Photoresist contrast ($\gamma$)

Photoresist contrast measures how sharply a photoresist changes its developed thickness as the exposure dose changes. For a positive photoresist, let $h(D)$ be the remaining thickness after a specified development process and $h_0$ the initial thickness. A contrast curve plots $h(D)/h_0$ against ${\log }_{10}D$. The measured photoresist contrast is the magnitude of the slope near clearing; it need not describe the entire curve.^[ref]

In an idealized model, the thickness decreases linearly with ${\log }_{10}D$ between a retention exposure dose $D_0$ and a clearing exposure dose $D_c$, where $0<D_0<D_c$. ThenOutside that interval the idealized thickness is clipped to $1$ or $0$. A large $\gamma$ means a smaller exposure dose ratio suffices to change from retention to clearing. This is why photoresist contrast influences the critical modulation for a photoresist; the inverse-logarithm formula assumes this particular contrast-curve model.

The critical modulation for a photoresist is the smallest aerial-image modulation that permits a common exposure time to clear bright regions while retaining full thickness in dark regions. Consider a positive photoresist with retention exposure dose $D_0>0$ and clearing exposure dose $D_c>D_0$. Write the extreme image intensities asFor exposure time $t$, the requirements are $t{I}_{\max }\ge D_c$ and $t{I}_{\min }\le D_0$. A feasible $t$ exists exactly whenwith $M=1$ handled directly since the dark-region exposure dose is zero. Rearrangement givesThis derivation assumes exposure dose is intensity times exposure time and uses fixed retention and clearing thresholds. In the idealized linear contrast-curve model of photoresist contrast, $D_c/D_0=10^{1/\gamma }$, soFor example, $D_c=4D_0$ gives $M_c=3/5$. Greater photoresist contrast lowers the required aerial-image modulation in this model.^[ref]