation method is effectively applied, as described in Example 8.
The first and second Cu-diffusion barrier films can be formed using a film deposition apparatus 50 illustrated in
In
The inner space 51A of the processing chamber 51 is connected to an evacuation port 55, and evacuated by evacuation means 53, such as a turbo molecular pump. Thus, the inner space 51A of the processing chamber 51 is maintained under reduced pressure. The wafer W to be processed is transported into or out of the processing chamber 51 through a gate valve (not shown).
The processing chamber 51 is furnished with an opening 51B, which opening is connected to a gas supply tube 51C through which the first source gas and the second source gas are introduced into the chamber 51.
A gas supply line 60 extends from the gas supply tube 51C to supply the first source gas. The gas supply line 60 is connected to a halogen compound gas supply line 62 via the valve 62a, and to a metal-organic gas supply line 61 via the valve 61a.
The metal-organic gas supply line 61 is connected to a vaporizer 61A, which vaporizer is further connected to a gas line 63 furnished with valves 63a, 63b, and 63c and a liquid mass flow controller 63A. The gas line 63 is coupled with a tank 66 containing a source material 66A of the metal-organic first source gas. An example of the source material 66A is Taimata (registered trademark), consisting of Ta(NC(CH3)2C2H5) (N(CH3)2)3.
The tank 66 is also connected to a gas line 65 furnished with a valve 65a, through which an inert gas, such as helium (He) gas, is supplied to the tank 55. The material 66A contained in the tank 66 is pressurized by the inert gas, and heated to 50° C. by a heater (not shown).
The pressurized and heated material 66A is supplied to the vaporizer 61A under the control of the liquid mass flow controller 63A. The vaporizer 61A is connected to the gas line 64 furnished with the valve 64a and 64b, as well as the mass flow controller 64A. The evaporated material 66A is supplied from the vaporizer 61A to the processing chamber 51, together with a carrier gas (such as argon (Ar) gas) supplied from the gas line 64, through the gas lines 61, the gas line 60, and the gas supply tube 51C.
The material 66A may be supplied after it is dissolved in an organic solvent, such as octane or hexane. In this case, the tank 66 may not be heated. By stirring the organic solvent using an agitation stick, the material 66A is dissolved uniformly in the organic solvent.
The gas line 62 is connected to a gas line 68 furnished with valves 68a, 68b, and 68c and the mass flow controller 68A. The gas line 68 is connected to a tank 69 that contains a halogen material 69A of the first source gas, such as TaCl5.
The tank 69 is heated to, for example, 150° C., to evaporate the material 69A consisting of TaCl5. The evaporated material 69A is supplied under the control of the mass flow controller 68A to the inner space 51 of the processing chamber, via the gas line 62, the gas line 60, and the gas supply tube 51C. Simultaneously, an inert gas (e.g., Ar gas) may be supplied to the processing chamber 51, together with this evaporated source gas, through the gas line 67 furnished with valves 67a and 67b and the mass flow controller 67A.
The gas supply tube 51C is also connected to the gas line 57 via a plasma source 54 (which is described below). The gas line 57 is then branched into a gas line 58 and a gas line 59. The gas line 58 is furnished with valves 58a and 58b and a mass flow controller 58A, and is used to supply the second source gas consisting of, for example, H2 to the plasma source 54. The gas line 59 is furnished with valves 59a and 59b and a mass flow controller 59A, and is used to supply a carrier gas (such as Ar gas) to the plasma source 54.
The plasma source 54 is made of a dielectric material, such as Al2O3, quartz, SiN or BN, and it has a substantially cylindrical shape. A coil 54a is wound around the cylindrical plasma source 54, and connected to a high-frequency power supply 56. High frequency power is applied to the coil 54a by the power supply 56 to excite the second source gas supplied into the plasma source 54 into plasma, as necessary. Reactive species, such as ions and radicals, are produced from the plasma-activated gas, and supplied to the processing chamber 51 via the gas supply tube 51C.
The plasma is generated by an inductively coupled plasma (ICP) generator at a high frequency of 13.56 MHz. Plasma excitation may be performed by a parallel plate plasma system or an ECR plasma system. The plasma may be generated at a lower frequency, such as 400 kHz or 800 kHz, or alternatively, radio waves or microwaves (2.45 GHz) may be used. Any suitable method or frequency can be employed as long as the gas is dissociated into excited plasma.
The operations of the film deposition apparatus 50, including the opening and closing of the valves and plasma excitation of the plasma source 54, are comprehensively controlled by a controller (not shown).
Next, explanation is made of formation of the Cu-diffusion barrier film using the film deposition apparatus 50.
In step S401, the wafer W to be processed is transported into the film deposition apparatus 50.
Then, in step S402, the wafer W is placed onto the wafer stage 52.
Then, in step S403, the wafer W is heated by the heater set inside the wafer stage 52, and is maintained at about 270° C. in this step and the subsequent steps.
Then, in step S404, the valves 65a, 63a, 63b, 63c, and 61a are opened to apply pressure to the tank 66 and supply liquid material 66A consisting of Ta(NC(CH3)2C2H5) (N(CH3)2)3 through the gas line 63, under the control of the mass flow controller 63A, so as to supply the liquid material 66A to the vaporizer 61A at 20 mg/min.
The evaporated material 66A is supplied to the processing chamber 51, together with argon (Ar) of 200 sccm supplied to the vaporizer 61A through the gas line 64.
At the same time, the valves 59a and 59b are also opened to supply argon (Ar) gas at 100 sccm under the control of the mass flow controller 59A to the processing chamber 51 through the gas line 57. This Argon gas flow prevents the evaporated material 66A from flowing back to the plasma source 54 through the gas supply tube 51C.
The material 66A is supplied and adsorbed onto the wafer W.
Then, in step S405, the valves 65a, 63a, 63b, 63c, and 61a are closed to stop supplying the material 66A to the processing chamber 51. The residual material 66A remaining in the processing chamber 51, without being adsorbed onto the wafer W, is purged from the evacuation port 55.
In this step, the valves 58a and 58b are opened to supply H2 gas to the processing chamber 51 at 200 sccm under the control of the mass flow controller 58A, through gas line 57. The mass flow controller 59A is also controlled to adjust the argon (Ar) gas flow through the gas line 57 to 200 sccm.
Then, in step S406, high-frequency power of 800 W is applied to the coil 54a to perform plasma excitation in the plasma source 54. Since the H2 gas supply has already started in the previous step S405, the mass flow of the H2 gas is stable at the beginning of step S406, and plasma excitation is performed promptly upon application of the high-frequency power.
Then, in step S407, the argon (Ar) gas supply through the gas line 57 is stopped, such that only H2 gas is supplied to the plasma source 54. In the plasma source 54, the hydrogen gas is dissociated into H+/H* (hydrogen ions and hydrogen radicals), and the plasma-activated hydrogen is supplied into the processing chamber 51. The ions and radicals H+/H* react with the material 66A adsorbed onto the substrate to form a Ta(C)N film. By stopping the argon (Ar) gas supply, the hydrogen ions and radicals (H+/H*) reach the peripheral portion of the substrate, and the reaction with the material 66A is promoted.
Then, in step S408, the valves 58a and 58b are closed to stop supplying the hydrogen gas to the plasma source 54, that is, to stop supplying the hydrogen ions and radicals to the processing chamber 51. The residual reactive species H+/H*, H2, or by-product materials of the reaction are purged out of the chamber 51 through the evacuation port 55.
The processing times in steps S404, S405, S406, S407, and S408 are 3 seconds, 3 seconds, 10 seconds, 10 seconds, and 1 second, respectively.
Then, in step S409, the process returns to step S404 to repeat steps S404 through S408 until a desired thickness of Ta(C)N film (the first Cu-diffusion barrier film) is obtained. After the necessary number of repetitions, the process proceeds to step S410.
Then, in step S410, the valves 68a, 68b, 68c, and 62a are opened to supply the material 69A, which is evaporated TaCl5, to the processing chamber 51 at 3 sccm under the control of mass flow controller 68A.
Simultaneously, the valves 59a and 59b are opened to supply argon (Ar) gas at 200 sccm under the control of the mass flow controller 59A, to the processing chamber 51 through the gas supply line 57. This argon (Ar) gas flow prevents the evaporated material 69A from flowing back to the plasma source 54 through the gas supply tube 51C.
The evaporated material 69A is supplied and adsorbed onto the substrate.
Then, in step S411, the valves 68a, 68b, 68c, and 62a are closed to stop supplying the material 69A to the processing chamber 51. The residual material 69A remaining in the processing chamber 51, without being adsorbed onto the wafer W, is purged from the evacuation port 55.
Then, in step S412, the argon (Ar) gas supply through the gas line 57 is stopped, and the valves 58a and 58b are opened to supply H2 gas to the plasma source 54 at 750 sccm under the control of the mass flow controller 58A, through gas line 58. High-frequency power of 1000 W is applied to the coil 54a to perform plasma excitation in the plasma source 54.
In the plasma source 54, the hydrogen gas is dissociated into H+/H* (hydrogen ions and hydrogen radicals), and the plasma-activated hydrogen is supplied into the processing chamber 51. The ions and radicals H+/H* react with the material 69A adsorbed onto the substrate to form a tantalum (Ta) film.
Then, in step S413, the application of the high-frequency power is stopped, and the valves 58a and 58b are closed to stop supplying the hydrogen gas to the plasma source 54, that is, to stop supplying the hydrogen ions and radicals to the processing chamber 51. The residual reactive species H+/H*, H2, or by-product materials of the reaction are purged out of the chamber 51 through the evacuation port 55.
Then, in step S414, the process returns to step S410 to repeat steps S410 through S413 until a desired thickness of Ta film (the second Cu-diffusion barrier film) is obtained. After the necessary number of repetitions, the process proceeds to step S415.
Then, in step S416, the processed wafer W is transported out of the processing chamber 51.
Then, in step S417, the wafer is transported into a copper (Cu) film deposition apparatus, such as a plating apparatus, a PVD apparatus, or a CVD apparatus, to form a copper (Cu) film over the second Cu-diffusion barrier film.
The second Cu-diffusion barrier film 503 consisting of tantalum (Ta) with a thickness of 3 nm is formed over the first Cu-diffusion barrier film 502, by repeating the process “b” 300 times under the conditions illustrated in
The double-layered Cu-diffusion barrier film consisting of the first and second Cu-diffusion barrier films may be formed using a film deposition apparatus 70 shown in
In
The inner space 71A of the processing chamber 71 is connected to an evacuation port 75, and evacuated by evacuation means (not shown) to maintain the inner space 71A of the processing chamber 71 under reduced pressure. The wafer W to be processed is transported into or out of the processing chamber 71 through a gate valve (not shown).
A substantially cylindrical shower head 73 is provided in the processing chamber 71 so as to face the wafer stage 72. An insulator 76 made, for example, quartz or ceramics (such as SiN or AlN), is provided so as to cover the shower head 73, leaving the bottom facing the wafer stage 72 uncovered.
An opening is provided to the processing chamber 71, through which opening an insulator 74 made of a dielectric material is inserted. A lead 77a connected to a high-frequency power supply 77 penetrates through the insulator 74 such that the other end of the lead 77a is connected to the shower head 73. High frequency power is applied to the shower head 73 via the lead 77a.
An insulator 60A made of a dielectric material, such as quartz or ceramics (SiN, AlN, or Al2O3), is inserted in the gas line 60. Accordingly, the gas line 60 is connected to the shower head 73 via the insulator 60A. The insulator 60A electrically insulates the gas line 60, through which materials 66A and 69A are supplied, from the shower head 73.
Similarly, an insulator 57A made of a dielectric material, such as quartz or ceramics (SiN, AlN, or Al2O3), is inserted in the gas line 57. Accordingly, the gas line 57 is connected to the shower head 73 via the insulator 57A. The insulator 57A electrically insulates the gas line 57, through which H2 gas and Ar gas are supplied, from the shower head 73. A gas containing a hydrogen compound may be supplied, in addition to the hydrogen (H2) gas, through the gas line 57.
When supplying H2 gas or Ar gas to the processing chamber 71, a high power is applied to the shower head 73 to perform plasma excitation by the high-frequency power supply 77, as necessary. In this structure, plasma excitation is performed in the inner space 71A of the processing chamber 71 to dissociate the H2 gas.
Using the film deposition apparatus 70, the first Cu-diffusion barrier film made of Ta(C)N and the second Cu-diffusion barrier film made of tantalum (Ta) can be formed, in a manner similar to Example 12. The film deposition apparatus 70 is capable of carrying out the film formation process shown in Examples 1 through 3.
With the present invention, a Cu-diffusion barrier film with satisfactory quality can be formed without damaging underlying films.
The formed Cu-diffusion barrier film contains a lesser amount of impurities, has good crystal orientation, and satisfactory coverage over a minute pattern.
Although the invention has been described using specific examples, the invention is not limited to these examples, and there are many modifications and substitutions within the scope of the invention defined by the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2003-170950 | Jun 2003 | JP | national |
| 2004-070144 | Mar 2004 | JP | national |
This application is a continuation application filed under 35 USC 111(a) claiming benefit under 35 USC 120 and 365(c) of PCT application JP04/006060, filed Apr. 27, 2004, the entire contents of which are incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/JP04/06060 | Apr 2004 | US |
| Child | 11231962 | Sep 2005 | US |
