The 'via first-trench second' dual damascene technology is currently being explored by several major semiconductor manufacturers due to lithography constraints of printing small contacts on extremely non-planar topology (trench first technology). Typical via holes are 0.30 - 0.50 micrometer and 0.18 - 0.25 micrometer with aspect ratios of 3 to 6 for i-line and DUV exposures, respectively. The novel approach utilizes an organic material to fill via holes to a desired level with some planarization of the topographic pattern. Numbers of novel polymers have been synthesized and evaluated to fulfill the requirements for the dual damascene process. These polymers showed good coating and planarizing properties. By modifying the formulations such as polymer molecular weight, viscosity, solvents, and cross linker and thermal acid generator additives, as well as dispense and casting process conditions, the polymers were able to fill the via holes in 20 to 80% with good fill profile. Further, these polymers were incorporated with chromophores, which are highly absorptive at 365 nm and 248 nm wavelength. Similar to the bottom antireflective coating, these polymer coatings can effectively reduce or eliminate substrate reflection, swing effect and other problems caused by thin film interference. Our progress in this study has led us to the development of AZ<SUP>R</SUP> EXP HERB<SUP>TM</SUP> B.A.R.C. for 365 nm exposure and the commercialization of AZ<SUP>R</SUP> EXP KrF 17B 80 B.A.R.C. for 248 nm exposure. This paper will focus on development and process modification of these novel materials.
Mitsubishi Electric Corporation (MELCO) has developed an advanced microlithographic process for producing 0.1 micrometer contact holes (CH). A chemical shrink technology, RELACS<SUP>TM</SUP> (Resolution Enhancement Lithography Assisted by Chemical Shrink), utilizes the crosslinking reaction catalyzed by the acid component existing in a predefined resist pattern. This 'RELACS<SUP>TM</SUP>' process is a hole shrinking procedure that includes simple coating, baking, and rinse steps applied after conventional photolithography. This paper examines the process parameters affecting shrinkage of CH size. We subsequently evaluated the dependency of CH shrinkage on resist formulation. We conducted investigations of shrink magnitude dependency on each process parameter. (1) Photoresist lithography process: CH size, exposure dose, post development bake temperature. (2) AZ<SUP>R</SUP> R200 [a product of Clariant (Japan) K.K.] RELACS<SUP>TM</SUP> process: Soft bake temperature, film thickness, mixing bake temperature (diffusion bake temperature), etc. We found that the mixing bake condition (diffusion bake temperature) is one of most critical parameters to affect the amount of CH shrink. Additionally, the structural influence of photoacid generators on shrinkage performance was also investigated in both high and low activation energy resist systems. The shrinkage behavior by the photoacid generator of the resist is considered in terms of the structure (molecular volume) of the photogenerated acid and its acidity (pKa). The results of these studies are discussed in terms of base polymer influence on shrinkage performance and tendency. Process impact of the structure and acidity of the photogenerated acid is explored. Though the experimental acetal type KrF positive resist (low activation energy system) can achieve around 0.1 micrometer CH after RELACS<SUP>TM</SUP> processing under the optimized condition, the experimental acrylate type positive resist (high activation energy system) showed less shrinkage under the same process condition. The shrinkage performance of RELACS<SUP>TM</SUP> process largely depends on the resist chemistry used as the underlying layer. Further, shrinkage degree can be controlled by process optimization even for the high activation energy type photoresist.
This paper will evaluate the potential improvements with the addition of an aqueous based top antireflective coating to a Deep-UV process. This antireflective coating (ARC) is resistant to intermixing with the resists and is removed during the normal develop operation. A logic IC with 0.15- 0.18 micrometers design rules will be the primary test vehicle, concentrating on Contact/VIA levels. Performance will be compare with and without the top ARC. Uniformity of measured critical dimensions (CD) will be compared. CD/dose swing curve suppression, as a function of substrate and resist thickness, will be documented. The Deep-UV resist/top ARC application will be optimized, as well, to maximize throughput.
The use of bottom antireflective coatings (BARCs) as a means for controlling substrate reflectivity and thin film effects, has become commonplace in today's wafer fabs. In an effort to simplify process integration, reduce environmental impact, and reduce processing costs, some next generation organic BARC materials have recently been introduced which are formulated with photoresist compatible solvent systems. This study examines the process effects of converting from the cyclohexanone based AZ<SUP>TM</SUP> BARLi<SUP>TM</SUP> anti-reflective coating, to the recently introduced PGME/Ethyl Lactate based AZ<SUP>TM</SUP> BARLi<SUP>TM</SUP> II anti-reflective coating. We will present a comparison of the optical properties of the two products, and examine i-line lithographic process effects including process latitudes, CD distributions, and coat defects, as well as post etch CD distributions, and dye sublimation during cure.
Process improvements attributed to the use of bottom anti- reflective coatings (B.A.R.C.s) are well documented. As our experience with these materials improves, so does our understanding of additional optimization. Recent supplier experiments suggest an increase in the thickness of AZ<SUP>R</SUP> BARLi<SUP>TM</SUP> (bottom anti-reflective layer i-line) solution to reduce photoresist swing curve ratios. Also, changes in thin film stack on common substrates can adversely affect the degree of photoresist reflective notching. It is therefore of extreme importance to determine optimum thickness(es) of a B.A.R.C. material to ensure maximum process potential. We document several process effects in the conversion of a SRAM test device (0.38 - 0.45 micrometers) from a 650 angstrom to a 2000 angstrom BARLi<SUP>TM</SUP> film thickness using conventional i-line photolithography. Critical dimension (CD) uniformity and depth of focus (DOF) are evaluated. Defect density between the two processes are compared before and after etch employing optical metrology and electrical test structures. Sensitivity of overlay as a function of BARLi<SUP>TM</SUP> film thickness is investigated as well.
In optical projection lithography of all types, optimum performance depends on the design and precise alignment of the source(s) and optical components that illuminate photomasks, as well as those for the projection lens. In this paper, we illustrate the effects of abnormalities in the illumination system; these abnormalities include asymmetric nonuniformity of the light source, obscurations, aberrations of the illumination optics, and telecentric error. 'Complex' illumination describes cases wherein all or part of the field of a stepper or scanning tool is illuminated asymmetrically. The interaction of complex illumination at the photomask with defocus or aberrations generates interference effects in the same manner as phase shifting or off-axis illumination, thereby modulating the image and, in many cases, shifting the image from its intended location. We calculate, from scalar coherence theory, quantitative influences on overlay for 0.35 micrometers lithography, and we determine selected tolerances for source uniformity and symmetry as a function of wavelength and coherence parameter. The effects of complex illumination are object-dependent, and we describe the variation with mask polarity, feature size, and proximity. We will consider the use of phase masks, the use of a scanned source and projection lens, and the use of off-axis illumination as special cases and describe their interaction with complex illumination in lithography. With the use of simulation software for lithography, we demonstrate the effects of complex illumination within photoresist patterns. We show that, for expected performance of illumination in a well-characterized step and repeat or scanning tool, the effects of complex illumination are seen to be small in comparison to expected alignment tolerances. For selected cases, we demonstrate that abnormalities arising from obstructed or incorrectly positioned components cause significant errors.
A process using a bottom-side antireflective coating, AZ BARLi, has been studied for 0.50 micrometers and sub-0.5 micrometers features using I-line photolithography. Significant improvements were demonstrated for such process parameters as CD swing curve ratio, exposure latitude, and reflective notching of the photoresist. Extensive characterization was done on defects observed between the BARLi and photoresist coatings, and a process developed for their elimination. Factors which had significant effects on the observed number of defects, and their distribution, were the type of photoresist coat program used, solvent treatment of the BARLi surface, and a high temperature bake after photoresist coat. Data is presented for a complete process, which includes plasma etching the BARLi antireflective coating.