Objectives. This study was conducted to investigate the depths of poly
merization inhibition by oxygen, the shear bond strengths to enamel an
d dentin, and the marginal adaptation in dentin cavities of experiment
al adhesives containing BisGMA/HEMA or mixtures of low-inhibition BisG
MA-dicarbonate with HEMA or with HEMA-carbonate at ratios of 100/0, 80
/20, 60/40, 50/50, and 40/60 by weight. Methods. The inhibition layer
thickness was determined microscopically as non-polymerized surface fi
lm thickness on three samples each. For bond strength testing, composi
te cylinders were bonded to enamel and dentin following H3PO4 conditio
ning with the adhesives dissolved in acetone; marginal adaptation was
evaluated in cylindrical dentin cavities. Five specimens each were sto
red in water for 24 h prior to testing, Shear bond strength results we
re compared by ANOVA and Duncan's test (p < 0.05), the maximum gap wid
ths by Kruskal-Wallis, ANOVA and Wilcoxon's two-sample test (p < 0.01)
. Results. inhibition layer thickness was significantly smaller for mo
dified than for conventional BisGMA/HEMA monomers and increased signif
icantly with HEMA content. In the BisGMA-dicarbonate/HEMA-carbonate gr
oup, the inhibited layer was thinner than 1 mu m. Apart from the 40 an
d 50% BisGMA-dicarbonate/HEMA-carbonate mixtures, enamel bond strength
s were not significantly different. The highest bond strengths to dent
in were 15 and 12 MPa at 50 BisGMA and 50 BisGMA-dicarbonate/HEMA, res
pectively. The average bond strength far the BisGMA-dicarbonate/HEMA-c
arbonate group was 5 MPa. Adhesives with 40 and 60% HEMA in the conven
tional and 40 to 60% HEMA in the dicarbonate/HEMA group showed signifi
cantly better marginal adaptation than all others (p < 0.05). Signific
ance. Carbonate-modified low-inhibition monomers have no advantage as
enamel/dentin adhesives compared with conventional BisGMA/HEMA-based r
esins.