DNQ–novolac photoresists revisited: ¹H and ¹³C NMR evidence for a novel photoreaction mechanism

Abstract

Photoactive compounds, such as diazonaphthoquinone (DNQ) esters, blended with novolac resins, solvents and certain additives, serve as photoresists. These are used for printing of electronic circuits at the micron or sub-micron level. Patterns are generated based on changes in the physical and chemical properties of the exposed and unexposed photoresist surfaces (printed circuit boards). The huge polarity change between the exposed and unexposed photoresists is exploited in the technique of microlithography. It is believed that the large polarity difference is due to acid formation in the exposed photoresist by a photochemical reaction of DNQ on exposure to light. However, it has also been suggested that in the unexposed part of a photoresist, the novolac resin undergoes an azo coupling reaction with DNQ, leading to an increase in the molecular weight of the resin, rendering it more insoluble in base. The protons in the para positions of the m-cresol units incorporated in the novolac resin are believed to take part in this azo coupling reaction with DNQ. In this paper, we propose a novel mechanism of action of positive photoresists in the unexposed part of photoresists for dissolution inhibition using molecular modelling, 1 H NMR, 13 C NMR and DEPT-135 NMR spectroscopic techniques. Our results enable us to propose that the diazo group of DNQ attacks the methylene bridges rather than the aromatic moiety of the resin. This mechanism explains the pattern formation observed using even p-cresol-based resins, where no free para positions are present in the aromatic ring.

Photo‐induced cross‐linking mechanism in azide–novolac negative photoresists: molecular level investigation using NMR spectroscopy

“…The signals at υ 6.8-7.05 ppm have been assigned by us earlier for the para hydrogens of m-cresol incorporated in the novolac resin. 4,20 The absence of any change in this region in the exposed part of the negative photoresist indicates the absence of any cross-linking reaction with the aromatic moiety. The intensities of the hydroxyl groups diminished drastically in the 1 H NMR spectrum of the exposed photoresist compared with its unexposed counterpart.…”

“…17,18 The other novolac oligomer used was also made by a two-step procedure. 4,16 The negative photoresists were made by mixing one part of the novolac oligomer (M w D 574, M n D 360 and I p D 1.56) and one part of 1-azido-2,5-dichloro-3,6-dimethyl-4-nitrobenzene. Benzene was used as the solvent for the photoresist.…”

“…We have investigated this at the molecular level using NMR spectroscopy. As mentioned in our earlier paper, 4 ethyl cellosolve acetate has been replaced by benzene as solvent for the photoresist to eliminate any interference due to the solvent in the crucial NMR chemical shift region. The ratio of azide to novolac resin, which in photoresist formulations is generally 1 : 5, has been increased to 1 : 1 for enhancement of the signal intensity arising out of the interaction of azide and the novolac resin.…”

“…In an earlier paper, we proposed a novel photochemical mechanism in the unexposed part of the positive photoresist during development, which bring about a huge polarity change between the exposed and unexposed parts of a photoresist. 4 In this paper, our focus is on the mechanism of action of a negative photoresist.…”

Quantitative estimation of the degree of derivatization: an alternative methodology using 1‐D and 2‐D NMR experiments

“…15,16 A 10.8 g (0.1 mol) amount of m-cresol, 16.2 g (0.2 mol) of 37% formaldehyde and 4.0 g (0.1 mol) of sodium hydroxide were stirred mechanically at room temperature (25°C) for 48 h. Then 10.8 g (0.1 mol) of m-cresol were added and then diluted with hydrochloric acid (1 : 1) in an ice-bath and the pH brought down to 4-4.5. The bishydroxymethylated cresol (BHMC) was separated and the organic layer was washed twice thoroughly with water containing 1% oxalic acid.…”