Rodney J. Hurditch

Selectively DNQ-esterified PAC for high-performance positive photoresists

DNQ-PACs with varying number of OH groups unesterified were examined for their imaging performance in novolac-based positive resists by means of a dissolution rate study. PACs generally lost dissolution inhibition with increasing numbers of the unesterified OH groups when compared to fully esterified PACs, whereas certain particular PACs still retained strong inhibition even when an OH was left unesterified. Such PACs lost inhibition quickly when one or more OH was left unesterified and gave a large dissolution discrimination upon exposure that resulted in high resolution of the resist. High hydrophobicity and remote DNQ configuration of the PAC molecule, and probably the steric crowding around the OH group appear to be structural requirements for obtaining such a high performance PAC. The PACs also provided an advantage in good solubility to resist solvent, and can be practically made by a proposed selective DNQ esterification.

Studies of dissolution inhibition mechanism of DNQ-novolak resist (II): effect of extended ortho-ortho bond in novolak

A p-cresol trimer sequence was incorporated into a polymeric chain of novolak by copolymerization with m-cresol of a reactive precursor which was prepared by attaching two units of m-cresol to the terminal ortho positions of p-cresol trimer. The resulting novolak was characterized by 13C NMR and FTIR in an attempt to correlate novolak structure with dissolution inhibition function based on physicochemical analysis of molecular interactions between novolak and DNQ-PAC in solid films.

Experimental investigation of a novel dissolution model

Mack recently proposed a new kinetically based 5-parameter model for positive photoresist bulk dissolution. The present work tests its physical assumptions by comparing its predictions with DRM measurements of seven different PAC/novolak formulations, each at several different PAC loadings. Although substantial qualitative agreement between the predictions and experiments is seen, quantitative agreement is poor. The most likely explanations for the quantitative difficulties are: (1) the assumptions that inhibition by the PAC and enhancement by the photo-acid can be treated independently and multiplicatively, apparently fail, (2) side reactions and intermediates are neglected by the model, and (3) complex PAC isomeric distributions and associated complex inhibition/enhancement effects are also neglected. The dissolution rate equation derived by the model does exhibit excellent flexibility in fitting actual bulk dissolution rate curves. This reason alone is sufficient to recommend its inclusion in commonly used simulation programs such as PROLITH/2 and SAMPLE. the present work shows that significant errors can result in simulations by using simple dissolution rate equation which are unable to accurately describe observed bulk dissolution data.

Physical and chemical factors governing dissolution of novolak resin films

The authors have investigated the dissolution of a series of novolac resins with different structures and properties with various quaternary ammonium hydroxides using a new technique called Spectroscopic Dissolution Rate Monitoring (SDRM). This technique allows direct spectral measurement of an absorption band attributed to cation complex formation. The rate of formation of this complex is used as an approximation or surrogate for cation diffusion. The rate of formation of the complex is approximately the same as the rate of dissolution when using a rinse. This evidence supports cation diffusion as the rate determining step even in TMAH development of high ortho, ortho bonded systems. Our studies also suggest that polymer flexibility and microstructure exert a strong influence on the cation diffusion rate.

Effect of PAC structure and resist morphology on the control of surface inhibition in positive photoresist systems

The influence of PAC structure and morphology on the surface induction in novolak based resist systems was investigated. Variation in surface induction for a given PAC is caused by changes in the morphology between the surface and the bulk of the resist film resulting from solvent depletion and is characteristic of the novolak. PAC structure, particularly its functionality [i.e., maximum amount of diazonaphthoqunione (DNQ) esterifiable OH], was found to greatly influence the amount of surface inhibition.

Interaction of novolak oligomers with hydrogen bond acceptors

Hydrogen bonding between several novolak oligomers as donors and phenylsulfoxide as acceptor was investigated as a model for the interaction of inhibitors with o,o-connected phenolic blocks in practical novolak resins. It was found that the attraction of the novolak dimer for the acceptor was more than twice as strong as that of the corresponding monomeric phenol. In a non-polar solvent, carbon tetrachloride, the higher oligomers assume cyclic conformations where all hydrogen bonds are internally saturated. There are, however, indications, that in sufficiently polar solvents these cyclic structures open up and that the acyclic oligomers then interact strongly with hydrogen bond acceptors.

I-line photolithography: an investigation of resist bleachability and process performance

The present work presents an investigation of the characteristics of a broad band capability photoresist in terms of photo-active compound (PAC) bleachability and process performances. The difference in performance on exposure to 436 and 365 nm wavelengths of a commercially available photoresist, Olin Hunt H1PR-6512, has been evaluated as a function of some typical lithographic parameters by using FT-IR spectroscopy and actinometry. This analysis contributed to the development of a submicron I-line process, which in conclusion is shown in a production environment application.