This lecture from a "Chemical Processes for Micro and Nano Fabrication" course provides a foundational explanation of the role of mask illumination in optical lithography. The discussion progresses from simple, normally incident light to the more complex and powerful techniques used in modern semiconductor manufacturing. The core concept presented is that by manipulating the angle of illumination—a technique known as oblique or off-axis illumination—it is possible to shift the mask's diffraction pattern relative to the objective lens. This shift enables "two-beam imaging," a method that can theoretically double the resolving power of a lithography system compared to traditional "three-beam imaging," pushing the resolution limit from 0.5 λ/NA down to 0.25 λ/NA. The lecture defines the practical implementation of these techniques through the concepts of spatial coherence (coherent, incoherent, and partially coherent light) and introduces the partial coherence factor (sigma, σ) as the key parameter for controlling the range of illumination angles. Finally, it describes advanced off-axis illumination (OAI) shapes like annular, quadripole, and dipole, which are engineered to optimize resolution for specific mask patterns.
Shaping Light to Print Smaller Features: A Guide to Lithographic Illumination
0:46 The Effect of Tilting Light: The lecture begins by explaining that tilting the illumination angle (oblique illumination) doesn't change the diffraction pattern itself, but rather shifts its position relative to the objective lens. This shift is critical because it determines which diffracted orders are captured to form the final image.
5:30 Defining and Improving Resolution: Resolution is defined as the smallest feature that can be printed. To form any image at all, at least two diffracted orders (e.g., the 0 and +1 orders) must be captured by the lens. With standard, non-tilted illumination, this typically requires capturing three orders (0, +1, and -1), which limits the theoretical minimum half-pitch to 0.5 * λ/NA (where K1=0.5).
9:11 Doubling Resolution with Two-Beam Imaging: A key takeaway is that by precisely tilting the illumination, it's possible to capture only two diffracted orders (e.g., the 0 and -1 orders). This technique, known as "two-beam imaging," dramatically improves the theoretical resolution limit, cutting the minimum half-pitch in half to 0.25 * λ/NA (K1=0.25).
12:47 Creating Illumination in Practice: The lecture explains how different illumination types are generated. A single point source creates coherent light from one direction. Moving that point off the central axis creates tilted (oblique) illumination. In practice, an extended light source is used, which acts as a collection of many point sources, creating a range of illumination angles.
16:56 Understanding Spatial Coherence: The range of illumination angles is formally described by spatial coherence.
Coherent: Light arriving from a single direction (like from an ideal point source).
Partially Coherent: Light arriving from a finite, controlled range of angles. This is what is used in real lithography tools.
Incoherent: Light arriving from all possible angles (like from an infinitely large source).
25:07 Quantifying Illumination with Sigma (σ): The degree of partial coherence is quantified by the partial coherence factor, or sigma (σ). It is defined as the ratio of the illumination's angular range to the objective lens's numerical aperture (NA). A sigma of 0 represents perfectly coherent light, while typical lithography processes use sigma values between 0.3 and 0.8.
29:19 Advanced Off-Axis Illumination (OAI): To force the two-beam imaging needed for high resolution, lithographers use specialized, non-circular source shapes. These include annular (donut-shaped), quadripole (four spots), and dipole (two spots) sources. This technique, called Off-Axis Illumination (OAI), is designed to enhance resolution for specific mask patterns by eliminating normally incident light.
31:36 Calculating the Final Image: For partially coherent illumination, the final image is the sum of the intensities of images produced by every individual point within the source. This complex calculation is performed by specialized lithography simulation software.