This year the BACUS (Bay Area Chrome Users Society) meeting also known as Photomask Technology, was combined with the EUV Lithography Conference, and was held at Monterey, California in September this year. One area pertinent to both EUV and photomasks is pellicles. Pellicles protect masks from encountering particles. Particles in contact with the mask block the exposure of covered features, which can include many lines or contacts/vias. Actually, even a small (<100 nm) particle landing within a contact/via feature on the mask is bad enough. At this conference, a number of materials were evaluated as potential new pellicle materials.
PTB and IMEC (paper 104510R) discussed nanostructured SiN membranes and CNT-based membranes. The holes in the membranes allow reduced absorption. However, they also cause light scattering, which are an extra transmission loss mechanism (besides absorption).
ASML (paper 104510O) presented work with Philips and Univ. of Twente on evaluating its current material choice, polysilicon (pSi), as well as a couple of alternatives, metal silicides and graphite. (Graphite is popular at Samsung, by the way.)
Pellicles need to satisfy a diverse set of requirements. The first requirement is 90% transmission (one-way). This requires pSi thickness of 30 nm or less. With such thin films, the deflection or sag needs to be carefully evaluated. 50 nm pSi thickness (current value) could lead to 100 um deflection over 100 mm x 100 mm (Mechanical deflection of a free-standing pellicle for extreme ultraviolet lithography - ScienceDirect) Current evaluation sizes were 1 cm x 1 cm, which doesn't represent a real-use load yet. Besides mechanical stability, thermal and chemical (hydrogen exposure) stability are necessary as well. Thinner films dissipate heat by conduction less effectively.
Metal-boride caps were proposed as transmissive and emissive protection for the pellicle, but at 530 C there is some loss of Boron detected. A temperature of 400-500 C is expected with a 250 W source.
Graphite films at ASML were more fragile (easily perforated) compared to pSi or metal silicides, and are also particularly vulnerable to hydrogen (as are CNTs). Metal silicides have lower transmission than pSi so need to be thinner but their thermal expansion is more severe as well.
Therefore, to date, no full successful demonstration of an EUV pellicle has been achieved yet.
PTB and IMEC (paper 104510R) discussed nanostructured SiN membranes and CNT-based membranes. The holes in the membranes allow reduced absorption. However, they also cause light scattering, which are an extra transmission loss mechanism (besides absorption).
ASML (paper 104510O) presented work with Philips and Univ. of Twente on evaluating its current material choice, polysilicon (pSi), as well as a couple of alternatives, metal silicides and graphite. (Graphite is popular at Samsung, by the way.)
Pellicles need to satisfy a diverse set of requirements. The first requirement is 90% transmission (one-way). This requires pSi thickness of 30 nm or less. With such thin films, the deflection or sag needs to be carefully evaluated. 50 nm pSi thickness (current value) could lead to 100 um deflection over 100 mm x 100 mm (Mechanical deflection of a free-standing pellicle for extreme ultraviolet lithography - ScienceDirect) Current evaluation sizes were 1 cm x 1 cm, which doesn't represent a real-use load yet. Besides mechanical stability, thermal and chemical (hydrogen exposure) stability are necessary as well. Thinner films dissipate heat by conduction less effectively.
Metal-boride caps were proposed as transmissive and emissive protection for the pellicle, but at 530 C there is some loss of Boron detected. A temperature of 400-500 C is expected with a 250 W source.
Graphite films at ASML were more fragile (easily perforated) compared to pSi or metal silicides, and are also particularly vulnerable to hydrogen (as are CNTs). Metal silicides have lower transmission than pSi so need to be thinner but their thermal expansion is more severe as well.
Therefore, to date, no full successful demonstration of an EUV pellicle has been achieved yet.
Last edited:
