Integration Issues for Low Dielectric Constant Materials in each Generation of ULSI'S

1999 ◽  
Vol 565 ◽  
Author(s):  
Hideki Gomi ◽  
Koji Kishimoto ◽  
Tatsuya Usami ◽  
Ken-ichi Koyanagi ◽  
Takashi Yokoyama ◽  
...  

AbstractThe technologies utilizing Fluorinated Silicon Oxide (FSG, k=3.6) and Hydrogen Silsesquioxane (HSQ, k=3.0) have been established for 0.25-μm and 0.18-μm generation ULSIs. However, low-k materials for the next generation ULSIs, which have a dielectric constant of less than 3.0, have not become mature yet. In this paper, we review process integration issues in applying FSG and HSQ, and describe integration results and device performance using Fluorinated Amorphous Carbon (a-C:F, k=2.5) as one of the promising low-k materials for the next generation ULSIs.

1999 ◽  
Vol 564 ◽  
Author(s):  
Hideki Gomi ◽  
Koji Kishimoto ◽  
Tatsuya Usami ◽  
Ken-ichi Koyanagi ◽  
Takashi Yokoyama ◽  
...  

AbstractThe technologies utilizing Fluorinated Silicon Oxide (FSG, k=3.6) and Hydrogen Silsesquioxane (HSQ, k=3.0) have been established for 0.25-µm and 0.1 8-µm generation ULSIs. However, low-k materials for the next generation ULSIs, which have a dielectric constant of less than 3.0, have not become mature yet. In this paper, we review process integration issues in applying FSG and HSQ, and describe integration results and device performance using Fluorinated Amorphous Carbon (a-C:F, k=2.5) as one of the promising low-k materials for the next generation ULSIs.


1997 ◽  
Vol 476 ◽  
Author(s):  
Nigel P. Hacker ◽  
Gary Davis ◽  
Lisa Figge ◽  
Todd Krajewski ◽  
Scott Lefferts ◽  
...  

Low dielectric constant materials (k < 3.0) have the advantage that higher performance IC devices may be manufactured with minimal increases in chip size. The reduced capacitance given by these materials permits shrinking spacing between metal lines to below 0.25 μm and the ability to decrease the number of levels of metal in a device. The technologies being considered for low k applications are CVD or spin-on of inorganic or organic polymeric materials. Traditional spin-on silicates or siloxanes have been used as planarizing dielectrics during the last 15 years and usually have k > 3.0.


1999 ◽  
Vol 565 ◽  
Author(s):  
Y. Shimogaki ◽  
S. W. Lim ◽  
E. G. Loh ◽  
Y. Nakano ◽  
K. Tada ◽  
...  

AbstractLow dielectric constant F-doped silicon oxide films (SiO:F) can be prepared by adding fluorine source, like as CF4 to the conventional PECVD processes. We could obtain SiO:F films with dielectric constant as low as 2.6 from the reaction mixture of SiH4/N2 O/CF4. The structural changes of the oxides were sensitively detected by Raman spectroscopy. The three-fold ring and network structure of the silicon oxides were selectively decreased by adding fluorine into the film. These structural changes contribute to the decrease ionic polarization of the film, but it was not the major factor for the low dielectric constant. The addition of fluorine was very effective to eliminate the Si-OH in the film and the disappearance of the Si-OH was the key factor to obtain low dielectric constant. A kinetic analysis of the process was also performed to investigate the reaction mechanism. We focused on the effect of gas flow rate, i.e. the residence time of the precursors in the reactor, on growth rate and step coverage of SiO:F films. It revealed that there exists two species to form SiO:F films. One is the reactive species which contributes to increase the growth rate and the other one is the less reactive species which contributes to have uniform step coverage. The same approach was made on the PECVD process to produce low-k C:F films from C2F4, and we found ionic species is the main precursor to form C:F films.


1996 ◽  
Vol 443 ◽  
Author(s):  
Neil H. Hendricks

AbstractFor over two years, intensive efforts at SEMATECH and elsewhere have focused on identifying low dielectric constant (low ε) materials which possess all of the required properties and processing characteristics needed for integration into standard IC fabrication lines. To date, no material candidate has been shown to satisfy this impressive list of requirements. For some candidates, drawbacks related to material properties such as poor thermal stability or electrical performance have been identified; in other cases, problems in process integration, for example difficulties in patterning have stalled progress.In this paper, most of the current leading candidates for the low ε IC IMC application are identified and discussed. An attempt is made to correlate structure/property relationships in these materials with their relative attributes and deficiencies as they relate to the IMD application. Key differences in chemistry and property/processing characteristics are contrasted for low c silicon-oxygen polymers and for purely organic polymers. Novel dielectrics such as porous organic and inorganic thin films are also discussed in terms of their properties and associated process integration challenges. Since the needs for global planarization and low c IMD are occurring within roughly the same generation of minimum feature size (˜ 0.25 μm), the chemical mechanical polishing (CMP) of low dielectric constant thin films and/or of SiO2 layers deposited above them is briefly discussed. Both subtractive metalization and damascene processes are included, and the required low dielectric constant film properties and processing characteristics are contrasted for each process. Finally, the author's views on future trends in low dielectric constant materials development are presented, with an emphasis on identifying the types of chemical structures which may prove viable for this most demanding of all polymer film applications.


1998 ◽  
Vol 511 ◽  
Author(s):  
T. E. F. M. Standaert ◽  
P. J. Matsuo ◽  
S. D. Allen ◽  
G. S. Oehrlein ◽  
T. J. Dalton ◽  
...  

ABSTRACTThe patterning of several novel low dielectric constant (K) materials has been studied in a high-density plasma (HDP) tool. Recent results obtained on oxide-like materials, such as fluorinated oxide, hydrogen silsesquioxane (HSQ), and methyl silsesquioxane (MSQ), are reviewed. These materials can be successfully patterned using a fluorocarbon etching chemistry. The etching is in this case controlled by a thin fluorocarbon film at the surface. The patterning of polymer dielectrics can be performed in an oxygen etching chemistry. As an example, the patterning of Parylene-N in an oxygen chemistry is discussed. In this case, the ion and the oxygen radical flux need to be properly controlled to obtain a directional etching process. After the dielectric etch, either in a fluorocarbon or oxygen based chemistry, fluorocarbons and oxygen contamination remain at the exposed metal surfaces. We recently demonstrated how a plasma treatment following the dielectric etch reduces these contaminants. The results of this treatment on copper surfaces and the resulting modification to the dielectric are reviewed.


1999 ◽  
Vol 565 ◽  
Author(s):  
J. N. Bremmer ◽  
D. Gray ◽  
Y. Liu ◽  
K. Gruszynski ◽  
S. Marcus

AbstractLow dielectric constant hydrogen silsesquioxane films were achieved by rapid thermal cure processing with production viable equipment. A reduced dielectric constant of k = 2.5–2.6 is demonstrated by optimizing rapid thermal cure process conditions to produce low density hydrogen silsesquioxane thin films. This is a significant reduction relative to production proven furnace cure processed hydrogen silsesquioxane with k = 2.9. Concurrent with reduced k performance is a characteristic film expansion which contributes to formation of a low density structure. A mechanism for film expansion and relevance to low k performance is described; and issues relative to integration of rapid thermal processed low k hydrogen silsesquioxane are discussed.


2014 ◽  
Vol 1692 ◽  
Author(s):  
Maxime Darnon ◽  
Nicolas Posseme ◽  
Thierry Chevolleau ◽  
Thibaut L. David

ABSTRACTTo improve the integrated circuits’ performance and continue the downscaling of dimensions, it is necessary to use low dielectric constant materials as interconnects insulators. Current porous SiCOH low-k dielectrics are now reaching their limits since their porosity enables the diffusion of species that modify the inner surface of the pores. To further reduce the dielectric constant, it is necessary to change paradigm in interconnects fabrication. In this paper, we discuss the most promising innovations in terms of process, materials and architectures to reduce the interconnects insulators dielectric constant.


Sign in / Sign up

Export Citation Format

Share Document