JP2012186208A - Wiring formation method and wiring formation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a wiring formation method which improves electric characteristics in a wiring structure where a carbon nano-tube penetrating through an insulation layer connects with a conductive layer on which the insulation layer is laminated, and to provide a wiring formation device using the wiring formation method.SOLUTION: Catalyst layers 36, 37 are formed in holes 35 penetrating through an insulation layer 34 laminated on a lower wiring layer 32 so as to include an entire inner surface of the hole 35. Then, a sheath Sh is formed in each hole 35 and plasma is generated so that the thickness of the sheath Sh on an inner wall surface 35a of the hole 35 is smaller than the thickness of the sheath Sh on a bottom wall surface 35b of the hole 35. Further, the catalyst layers 36, 37 formed on the inner wall surface 35a of the hole 35 are removed by spatter particles Sp in the plasma, and then a carbon nano-tube 38 is formed from the bottom wall surface 35b by using the catalyst layers 36, 37 remaining on the bottom wall surface 35b of the hole 35.

Description

  The present invention relates to a wiring forming method for connecting a carbon nanotube penetrating an insulating layer to a conductive layer in which insulating layers are laminated, and the wiring forming apparatus.

  Conventionally, as described in Patent Document 1, for example, as a wiring structure of a semiconductor device, a technique using a carbon nanotube as a wiring connecting a lower layer wiring and an upper layer wiring of the semiconductor device has been studied.

  When forming the carbon nanotube, first, a hole connected to the lower layer wiring is formed in the interlayer insulating layer on the lower layer wiring, and a catalyst layer is formed on the inner surface of the hole by, for example, sputtering. Next, carbon nanotubes are formed on the catalyst layer by a thermal CVD method using a hydrocarbon gas. At this time, since the entire inner surface of the hole is covered with the catalyst layer, the carbon nanotubes grown from the catalyst layer are formed on the entire inner surface of the hole so as to follow the catalyst layer. As a result, the inside of the hole is a mixture of carbon nanotubes that contribute to the connection between the electrodes by being formed on the bottom wall surface of the hole, and carbon nanotubes that are formed on the inner wall surface of the hole and do not contribute to the connection between the electrodes. Will do.

  In recent years, with the miniaturization of semiconductor devices, the above-mentioned miniaturization of holes has progressed, so the upper limit value of the density of carbon nanotubes that can be formed in the holes is also decreasing. For this reason, it is becoming difficult to ignore the influence of carbon nanotubes that do not contribute to the connection between the electrodes as described above on the electrical characteristics of the wiring structure, such as electrical conductivity and allowable current density. Therefore, an attempt has been made to form the carbon nanotube only on the bottom wall surface of the hole by removing the catalyst layer formed on the inner wall surface of the hole by ion milling before forming the carbon nanotube.

JP 2009-298640 A

  Here, the aspect which removes a part of catalyst layer by the said ion milling is typically shown in FIG. As shown in FIG. 6, the lower wiring layer 42 formed on the support substrate 41 is exposed as the bottom wall surface 44 b of the hole 44 formed in the insulating layer 43. At this time, in order to selectively remove only the catalyst layer 45 formed on the inner wall surface 44a of the hole 44 from the catalyst layer 45 formed on the inner surface of the hole 44, only the inner wall surface 44a of the hole 44 is removed. It is necessary to set the trajectory of the milling particle P so that the milling particle P collides.

However, in order to advance the ion milling so that the milling particles P collide with the entire inner wall surface 44a of the hole 44 and the milling particles P do not collide with the bottom wall surface 44b of the hole 44, The arrangement of the support substrate 41 and the directivity of the milling particles P are very limited. Even if the milling particles P can be selectively incident only on the inner wall surface 44a of the hole 44, the milling particles P recoiled on the inner wall surface 44a often collide with the bottom wall surface 44b. . Furthermore, the metal particles 45 a ejected from the catalyst layer 45 by the collision of the milling particles P often adhere to the bottom wall surface 44 b of the hole 44. As a result, in the removal of the catalyst layer 45 by ion milling, it becomes difficult to control the thickness of the catalyst layer 45 formed on the bottom wall surface 44b and the density of the catalyst layer 45. After all, in recent years when the hole miniaturization progresses, it is difficult to suppress the growth of carbon nanotubes from the inner wall surface 44a of the hole 44. Therefore, the carbon nanotubes are selectively formed from the bottom wall surface 44b of the hole 44. Thus, development of a method capable of improving the electrical characteristics of the wiring structure is desired.

  Such a problem is not limited to the wiring structure having carbon nanotubes connecting the wiring layers formed above and below the insulating layer on the semiconductor substrate, and penetrates the insulating layer through the conductive layer in which the insulating layers are stacked. The wiring structure to which the carbon nanotubes are connected is generally common.

  The present invention has been made in view of the above circumstances, and an object thereof is to improve electrical characteristics in a wiring structure in which a carbon nanotube penetrating through an insulating layer is connected to a conductive layer in which the insulating layer is laminated. An object of the present invention is to provide a wiring forming method and a wiring forming apparatus using the method.

  Hereinafter, means for solving the above-described problems and the effects thereof will be described.

  The invention according to claim 1 is a wiring forming method in which carbon nanotubes are connected to a conductive layer, the step of forming a recess penetrating in the stacking direction in an insulating layer stacked on the conductive layer, and the recess Forming a catalyst layer on the insulating layer so as to cover the entire inner surface, a sheath is formed inside the recess, and the thickness of the sheath relative to the inner wall surface of the recess is relative to the bottom wall surface of the recess Using a step of generating plasma so as to be smaller than the thickness of the sheath and removing the catalyst layer formed on the inner wall surface by ions in the plasma, and the catalyst layer left on the bottom wall surface of the recess And a step of forming carbon nanotubes from the bottom wall surface.

  In the first aspect of the present invention, after the catalyst layer is formed on the entire inner wall of the recess, a plasma in which a sheath is formed in the recess is generated, and the catalyst layer formed on the inner wall surface of the recess Ions are collided. At this time, plasma is generated such that the thickness of the sheath with respect to the inner wall surface of the recess is smaller than the thickness of the sheath with respect to the bottom wall surface of the recess. Therefore, the collision of ions accelerated by the sheath preferentially proceeds with respect to the inner wall surface of the recess. As a result, it becomes easy to selectively leave the catalyst layer only on the bottom wall surface of the recess, and thus to make the dominant growth direction of the carbon nanotubes the stacking direction. And since the growth direction of a carbon nanotube can be arrange | equalized, it becomes possible to improve the electrical property of a wiring structure.

  According to a second aspect of the present invention, in the wiring forming method according to the first aspect, after the carbon nanotube is formed, a plasma is generated on the insulating layer, and the plasma is applied to the carbon nanotube protruding from the opening of the recess. The gist of the invention is to have a step of removing the tip side of the carbon nanotube by colliding ions therein.

  In the invention described in claim 2, the tip of the carbon nanotube is removed by collision of ions in the plasma. For this reason, among the carbon nanotubes, charge preferentially accumulates at a portion protruding from the opening of the recess, and thus the carbon nanotubes are sequentially removed from the tip side of the carbon nanotube. Therefore, it becomes easy to selectively remove only the carbon nanotubes protruding from the opening.

According to a third aspect of the present invention, in the wiring forming method according to the second aspect, the step of laminating a conductive layer connected to the carbon nanotube on the insulating layer after the step of removing the tip of the carbon nanotube. Furthermore, the gist is that the step of removing the tip of the carbon nanotube and the step of laminating a conductive layer on the insulating layer are performed in a common vacuum system.

  In the invention according to claim 3, even if the cap structure is opened by the carbon nanotube tip removal step, the air has the electrical characteristics of the carbon nanotubes reduced by avoiding air from entering the carbon nanotubes. Can be avoided.

  Invention of Claim 4 is the wiring formation method as described in any one of Claims 1-3, The said catalyst layer is formed from a lower catalyst layer and an upper catalyst layer, The said lower catalyst layer is It is formed of at least one of titanium, tantalum, vanadium, niobium, nitrides of these transition metals, and oxides of the transition metals, and the upper catalyst layer is formed of any of cobalt, nickel, and iron. This is the gist.

  The inventors of the present invention have a catalyst layer comprising a lower catalyst layer formed of at least one of titanium, tantalum, vanadium, niobium, nitrides of these transition metals, and oxides of the transition metals, and cobalt, nickel, and iron. It has been found that the temperature of the substrate when forming the carbon nanotubes can be lowered by forming from the upper catalyst layer formed by any of the above.

  According to the fourth aspect of the present invention, since the catalyst layer is formed by combining the metal layers described above, the temperature of the substrate when forming the carbon nanotube can be lowered. Therefore, according to the present invention, it is possible to reduce the thermal history when the wiring structure is formed.

The gist of a fifth aspect of the present invention is that the catalyst layer is formed by a sputtering method in the wiring formation method according to any one of the first to fourth aspects.
In the invention according to claim 5, since the catalyst layer is formed by the sputtering method, the catalyst is formed on the substrate having a large area, compared with the case where the catalyst layer is formed by another film forming method such as an electron beam evaporation method. It is possible to form layers and to form catalyst layers on a plurality of substrates at once.

The gist of a sixth aspect of the present invention is that the catalyst layer is formed by an electron beam evaporation method in the wiring forming method according to any one of the first to fourth aspects.
In the invention according to claim 6, since the catalyst layer is formed by the electron beam evaporation method, the catalyst layer is selected rather than forming the catalyst layer by other film forming methods such as a sputtering method or various CVD methods. Can be formed. Therefore, since it becomes easy to selectively form the catalyst layer with respect to the bottom wall surface of the recess from the time of formation of the catalyst layer, the catalyst formed on the inner wall surface of the recess by passing through the step of removing the catalyst layer. The layer is more reliably removed.
The gist of a seventh aspect of the present invention is that the conductive layer on the insulating layer is formed by a sputtering method in the wiring forming method according to any one of the third to sixth aspects.
The gist of an eighth aspect of the present invention is that the conductive layer on the insulating layer is formed by an electron beam evaporation method in the wiring formation method according to any one of the third to sixth aspects.
As in the methods according to the seventh and eighth aspects, the conductive layer of the wiring structure can be formed by sputtering or electron beam evaporation.

The invention according to claim 9 is a wiring forming apparatus for forming a wiring structure in which a carbon nanotube is connected to a conductive layer in which an insulating layer is laminated via a recess penetrating the insulating layer, A chamber in which a catalyst layer is formed on the insulating layer so as to include the entire surface, a sheath is formed inside the recess, and the thickness of the sheath relative to the inner wall surface of the recess is the sheath relative to the bottom wall surface of the recess Using a chamber that generates plasma to be smaller than the thickness of the inner wall surface and removes the catalyst layer formed on the inner wall surface by ions in the plasma, and the catalyst layer left on the bottom wall surface of the recess. And a chamber for forming carbon nanotubes from the bottom wall surface.

  According to the ninth aspect of the present invention, a chamber is provided that generates plasma in which a sheath is formed in the recess, and collides ions in the plasma with the catalyst layer formed on the inner wall surface of the recess. The chamber generates plasma such that the thickness of the sheath with respect to the inner wall surface of the recess is smaller than the thickness of the sheath with respect to the bottom wall surface of the recess. Therefore, the collision of ions accelerated by the sheath preferentially proceeds with respect to the inner wall surface of the recess. As a result, it becomes easy to selectively leave the catalyst layer only on the bottom wall surface of the recess, and thus to make the dominant growth direction of the carbon nanotubes the stacking direction. And since the growth direction of a carbon nanotube can be arrange | equalized, it becomes possible to improve the electrical property of a wiring structure.

The figure which shows schematic structure of the cluster apparatus by which the wiring formation apparatus of one Embodiment which actualized the wiring formation apparatus in this invention is mounted. The block diagram which shows schematic structure of the reverse sputtering chamber which comprises wiring formation apparatus. (A) (b) (c) (d) Process drawing which shows the formation process of a wiring structure in order. (A) (b) (c) (d) (e) Process drawing which shows the formation process of a wiring structure in order. (A) (b) Process drawing which shows the formation process of a wiring structure in order. The figure which shows typically the selective removal method of the conventional catalyst layer.

  Hereinafter, an embodiment embodying a wiring forming method and a wiring forming apparatus of the present invention will be described with reference to FIGS. First, a cluster device equipped with a wiring forming apparatus will be described with reference to FIG.

  A carry-in / out chamber 11, a catalyst layer forming chamber 13, a reverse sputtering chamber 14, a thermal CVD chamber 15, and a conductive layer forming chamber 16 are connected to the transfer chamber 12 of the cluster apparatus 10. Connected to each of the chambers 11 to 16 is an exhaust section composed of a vacuum pump and a pressure adjustment valve that depressurize the chambers 11 to 16 to a common predetermined pressure. The chambers 11 to 16 form a common vacuum system. When the substrate S to be processed is loaded / unloaded via the loading / unloading chamber 11, the loading / unloading chamber 11 is moved by the gate valve provided between the loading / unloading chamber 11 and the transfer chamber 12. , And are isolated from the other chambers 12-16. Therefore, the chambers 12 to 16 are maintained in a predetermined vacuum state.

  The carry-in / out chamber 11 carries the substrate S transferred from the outside into the cluster apparatus 10 and carries out the substrate S processed in the cluster apparatus 10 to the outside. The substrate S carried in from the carry-in / out chamber 11 is an insulation made of a copper wiring constituting a conductive layer on a silicon substrate, a tantalum nitride layer constituting a conductive layer having a diffusion preventing function, and a low dielectric constant material. The layers are sequentially stacked, and a hole is formed as a recess penetrating the insulating layer in the stacking direction. A plurality of holes are formed in the insulating layer, and each hole has a diameter of, for example, 160 nm and a depth of, for example, 200 nm.

  The transport chamber 12 transports the substrate S to the chambers 13 to 16 through the transport chamber 12, and transports the substrate S processed in the chambers 13 to 16 to the transport chamber 11. A robot 12a is mounted.

  The catalyst layer forming chamber 13 is mounted with a target made of titanium nitride and a target made of cobalt, and connected to a gas supply unit for supplying a sputtering gas, and also connected to a power source for generating plasma of the sputtering gas. ing. In the catalyst layer forming chamber 13, a titanium nitride layer and a cobalt layer as a catalyst layer are formed on the entire inner surface of the hole formed in the substrate S and the entire surface of the insulating layer.

  A substrate stage on which the substrate S is placed is built in the reverse sputtering chamber 14, and a gas supply unit for supplying a sputtering gas is connected to the reverse sputtering chamber 14. In the reverse sputtering chamber 14, sputtering gas plasma is formed, and a part of the catalyst layer formed on the substrate S is reversely sputtered. In the present embodiment, removing a part of the structure such as the catalyst layer once formed on the substrate S with sputtered particles is called reverse sputtering.

  The thermal CVD chamber 15 has a built-in substrate stage that holds the substrate S while heating it to a predetermined temperature, and is connected to a gas supply unit that supplies hydrocarbon gas to the substrate S. In the thermal CVD chamber 15, hydrocarbon gas is supplied in a state where the substrate S is heated to a predetermined temperature, and thereby carbon nanotubes are formed in the holes of the substrate S.

  The conductive layer forming chamber 16 is equipped with a target made of tantalum to which a gas supply unit for supplying a sputtering gas and a reactive gas is connected, and to which a power source for generating plasma of the sputtering gas is connected. In the conductive layer forming chamber 16, tantalum nitride as a conductive layer connected to the carbon nanotube is formed on the substrate S on which the carbon nanotube is formed by a reactive sputtering method.

  Next, the detailed configuration of the above-described reverse sputtering chamber 14 will be described with reference to FIG. A substrate stage 22 that holds the loaded substrate S is disposed inside the vacuum chamber 21 that constitutes the reverse sputtering chamber 14. A high frequency power supply 25 that outputs high frequency power having a frequency of 13.56 MHz, for example, is connected to the stage electrode 23 in the substrate stage 22 via a matching unit 24. The matching unit 24 includes a blocking capacitor for applying a negative bias voltage to the stage electrode 23, and matches the input impedance of the high frequency power supply 25 with the output impedance of the load. In addition, a shower plate 26 is installed above the vacuum chamber 21 so as to face the substrate stage 22. A sputter gas supply unit 27 that supplies a sputter gas, for example, an argon gas, is connected to the shower plate 26 through a gas supply port 21 a formed through the vacuum chamber 21.

  The vacuum chamber 21 is connected to an exhaust unit 28 that exhausts the fluid in the vacuum chamber 21 through an exhaust port 21 b formed through the vacuum chamber 21. The exhaust unit 28 includes various vacuum pumps and pressure control valves, and exhausts the fluid in the vacuum chamber 21 to the outside of the vacuum chamber 21 at a predetermined exhaust flow rate.

  When reverse sputtering is performed in the reverse sputtering chamber 14, the inside of the vacuum chamber 21 is set to a predetermined process pressure by the exhaust flow rate of the exhaust unit 28 and the flow rate of the gas supplied from the sputtering gas supply unit 27. The Next, the high frequency power supply 25 outputs high frequency power to the stage electrode 23, whereby argon gas plasma is generated between the substrate stage 22 and the shower plate 26. At this time, a negative bias voltage is applied to the stage electrode 23 as the plasma is generated, and positive ions contained in the argon gas plasma are attracted to the surface of the substrate S by such a bias voltage. Thereby, a part of the catalyst layer formed on the surface of the substrate S or the inner surface of the hole is reverse sputtered.

  Next, a wiring forming method performed in the cluster device 10 as described above will be described with reference to FIGS. 3 to 5, description will be made using a surface obtained by cutting the substrate S along a direction parallel to the stacking direction.

  In order to form a wiring structure in which two conductive layers are connected by carbon nanotubes, first, a substrate S as shown in FIG. 3A is prepared. That is, in the substrate S, first, on the support substrate 31 made of silicon, the lower wiring layer 32 made of copper constituting the conductive layer and the barrier layer made of tantalum nitride for preventing diffusion which also forms the conductive layer. 33 and an insulating layer 34 made of an insulating material having a low dielectric constant are sequentially stacked. Next, a plurality of holes 35 are formed in the insulating layer 34 as recesses that expose the surface of the barrier layer 33 by penetrating the insulating layer 34 from the surface 34 a in the stacking direction. The substrate S has a diameter of, for example, 8 inches. The diameter of each hole 35 formed in the substrate S is 160 nm, and the depth is 200 nm.

  When such a substrate S is loaded into the cluster apparatus 10 from the loading / unloading chamber 11, the substrate S is loaded into the catalyst layer forming chamber 13 via the transfer chamber 12 by the transfer robot 12a. When the substrate S is carried into the catalyst layer forming chamber 13, as shown in FIG. 3B, the entire inner surface of the hole 35 formed in the insulating layer 34, that is, the inner wall surface 35a and the bottom wall surface 35b, The lower catalyst layer 36 made of tantalum nitride is formed on the entire surface 34a of the insulating layer 34 so as to include the above. The lower catalyst layer 36 is formed with a thickness of 5 nm, for example.

  When the lower catalyst layer 36 is formed, the substrate S is transferred from the catalyst layer forming chamber 13 to the reverse sputtering chamber 14 by the transfer robot 12a. When the substrate S is carried into the reverse sputtering chamber 14, as shown in FIG. 3C, sputtered particles Sp, specifically argon ions, contained in the argon plasma formed in the vacuum chamber 21 are formed. Thus, the lower catalyst layer 36 is reverse sputtered.

At this time, among the process conditions of the reverse sputtering performed in the reverse sputtering chamber 14, the pressure in the chamber and the high frequency power are set so as to satisfy the following (a) and (b).
(A) A sheath Sh is formed inside the hole 35.
(B) The thickness Wa of the sheath Sh with respect to the inner wall surface 35a of the hole 35 is smaller than the thickness Wb of the sheath Sh with respect to the bottom wall surface 35b of the hole 35 (Wa <Wb).

  According to the reverse sputtering satisfying the above (a) and (b), the collision of the sputtered particles Sp accelerated by the sheath Sh causes the inner wall surface 35 a of the hole 35 and the surface of the insulating layer 34 to be more than the bottom wall surface 35 b of the hole 35. It advances preferentially with respect to 34a. As a result, as shown in FIG. 3D, the lower catalyst layer 36 can be selectively left only on the bottom wall surface 35 b of the hole 35. At this time, in the vicinity of the opening of the hole 35, most of the sputtered particles Sp accelerated by the sheath Sh are drawn into the inner wall surface 35a along the normal direction of the inner wall surface 35a. Therefore, in the vicinity of the opening of the hole 35, compared to the above-described ion milling method, particles sputtered by the sputtered particles Sp, particles sputtered by the sputtered particles Sp, and these are suppressed from traveling toward the bottom wall surface 35b. be able to.

The thickness distribution of the sheath Sh formed in the hole 35 varies depending on the shape of the hole 35. However, the thickness distribution of the sheath Sh can be generally adjusted by the plasma density at the time of reverse sputtering, and by extension, the pressure in the vacuum chamber 21 at the time of reverse sputtering and the magnitude of the high-frequency power output from the high-frequency power source 25. It is. The plasma conditions for obtaining the thickness distribution of the sheath Sh can be optimized according to the shape of the catalyst layer after reverse sputtering and the growth direction of the carbon nanotubes after reverse sputtering. Can also be obtained.
For example, when only the lower catalyst layer 36 on the bottom wall surface 35b is left with respect to the hole 35 having the above-described size, the following process conditions may be mentioned.
Argon flow rate: 1 sccm to 100 sccm
Process pressure: 0.1 Pa to 10 Pa
・ High frequency power: 10W to 1000W

  When the reverse sputtering of the lower catalyst layer 36 formed on the substrate S is completed, the substrate S is loaded again into the catalyst layer forming chamber 13 via the transfer chamber 12 by the transfer robot 12a. When the substrate S is carried into the catalyst layer forming chamber 13, as shown in FIG. 4A, the entire inner surface of the hole 35 formed in the insulating layer 34, that is, the inner wall surface 35 a, and the lower catalyst layer 36. The upper catalyst layer 37 of cobalt is formed on the entire surface 34a of the insulating layer 34 so as to include the bottom wall surface 35b on which is formed. The upper catalyst layer 37 is formed with a thickness of 1 to 1.5 nm, for example.

  In the present embodiment, a catalyst layer composed of the lower catalyst layer 36 and the upper catalyst layer 37 is formed. According to such a two-layered catalyst layer, the temperature of the substrate S when forming the carbon nanotube can be lowered. Therefore, it is possible to reduce the thermal history when forming the wiring structure.

  When the upper catalyst layer 37 is formed, the substrate S is loaded again into the reverse sputtering chamber 14 via the transfer chamber 12 by the transfer robot 12a. When the substrate S is carried into the reverse sputtering chamber 14, the upper catalyst layer 37 is reversed by the sputtered particles Sp contained in the argon plasma formed in the vacuum chamber 21 as shown in FIG. 4B. Sputtered. Even when the upper catalyst layer 37 is reverse sputtered, the sheath Sh is formed inside the surface 34a of the insulating layer 34 and the hole 35 in the same manner as when the lower catalyst layer 36 is reverse sputtered. Therefore, the collision of the sputtered particles Sp accelerated by the sheath Sh preferentially proceeds with respect to the inner wall surface 35 a of the hole 35 and the surface 34 a of the insulating layer 34 rather than the bottom wall surface 35 b of the hole 35. As a result, as shown in FIG. 4C, the upper catalyst layer 37 can be selectively left only on the lower catalyst layer 36 formed on the bottom wall surface 35 b of the hole 35.

  When the reverse sputtering of the upper catalyst layer 37 is completed, the substrate S is transferred to the thermal CVD chamber 15 via the transfer chamber 12 by the transfer robot 12a. When the substrate S is carried into the thermal CVD chamber 15, the substrate S is heated to a predetermined temperature, and hydrogen sensitized gas is supplied to the substrate S. As a result, as shown in FIG. 4 (d), the lower catalyst layer 36 and the upper catalyst layer 37 formed on the bottom wall surface 35b of the hole 35 are parallel to the stacking direction and from the catalyst layers 36, 37. Carbon nanotubes 38 extending in the direction of leaving are formed. At this time, the carbon nanotubes 38 are formed in such a length that the upper end portion 38 a protrudes from the opening of the hole 35 formed in the surface 34 a of the insulating layer 34.

  When the formation of the carbon nanotubes 38 is completed, the substrate S is loaded again into the reverse sputtering chamber 14 via the transfer chamber 12 by the transfer robot 12a. When the substrate S is carried into the reverse sputtering chamber 14, as shown in FIG. 4 (e), argon gas plasma is generated in the vacuum chamber 21, whereby the surface 34 a of the insulating layer 34 and the carbon nanotubes 38. A sheath Sh is formed so as to cover the upper end portion 38a.

At this time, in the upper end portion 38 a of the carbon nanotube 38, the charge is preferentially accumulated as it is closer to the cap 38 b that is the tip of the carbon nanotube 38. Therefore, since the sputtered particles Sp are easily drawn toward the cap 38b, the upper end portion 38a is removed in order from the cap 38b side. Therefore, it becomes easy to selectively remove only the carbon nanotubes 38 protruding from the surface 34 a of the insulating layer 34.

  Note that the sputtering rate of the carbon nanotubes sputtered by reverse sputtering is determined in advance by experiments or the like, and the reverse sputtering for removing the upper end portion 38a of the carbon nanotube 38 is continued for the time necessary for removing the upper end portion 38a. The Thereby, as shown in FIG. 5A, the upper end portion 38a of the carbon nanotube 38 is removed. Since the cap 38b of the carbon nanotube 38 is removed by removing the upper end portion 38a in this manner, that is, the carbon nanotube 38 is opened, the carbon nanotube 38 and the upper wiring layer 39 connected thereto are connected. The conductivity is improved at the connection site.

  When the upper end portion 38a of the carbon nanotube 38 is removed, the substrate S is transferred to the conductive layer forming chamber 16 via the transfer chamber 12 by the transfer robot 12a. When the substrate S is carried into the conductive layer forming chamber 16, as shown in FIG. 5B, the surface 34a of the insulating layer 34 and the carbon are sputtered by the target disposed in the conductive layer forming chamber 16. An upper wiring layer 39 made of tantalum nitride is formed at the tip of the nanotube 38. In the present embodiment, the open ends of the carbon nanotubes 38 and the formation of the upper wiring layer 39 formed so as to cover the upper part thereof are performed in a common vacuum system. Therefore, air can be prevented from entering the opened carbon nanotubes 38. Therefore, it is possible to avoid the electrical characteristics of the carbon nanotubes 38 from being deteriorated by air.

  Through the above-described steps, a wiring structure having carbon nanotubes 38, a lower wiring layer 32 that is a conductive layer connected by the carbon nanotubes 38, and an upper wiring layer 39 is formed.

As described above, according to the embodiment, the effects listed below can be obtained.
(1) After the lower catalyst layer 36 and the upper catalyst layer 37 are formed on the entire inner surface of the hole 35, plasma in which the sheath Sh is formed in the hole 35 is generated, and thereby the inner wall surface 35a of the hole 35 is generated. Sputtered particles Sp in the plasma collide with each of the catalyst layers 36 and 37 formed in (1). At this time, the thickness Wa of the sheath Sh with respect to the inner wall surface 35a of the hole 35 is smaller than the thickness Wb of the sheath Sh with respect to the bottom wall surface 35b of the hole 35. For this reason, the collision of ions accelerated by the sheath Sh progresses more preferentially to the inner wall surface 35a of the hole 35 than to the bottom wall surface 35b of the hole 35. As a result, it becomes easy to selectively leave the catalyst layers 36 and 37 only on the bottom wall surface 35b of the hole 35 and to make the dominant growth direction of the carbon nanotubes 38 be the stacking direction. Since the growth direction of the carbon nanotubes 38 can be aligned, the electrical characteristics of the wiring structure can be improved.

  (2) Before the upper wiring layer 39 is formed, the upper end portion 38a of the carbon nanotube 38 is removed by the collision of the sputtered particles Sp in the plasma. For this reason, the charge preferentially accumulates in the portion of the carbon nanotube 38 protruding from the opening of the hole 35, so that the carbon nanotube 38 is removed in order from the tip side of the carbon nanotube 38. Therefore, it becomes easy to selectively remove only the carbon nanotubes 38 protruding from the opening.

(3) The removal of the upper end portion 38a of the carbon nanotube 38 and the formation of the upper wiring layer 39 are performed in a common vacuum system. As a result, even if the cap 38b is opened by removing the upper end portion 38a of the carbon nanotube 38, it is possible to suppress the entry of air into the carbon nanotube 38, thereby reducing the electrical characteristics of the carbon nanotube 38 by the air. Can be suppressed.

  (4) Before the carbon nanotubes 38 are formed, a two-layered catalyst layer composed of a titanium nitride lower catalyst layer 36 and a cobalt upper catalyst layer 37 is formed. Thereby, when the carbon nanotube 38 is formed by thermal CVD, the temperature of the substrate S can be lowered. Therefore, it is possible to reduce the thermal history when forming the wiring structure.

  (5) The lower catalyst layer 36 and the upper catalyst layer 37 are formed by sputtering. Therefore, it is possible to form the catalyst layers 36 and 37 on the substrate S having a large area, rather than forming the catalyst layers 36 and 37 by other film forming methods, for example, an electron beam evaporation method, It is also possible to form the catalyst layers 36 and 37 for a plurality of substrates S at a time.

In addition, the said embodiment can be changed and implemented suitably as follows.
The support substrate 31 may be a substrate made of an insulator such as quartz or sapphire in addition to a substrate made of a semiconductor other than silicon. In short, any substrate may be used as long as a concave portion penetrating the insulating layer is formed in the conductive layer on which the insulating layer is laminated.

  The lower wiring layer 32 and the upper wiring layer 39 may be any conductive wiring layer, and may be a layer formed of other than copper or tantalum nitride, for example, a layer made of aluminum. In short, the lower wiring layer may be a conductive layer that can be laminated with a catalyst layer for carbon nanotubes, and the upper wiring layer may be a conductive layer that can be connected to carbon nanotubes.

  The barrier layer 33 is not limited to tantalum nitride, but may be formed of a material that can suppress diffusion of the metal element forming the lower wiring layer 32 and a conductive material.

  The diameter of the hole 35 is 160 nm and the depth of the hole 35 is 200 nm. However, the diameter of the hole 35 and the depth of the hole 35 are not limited to this. The shape of the hole 35 is not limited to a circular hole shape, and may be a rectangular hole shape or may be a multistage hole shape. In short, the recess formed in the insulating layer may be anything that penetrates the insulating layer and is connected to the conductive layer that is the lower layer of the insulating layer.

  The lower catalyst layer 36 is formed by sputtering a tantalum nitride target with argon. However, the present invention is not limited thereto, and the lower catalyst layer 36 may be formed by sputtering a tantalum target with a nitrogen-containing gas.

  The lower catalyst layer 36 and the upper catalyst layer 37 are formed by sputtering. However, the present invention is not limited to this, and the catalyst layers 36 and 37 may be formed by electron beam evaporation. According to this, the following effects can be obtained.

  (6) For example, the thickness of the catalyst layers 36 and 37 in the recesses is made larger on the bottom wall surface than when the catalyst layers 36 and 37 are formed by other film forming methods such as sputtering and various CVD methods. It becomes easy. Therefore, it becomes easy to make the film thickness of the catalyst layers 36 and 37 formed on the inner wall surface 35 a of the hole 35 in advance larger than the film thickness of the catalyst layers 36 and 37 formed on the bottom wall surface 35 b of the hole 35. Therefore, the catalyst layers 36 and 37 formed on the inner wall surface 35a of the hole 35 are more reliably removed through the process of removing the catalyst layers 36 and 37.

The lower catalyst layer 36 is made of titanium nitride and the upper catalyst layer 37 is made of cobalt. However, the present invention is not limited to this, and the lower catalyst layer 36 and the upper catalyst layer 37 may be a combination of the following metals and metal compounds. That is, the lower catalyst layer 36 may be formed of at least one of titanium, tantalum, vanadium, niobium, nitrides of these transition metals, and oxides of the transition metals. The upper catalyst layer 37 may be formed of any one of the cobalt, nickel, and iron. The combination of the lower catalyst layer 36 and the upper catalyst layer 37 formed of such materials can provide the same effect as that obtained by the combination of the lower catalyst layer 36 of titanium nitride and the upper catalyst layer 37 of cobalt. It becomes like this.
In addition, the combination of targets arranged in the catalyst layer forming chamber 13 may be changed according to the combination of the catalyst layers as described above.

In the catalyst layer forming chamber 13, both a target for forming the lower catalyst layer 36 and a catalyst layer for forming the upper catalyst layer 37 are disposed. However, the present invention is not limited thereto, and the cluster apparatus 10 may be provided with a chamber for forming the lower catalyst layer 36 and a chamber for forming the upper catalyst layer 37.
The frequency of the high frequency supplied from the high frequency power supply 25 to the stage electrode 23 may be arbitrarily set as long as reverse sputtering that satisfies the above conditions (a) and (b) is possible.

  Although the conductive layer forming chamber 16 for forming the upper wiring layer 39 is embodied as a sputter chamber, it may be embodied as a chamber for forming the upper wiring layer 39 by electron beam evaporation.

  The cluster apparatus 10 includes a single reverse sputtering chamber 14, but has a configuration in which each chamber has a separate chamber for reverse sputtering only the lower catalyst layer 36, the upper catalyst layer 37, and the carbon nanotube 38. May be. Alternatively, the cluster apparatus 10 may have a configuration in which a chamber that reversely sputters both the catalyst layers 36 and 37 and a chamber that reversely sputters the carbon nanotubes 38 are separately provided.

  The cluster apparatus 10 has a chamber for reverse sputtering the carbon nanotubes 38 separately from the chamber used for reverse sputtering of the catalyst layers 36 and 37, and the reverse sputtering of the carbon nanotubes 38 is performed in the chamber with argon gas and oxygen gas. The reverse sputtering may be performed by a mixed gas. According to such a configuration and method, the time required for removing the upper end portion 38a of the carbon nanotube 38 can be shortened.

  The upper end portion 38a of the carbon nanotube 38 is removed by the reverse sputtering method, but may be removed by other methods such as a chemical mechanical polishing method (CMP method) or an ion milling method. Even with such a method, it is possible to improve the electrical characteristics of the wiring structure as long as the growth direction of the carbon nanotubes is aligned with the stacking direction.

The conductive layer forming chamber 16 for forming the upper wiring layer 39 can be embodied as a device that does not constitute a vacuum system common to the thermal CVD chamber for forming the carbon nanotubes. In addition, the reverse sputtering chamber for removing the upper end portion 38a of the carbon nanotube 38 can also be embodied as an apparatus that does not constitute a common vacuum system with the thermal CVD chamber for forming the carbon nanotube. Even with such a configuration, the electrical characteristics of the wiring structure can be improved as long as the growth direction of the carbon nanotubes is aligned with the stacking direction. Even in such a configuration, if the reverse sputtering chamber for removing the upper end portion 38a and the conductive layer forming chamber 16 for forming the upper wiring layer 39 constitute a common vacuum system, they are opened. It is possible to suppress the intrusion of air into the carbon nanotube 38.

  If the thickness Wa of the sheath Sh with respect to the inner wall surface 35a of the hole 35 is smaller than the thickness Wb of the sheath Sh with respect to the bottom wall surface 35b of the hole 35, the sputtered particles Sp are drawn into the bottom wall surface 35b of the hole 35. May be. Even with this configuration, as long as the thickness Wa is smaller than the thickness Wb, it is possible to preferentially remove the catalyst layer on the inner wall surface 35a over the catalyst layer on the bottom wall surface 35b.

  If the sputtered particles Sp are drawn into the bottom wall surface 35b of the hole 35, it is preferable that the energy of the sputtered particles Sp is smaller than the energy required for sputtering of the catalyst layer. That is, the thickness Wa of the sheath Sh with respect to the inner wall surface 35a of the hole 35 is such a thickness that the sputtered particles Sp drawn into the inner wall surface 35a have energy necessary for sputtering of the catalyst layer. On the other hand, the thickness Wb of the sheath Sh with respect to the bottom wall surface 35b of the hole 35 is preferably such that the sputtered particles Sp drawn into the bottom wall surface 35b do not have the energy required for sputtering of the catalyst layer. Furthermore, the thickness Wb of the sheath Sh with respect to the bottom wall surface 35b of the hole 35 is such that the sputtered particles Sp, which are ions in the plasma, collide with other particles in the hole 35 and do not reach the bottom wall surface 35b of the hole 35. Is preferable. However, the lower catalyst layer 36 or the upper catalyst layer 37 formed on the inner wall surface 35a of the hole 35 is removed, and the lower catalyst layer 36 or the upper catalyst layer 37 formed on the bottom wall surface 35b of the hole 35 remains. For example, a method in which the sputtered particles Sp reach the bottom wall surface of the hole 35 may be used.

  In the reverse sputtering chamber 14, reverse sputtering of the substrate S was performed using argon gas. The gas used for plasma generation is not limited to this, and may be a rare gas other than argon gas, for example, helium gas, neon gas, krypton gas, xenon gas, or an inert gas other than rare gas. In short, the lower catalyst layer 36 or the upper catalyst layer 37 formed on the inner wall surface 35a of the hole 35 is removed by the sputtered particles Sp in the plasma, and the lower catalyst layer 36 or the upper catalyst layer 37 is formed by radicals in the plasma. Any gas that is not etched may be used.

  The wiring structure has carbon nanotubes 38 that connect the lower wiring layer 32 and the upper wiring layer 39 formed above and below the insulating layer 34 on the support substrate 31 formed of a semiconductor such as silicon. . For example, the wiring structure may be a structure in which the active element and the lower wiring layer 32 are connected by carbon nanotubes, and a carbon nanotube penetrating the insulating layer in a conductive layer in which insulating layers are stacked. Any structure can be used as long as it is connected.

  DESCRIPTION OF SYMBOLS 10 ... Cluster apparatus, 11 ... Carry-in / out chamber, 12 ... Transfer chamber, 12a ... Transfer robot, 13 ... Catalyst layer formation chamber, 14 ... Reverse sputter chamber, 15 ... Thermal CVD chamber, 16 ... Conductive layer formation chamber, 21 ... Vacuum Tank, 21a ... Gas supply port, 21b ... Exhaust port, 22 ... Substrate stage, 23 ... Stage electrode, 24 ... Matching unit, 25 ... High frequency power supply, 26 ... Shower plate, 27 ... Sputter gas supply unit, 28 ... Exhaust unit, 31, 41 ... support substrate, 32, 42 ... lower wiring layer, 33 ... barrier layer, 34, 43 ... insulating layer, 34a, 43a ... surface, 35, 44 ... hole, 35a, 44a ... inner wall surface, 35b, 44b ... Bottom wall surface, 36 ... lower catalyst layer, 37 ... upper catalyst layer, 38 ... carbon nanotube, 38a ... upper end, 38b ... cap, 39 ... upper wiring layer, 45 Catalyst layer, 45a ... metal particles, P ... milling particles, S ... substrate, Sh ... sheath, Sp ... sputtered particles.

Claims (9)

A wiring formation method in which carbon nanotubes are connected to a conductive layer,
Forming a recess penetrating in the stacking direction in the insulating layer stacked on the conductive layer;
Forming a catalyst layer on the insulating layer so as to cover the entire inner surface of the recess;
The sheath is formed inside the recess, and plasma is generated so that the thickness of the sheath with respect to the inner wall surface of the recess is smaller than the thickness of the sheath with respect to the bottom wall surface of the recess, and the plasma is formed on the inner wall surface. Removing the catalyst layer with ions in the plasma;
Forming a carbon nanotube from the bottom wall surface using the catalyst layer left on the bottom wall surface of the recess. And a step of generating plasma on the insulating layer after forming the carbon nanotubes, causing ions in the plasma to collide with the carbon nanotubes protruding from the openings of the recesses, and removing the tip side of the carbon nanotubes. Item 4. The wiring forming method according to Item 1. A step of laminating a conductive layer connected to the carbon nanotube on the insulating layer after the step of removing the tip of the carbon nanotube;
The wiring formation method according to claim 2, wherein the step of removing the tip of the carbon nanotube and the step of laminating a conductive layer on the insulating layer are performed in a common vacuum system. The catalyst layer is formed of a lower catalyst layer and an upper catalyst layer,
The lower catalyst layer is formed of at least one of titanium, tantalum, vanadium, niobium, nitrides of these transition metals, and oxides of the transition metals,
The wiring formation method according to claim 1, wherein the upper catalyst layer is formed of any one of cobalt, nickel, and iron. The wiring formation method according to claim 1, wherein the catalyst layer is formed by a sputtering method. The wiring formation method according to any one of claims 1 to 4, wherein the catalyst layer is formed by an electron beam evaporation method. The wiring formation method according to claim 3, wherein the conductive layer on the insulating layer is formed by a sputtering method. The wiring forming method according to claim 3, wherein the conductive layer on the insulating layer is formed by an electron beam evaporation method. A wiring forming apparatus for forming a wiring structure in which a carbon nanotube is connected to a conductive layer in which an insulating layer is laminated via a recess penetrating the insulating layer,
A chamber for forming a catalyst layer on the insulating layer so that the entire inner surface of the recess is included;
The sheath is formed inside the recess, and plasma is generated so that the thickness of the sheath with respect to the inner wall surface of the recess is smaller than the thickness of the sheath with respect to the bottom wall surface of the recess, and the plasma is formed on the inner wall surface. A chamber for removing the catalyst layer by ions in the plasma;
And a chamber for forming carbon nanotubes from the bottom wall surface using the catalyst layer left on the bottom wall surface of the recess.

Images