Plastic and nonmetallic article shaping or treating: processes – Direct application of electrical or wave energy to work – Producing or treating porous product
Reexamination Certificate
1998-11-19
2001-02-13
Tentoni, Leo B. (Department: 1732)
Plastic and nonmetallic article shaping or treating: processes
Direct application of electrical or wave energy to work
Producing or treating porous product
C264S041000, C264S083000, C264S216000, C264S272170, C264S310000
Reexamination Certificate
active
06187248
ABSTRACT:
BACKGROUND OF THE INVENTION
The electronics industry utilizes dielectric materials as insulating layers between circuits and components of integrated circuits and associated electronic devices. Line dimensions are being reduced in order to increase speed and storage capability of microelectronic devices (e.g., computer chips). Microchip dimensions have undergone a significant decrease even in the past decade such that line widths previously >1 micron are being decreased to 0.25 microns, with future plans on the drawing boards of as low as 0.07 microns. As the line dimensions decrease, the requirements for preventing signal crossover (crosstalk) between the chip components become much more severe. These requirements can be summarized by the expression RC, where R=resistance of the conductive line and C=capacitance of the dielectric layer. Decreasing the dielectric constant will decrease the capacitance of the interlayer which will aid in minimizing RC.
Historically, silica with a dielectric constant of 4.2-4.5 has been employed as the interlayer dielectric (ILD). However at line dimensions of 0.25 microns and less silica is no longer acceptable, and is expected to be replaced by polymers as the ILD material of choice. Representative polymers which meet the demanding requirements for an ILD material include the poly(arylene ethers) with dielectric values in the range of 2.6-2.8, such as that described in U. S. Pat. No. 5,658,994. Ultimately materials with dielectric constant values of 2.0 or less will be required as the line dimensions continue to decrease. The only polymers (nonporous) which reach values near or below 2.0 are fluorocarbon polymers. Fluorocarbon polymers have other drawbacks such as poor adhesion, potential reaction with metals at high temperature, poor rigidity at high temperature, and in some cases lower thermal stability than acceptable. In order to achieve the desired property characteristics and low dielectric constant values, nanoporous polymeric materials may be used.
Incorporation of porosity to polymeric materials is long known to reduce the dielectric constant of polymers (L. Mascia, The Role of Additives in Plastics, 1974). Microporous polymers (e.g., cyanate ester resins) have been noted by Kiefer et al. (Macromolecules, 29 (1996) 8546) to exhibit reduced dielectric constant values versus dense precursors. The resultant pore diameters were in the range of 10-20 microns which is orders of magnitude too large for line widths <0.25 microns. If polymeric materials are going be used as the interlayer in applications which require dielectric constant &egr;<2.0 along with excellent stability, excellent adhesion, compatibility with the integration process (e.g., reactive ion etching, fabrication of thin films), ability to have global planarization, low water sorption, non-reactivity to metals, and constant dielectric constant up to and exceeding 1 GHz, then nanoporous films will be one of the alternatives to be considered.
The design of nanoporous polymers for these applications was reported by Hedrick and coworkers at IBM in several papers. Their procedure involved the synthesis of diblock copolymers comprised of a high T
g
, thermally stable block and a lower T
g
, thermally labile block. The resultant material phase separates into well-defined micellar domains. Upon exposure to temperatures above the decomposition temperature of the thermally labile block, but below the T
g
and thermal decomposition temperature of the high T
g
block, thermolysis of the former would occur to produce a nanoporous structure. Hedrick et al. (Polymer, 34 (1993) 4717) also noted the ability to generate porous polymers with nanometer sized pores using the above described technique employing poly(phenylquinoxaline) as the high T
g
stable block and poly(propylene oxide) or poly(methyl methacrylate) as the labile block. Charlier et al. (Polymer, 36, No. 5 (1995) 987) used this procedure employing poly(propylene oxide) as the labile block to produce nanoporous polyimides. Hedrick et al. (Chem. Mater., 10 (1998) 39) described polyimide nanofoams using aliphatic polyester based thermally labile blocks (e.g., poly(&egr;-caprolactone)) in polyimide block copolymers. Pore sizes of 6-7 nm were reported. While this procedure yields a nanoporous material, the dielectric constants reported are primarily above 2.0 and there appears to be a practical limit to the amount of porosity that can be achieved via this approach. The choice of polymeric materials with T
g
above 400 ° C., desired to be even >450° C., are very limited and generally involve very polar materials and/or materials with higher water sorption than desired.
One method for producing porous polymeric materials is termed phase inversion, commonly employed in the production of microporous membranes. The phase inversion process to produce porous membranes has been well described in the literature, including the book by Kesting and Fritzsche (Polymeric Gas Separation Membranes, 1993). The phase inversion process can involve a non-solvent induced phase separation or a temperature induced phase separation to produce pores in the micron to nanometer range. Saunders et al. (ASME, 53 (1994) 243) noted that spin coating on silicon wafers followed by non-solvent induced phase separation yielded microporous polyimide films with a reduced dielectric constant. A polyimide (Ultem 1000™) was spin coated onto a silicon wafer from a solution of 1,3-dimethoxybenzene followed by immersion in toluene, a non-solvent for the polymer. A specific example indicates film thickness of 22 microns with porosity of 68% and a pore size of 1.4 microns. The dielectric constant decreased from 3.15 for the dense film to 1.88 for the porous film. While this reference denotes a porous film with a reduced dielectric constant produced via non-solvent induced phase separation, the pore sizes being almost two orders of magnitude too large are considerably outside the realm of interest for present or future low dielectric constant interlayer materials. Therefore the methodology employed by this reference requires major changes in order to be able to achieve the desired film properties for an interlayer dielectric in microelectronic applications (e.g., computer chips). Young, et al., Desalination 103 (1995) pp 233-247 studied solvent based phase inversion for porous polymers and used combinations of solvents and nonsolvents to evaluate pore formation.
The prior art suffers from several disadvantages for using phase inversion for low dielectric films for microelectronics. One of the basic problems of the phase separation and spin coating process is that low viscosity (e.g., low solids content) solutions are required to yield thin films (~1&mgr;) with uniform wafer coverage. However the phase separation process using low viscosity, low solids solutions yields large pore structures; utilizing high solids solutions (>20 vol %) leads to exorbitantly thick films and inconsistent film thicknesses along the substrate. Also, the ability to tune porosity to a desired level with essentially constant film thickness is not possible with the current procedure due to the low viscosity requirements for the spin-on solution and the required non-volatile nature of the solvent, e.g., to prevent phase separation during casting. Another characteristic inherent with the prior art process when employing a low solids solution (as required for spin casting) is that the phase separation process leads to loss of adhesion to the desired substrate, often resulting in films detaching from the substrate during the non-solvent phase separation process. Poor adhesion is not acceptable and poor mechanical properties may lead to failure of the resulting interlayer during integration processing. A final problem which must be addressed for nanoporous ILDs arises from the temperature exposure inherent with typical microelectronic device manufacturing, e.g., up to 450° C. For a polymeric porous ILD with a T
g
below processing conditions, collapse of the porous structure will o
Burgoyne, Jr. William Franklin
Langsam Michael
O'Neill Mark Leonard
Robeson Lloyd Mahlon
Air Products and Chemicals Inc.
Chase Geoffrey L.
Tentoni Leo B.
LandOfFree
Nanoporous polymer films for extreme low and interlayer... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Nanoporous polymer films for extreme low and interlayer..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Nanoporous polymer films for extreme low and interlayer... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2564702