Patent Grant
8017025
The invention relates to a method for producing air gaps in a target layer successively comprising:
Nowadays, integrated electronic circuits present a non-negligible proportion of metallic elements. The latter can be used as interconnection elements, antennas, coils or inductors. With the reduction of the distances separating the metallic layers, the electrostatic or electromagnetic couplings increase between these elements, which has the effect of modifying their electrical behavior. In a general manner, these electrostatic couplings are modelled by the appearance of at least one stray capacitance between the metallic elements. Reducing the coupling requires the dielectric permittivity of the material situated between said metallic elements to be reduced.
For current generations of integrated circuits, the use of low-permittivity dielectric materials (about 2.3 to 2.7) is compatible with the performances sought for. These low-permittivity dielectric materials advantageously replace silicon oxides. However, for the forthcoming technological generations (nodes smaller than 32 nm), the performance requirements are considerably greater than the small improvement of the permittivity currently measured on deposited bulk materials. The use of air gaps in interconnection structures today seems inevitable. With such architectures, the air gaps present between the conductor lines give permittivities equal to 1.
A possible means at the present time for fabricating these air gaps is the use of non-conformal deposition of dielectric on interline layers of metallic material. This approach, described by Gosset et al. (“General review of issues and perspectives for advanced copper interconnections using air gap as ultra-low K material”, Proceedings of the IEEE, 2003, International Interconnect Technology Conference (2003) 65), enables the required air gaps to be obtained under certain fabrication conditions with an excess of complex and costly technological steps.
Another envisaged solution consists in integrating a sacrificial material between the metallic material patterns, thereby forming the intermetallic level. This material reacts after it has been integrated by means of various physico-chemical processes (thermal annealing, UV radiation treatment (“Benefits and Trade-offs in Multi-Level Air Gap Integration”, Spring MRS San Diego, April 2006), or chemical etching (U.S. Pat. No. 7,172,980). In this way, integration of the interconnection levels with the sacrificial material can be totally or partly achieved, and the air gaps can then be formed. However, this approach greatly limits the scope of materials able to be used for performing integration. The sacrificial material which has to make way for the air gap does in fact have to be compatible with semi-conductor technology, be degradable and, when it is removed, the physico-chemical processes used must not damage the electronic devices that have already been integrated (CMOS transistors, capacitors . . . ). Likewise, the materials adjacent to the sacrificial material must present a resistance to these physico-chemical processes in order not to be damaged, but some of these materials have to allow degradation and removal of the sacrificial material by being for example permeable to the multiple chemical processes.
The object of the invention is to provide a sequence of technological steps aiming to alleviate the above-mentioned shortcomings. More particularly, the object of the invention is to provide a fabrication method that is less complex and less costly.
According to the invention, this object is achieved by the fact that, before deposition of the insulating layer, the method comprises formation of nanotubes on the target layer, a majority of nanotubes passing through the insulating layer and forming passages between the target layer and the outside.
Other advantages and features will become more clearly apparent from the following description of particular embodiments of the invention given as non-restrictive examples only and represented in the accompanying drawings, in which:
In the first particular embodiment, as illustrated in
As illustrated in
As illustrated in
The structure thus formed is subjected to degradation agents 10 of the degradable material 5. This degradation can for example be performed by any known method, for example by a thermal method, by radiation (for example ultraviolet radiation) or by any suitable undulatory phenomenon or by chemical etching by an agent in liquid or gaseous form.
As illustrated in
Advantageously, if completely hollow nanotubes 6 are to be produced by catalytic means, the catalyst grains used for formation thereof can be eliminated for example by means of an acid solution (for example nitric acid HNO3).
Moreover, the nanotubes 6 can also be partially degraded before or during degradation of the degradable material 5 so as to enlarge the passage formed by the nanotubes 6. The nanotubes 6 can also be totally eliminated before degradation of material 5 is performed, the passages then being formed by holes present in the insulating layer 7. The nanotubes 6 can then be eliminated for example by plasma or by chemical solution depending on their nature. For example, for carbon nanotubes, an oxygen-based plasma will be used.
Formation of passages between the target layer 4 and the outside, formed in the insulating layer 7 by the nanotubes or corresponding holes, thus enables a chemical agent 10 to reach and degrade the degradable material 5 and, after degradation of material 5, enables the degradation residues to be eliminated through the insulating layer 7 to create air gaps 10 at the foot of each pass-through nanotube, as illustrated in
In the case where the degradable material 5 is formed by silicon oxide, degradation thereof can be achieved by means of hydrofluoric acid. The material of layers 3 and 7 is then chosen such as not to react with hydrofluoric acid and layers 3 and 7 are then for example made from silicon nitride or silicon carbide. Layer 7 can for example be made from BD1® marketed by Applied Materials.
In a second embodiment illustrated in
For example, to integrate air gaps in a metallic interconnection structure, the target layer 4 is formed by an alternation of patterns made from the degradable material 5 (for example silicon oxide) and a predefined material 8, for example metallic (for example copper). This alternation is performed in known manner, for example by the damascene method. Layer 2 can then comprise a set of active electronic devices, for example CMOS transistors, arranged on a silicon substrate 1. An encapsulation and planarization layer 3 can be arranged on this layer 2.
As illustrated in
The diameter of nanotubes 6 is chosen according to the required application and according to the degradation method and the sub-products arising therefrom, typically between 1 nm and 300 nm. In fact, the wider nanotube 6 is, the easier diffusion of the chemical species (agent 10) is. Likewise, the degradation sub-products are liable to be removed more rapidly if nanotube 6 is wide.
If very localized etching is required, the use of thin nanotubes 6 and of suitable degradation methods (relatively short chemical etching or annealing time) will be favored. If on the contrary, broad gaps are sought to be achieved under nanotubes 6, then wide nanotubes 6 will be produced, possibly with a highly concentrated chemical degradation agent 10, or a longer etching time will be used. The person skilled in the art will adapt the etching time to the chemical etching solution concentration.
The density of nanotubes 6 is also chosen according to the required application, the density of nanotubes 6 preferably being comprised between 0.3 nanotubes/nm2 and 10−4 nanotubes/nm2. If the nanotube density is too great, the insulating layer 7 does not coat each nanotube individually, and in this case the etching solution diffuses between the nanotubes 6.
As illustrated in
Advantageously, the insulating layer 7 is of Methylsilsesquioxane type and more particularly made from BD1® material marketed by Applied Materials. If the insulating layer 7 completely coats the nanotubes 6, an additional step can be implemented to free one end of nanotubes 6.
Once this architecture has been obtained, degradation of the degradable material 5 is performed as before. Degradation and elimination of material 5 is preferably total. As illustrated in
In a variant of this embodiment, material 5 is thermally degradable, the other materials being able to remain unchanged. The degradable material 5 is then for example of polymer type (for example made of polymethyl methacrylate PMMA, of pore-forming type made of Norbornadiene or alpha-terpinene . . . ). In this approach, the structure is then heated so as to perform thermal degradation of the degradable material 5. The degradation sub-products can then diffuse out of target layer 4 by means of the nanotubes 6 acting as passages.
In a third embodiment illustrated by
The additional degradable material 9 is preferably chosen such as to facilitate deposition of the nanotubes 6. For example, the degradable material 5 can be a thermally degradable material such as PMMA, p-SILK®, Spin On Glass or PhiLK®) particularly suitable for the required application. The additional degradable material 9 is then for example a chemically degradable material such as silicon oxide, which makes production of the nanotubes 6 much easier than the degradable material 5. The carbon nanotubes 6 are then formed on the additional degradable material 9.
As illustrated in
Superposition of two degradable materials having different degradation characteristics enables the degradation rate and the formation conditions of nanotubes 6 to be optimized.
| Number | Date | Country | Kind |
|---|---|---|---|
| 07 03487 | May 2007 | FR | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 7172980 | Torres et al. | Feb 2007 | B2 |
| 20050124172 | Townsend, III et al. | Jun 2005 | A1 |
| 20060278901 | Dangelo et al. | Dec 2006 | A1 |
| 20070248758 | Ward et al. | Oct 2007 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20080283497 A1 | Nov 2008 | US |
