Low K1 Factor Research Articles

Imaging optics on scanner for SMO generation process

Abstract The k1 factor continues to be driven downward, in order to enable the 22-nm feature generation and beyond. Such a low k1 factor tends to lead to extremely small process windows. For such demanding imaging challenges, it is not only necessary for each unit, contributing to the imaging system, to be driven to its ultimate performance capability, but also that active techniques that can expand the process window and the robustness of the imaging against various kinds of imaging parameters be implemented. One such technique is source and mask optimization (SMO). In this paper, we study the effect of SMO. Furthermore, we discuss how to realize the SMO solution in the imaging system setup on the scanner. The setup process includes freeform pupilgram generation (source intensity distribution on the pupil) and pupilgram adjustment.

DOF and Coherence Optimization in Sub-32 nm Contact Lithography

As lithography process steps into deep low k1 factor (~0.3 or smaller) generation, imaging quality and lithography process window (PW) are highly sensitive to the illumination pupil. Maintaining the depth of focus (DOF) becomes more challenging for sub-32 nm tech node micro-lithographic process especially for contact patterns, and carefully selection of accurate illumination setting is significant during the course of process development. Although Source Mask Optimization (SMO) has become one of the most important techniques available for extending ArF immersion lithography, for logic and foundry patterns, source optimization only is still an effective method to extend lithographic capability and maximize depth of focus. In this paper, we present our study on the determination of the optimized illumination mode for 28 nm contact lithography through both photolithographic simulation and experiments with different imaging coherence settings. This simulation and experimental study show that even a previously believed "tiny" sigma change, such as, 0.01 can lead to a significant effect on depth of focus, which can be as much as 30%. This indicates that photolithographic process optimization and development have entered an era where intensive simulation and parameter mapping becomes inevitable in order to achieve success. Also, the heavy dependence of an acceptable process on the illumination has put a more stringent requirement on the scanner's control on the illumination pupil's accuracy and stability.

Direct, in-scanner, aerial image sensing

With feature sizes now less than one half of the wavelength of the actinic radiation, determining the intensity distribution in the aerial image has become both more critical because of the low k1 factors used and more challenging because of the limitations of scalar diffraction theory. Here we describe ways of measuring in situ the intensity distribution generated at the wafer surface by the wafer exposure tool (‘scanner’ or ‘stepper’) and that eliminate the delays and vagaries of resist processing. Some early results show a 2D image generated by an aerial image sensor loaded into an unmodified 193 nm scanner. Nominal feature sizes are down to 67 nm. One additional advantage of this technique over the slower alternative of SEM viewing of the resist image is the gray-scale of the intensity image from the sensor. Applications can include assurance of the correctness of the stepper-mask combination, verification and development of the modeling of the aerial image (including the effect of resolution enhancing techniques) and the qualification of incoming masks.

CURRENT ISSUES AND FUTURE PROSPECTS OF LITHOGRAPHY

The resolution limit of optical lithography now looms on the horizon. We are using very complicated masks and sophisticated exposure tools with very low k1 factors. To deal with this situation, liquid immersion lithography is now under intensive development. It can extend the application of the optical lithography for one or two generation, but the k1 factor remains very small. To obtain a significantly larger k1 value, we shall move to the non-optical lithography techniques. EUV lithography is the strongest candidate. Many problems of EUV still remain though we have remarkable achievements recently. In addition to the mass production technology, methods such as maskless lithography (ML2) and nano-imprint lithography were also developed for small-scale production and some other applications.