Abrading – Machine – Combined
Reexamination Certificate
2002-05-16
2004-01-13
Wilson, Lee D. (Department: 3723)
Abrading
Machine
Combined
C451S288000, C451S285000, C451S057000
Reexamination Certificate
active
06676492
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field
This invention relates to chemical mechanical polishing, and more particularly to improved chemical mechanical polishing with improved reproducibility, versatility, productivity, robustness, and low cost.
2. Prior Art
CMP is uniquely capable of removing thick metal films while leaving intact features inset and surrounding dielectric films. This process has become an enabling technology for both advanced tungsten plug and copper demanscence process. It is as crucial as metal deposition or lithography aiming to achieve global planarity. CMP is no longer a niche application with the same fixed equipment, material, and process for all various device designs, material and process selections. In particular, device miniaturization and the coming of multi-metal architectures and techniques such as the emerging copper dual damascene are seriously challenging. These challenges force CMP technology including platforms, chemistries, pads and slurries to rapidly and radically evolve and improve.
Current CMP is not perfect. It must be carefully controlled for it to be optimized. A poorly executed CMP can generate extreme metal dishing in wide structures or dielectric erosion in high-density regions of smaller features. Abrasive particle containing slurries generate scratches or gouges in the inlaid structures. If a substrate is improperly post-CMP cleaned, the slurry particles can be included in subsequent dielectric deposits and depress yield. The process also reveals and highlights preexisting defects such as seams and voids encapsulated in vias and trenches during earlier copper electroplating processes, or delamination of barrier/seed layers from surrounding dielectrics not easily detected in preceding operations. Particles trapped into underlying dielectrics, barrier and seed films will appear only after the metal that coats them is removed. Shifts in film microstructure through the thickness of the deposit may affect CMP rates, or lead to effects such as pull-out of grains causing pits in the surfaces or perimeters of the inlaid metal.
CMP technology including equipment, material and processes cannot meet the needs for metal thinning, planarization and defect elimination. There is no slurry meeting all the CMP requirements. Additional problems exist as to stability and shelf life of the slurry, lot-to-lot variability of certain slurry products. First-step cannot always stop at the barrier layer. Second-step polish often introduces dishing, erosion, and non-uniformity.
CMP's next challenges include:
1) Adapting to smaller device features and large wafer sizes such as 300 mm;
2) Smart processing automation such as with real-time, in-situ monitoring and feed-back control, and computerized R&D for self-optimized process control;
3) Minimizing defect formation including planarity, metal thinning, nonuniform polishing, erosion, corrosion, pits, delamination, planarization, oxide and total metal loss, scratches, ruptures, topography issues with damascene structures, excessive edge exclusion below 3 mm, too much down-force pressure during CMP especially with copper and ultralow-k dielectric materials, and other damages;
4) Minimizing copper and oxide loss in double-damascene process;
5) Tailoring and integrating equipment, material, and process to new materials such as low-k films for low cost but with minimum size and complexity, maximum productivity, endpoint control, design flexibility, improved deposition rates, versatility, reliability, and robustness; and
6) Improving slurry stability, uniformity, deterioration during processing, shelf life, and lot-to-lot variability, all customized to meet specific process needs. Special attention should be paid to possible gel formation and agglomeration of the slurries, definite knowledge of chemistry and particle interaction, time-sensitive chemicals like an oxidizer, consistent concentration of the delivered slurry, controlling particle size-distribution in mixed powders and excessive settling associated with certain particles. These improvements are necessary for rapid development, characterization, and optimization of a specific robust CMP equipment and process for each customer's product.
The CMP method of surface planarization is a dominant technology in polishing glass. It also meets planarization requirements in the <0.35 Dm (micron) feature sized multi-level devices and interconnects in the semiconductor industry. The CMP method is a preferred technology to carry out global planarization for various integrated circuits (IC). Planarized surfaces have become key to the success of advanced semiconductor devices and circuits, particularly for high-density multi-level interconnects.
In IC manufacturing, CMP involves competing requirements at various length scales, e.g., uniform removal at the wafer scale, but non-uniform removal of protruding surfaces or areas to achieve planarization at the feature scale. The process, developed so far through trial-and-error, involves a synergistic interaction of many factors: fluid flow, fluid chemistry, slurry particle material, surface dissolution, and wafer material. Ideally, the grinding and polishing method and equipment should provide: high uniformity and selectivity, low defect levels, high removal rate, low-pressure/high-speed capability, short product development time, and low cost. Also, the solid grinding/polishing abrasive materials should always remain as sharp as possible (always sharp), efficient, long-lasting, and low in initial and operating costs for rapid, reproducible grinding and polishing operations.
Grinding, polishing, or planarizing is widely used in many industries such as automotive, electronics, optical, machinery, metallurgical, medical, and glass. The quality and performance of an automobile, electronic components, optical instruments, precision machinery, glass plate, metallurgical material, or biomedical samples often critically depend on the cost and quality, e.g., flatness, surface finish, and reproducibility, of the planarized material. A perfectly planarized sample is often not available, too costly, or even impossible to obtain.
Making a modern 0.25 Dm CMOS IC chip requires 13 planarizing steps. A single major defect in any one step can result in the rejection of the entire chip lot. Even if each planarizing step has a yield of 99%, the final product yield loss from the 13 planarizing steps alone is over 12.2%. Raising yields from 99% to 99.5% in the planarizing steps still incurs a planarizing loss of 6,3%. This is still a big production and financial loss.
A planarizing machine is often used to obtain a planar, smooth outer surface on a material. The prior-art grinding or polishing machine often comprises a rotating wheel for mounting the material thereon. A colloidal liquid or liquid abrasive suspension is provided to wet the wheel and to hold/mount the material against the rotating wheel. The liquid suspension comprises a liquid suspension medium and a plurality of solid abrasive particles suspended therein. The liquid suspension is fed onto the wheel to chemically and mechanically grind or polish off surface layers of the mounted material. Both manual and automatic planarizing machines have been known in the art for quite some time.
But these machines are not satisfactory in many respects. The liquid suspension is costly but it is not reproducible; has short shelf lives; deteriorates in performance during use, transit, or even storage; and does not reliably produce quality product results. The solid abrasive particles wear out rapidly degrading the planarizing results. The solid abrasive particles also often agglomerate or break up into smaller pieces. Changes in particle size alone lead to loss of control of the desired surface finish. A large size distribution of the solid abrasive particles produces a wide variety of surface finishes of differing smoothness and qualities, hampering product yield and reproducibility.
The hard, sharp, fragile, and brittle working edges and points on the solid abrasive particles are ea
Poulos, III James A.
Wilson Lee D.
LandOfFree
Chemical mechanical polishing does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Chemical mechanical polishing, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Chemical mechanical polishing will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3210951