The purpose of this study is to evaluate the hypothesis that replacing 2-hydroxyethyl methacrylate (HEMA) for surfactant dimethacrylates (SD) does not affect the immediate and long-term microtensile bond strength (mu TBS) of experimental two-step self-etch HEMA-free adhesive systems applied on dentin. Five experimental HEMA-free two-step self-etching systems containing different SD (ethoxylated bisphenol A diglycidyl dimethacrylate (Bis-EMA 10, B10), Bis-EMA 30 (B30), poly-ethyleneglycol (400) dimethacrylate (PEG 400, P400), PEG 1000 (P1000), and PEG 400 urethane dimethacrylate (UDMA) (UP400)) and a HEMA-containing system (control) (HA) were formulated. Specimens were subjected to the mu TBS test after 24 h and 6 and 12 months of storage. Data (in megapascals) were analyzed by Kruskal-Wallis and Dunn tests (alpha = 0.05). Medians of the mu TBS data after 24 h of storage are: HA = 57.2(A), B10 = 26.2(BC), B30 = 24.0(C), P400 = 32.6(BC), P1000 = 37.3(B), and UP400 = 57.9(A); after 6 months are: HA = 47.9(A), B10 = 18.5(B), B30 = 7.8(C), P400 = 16.1(B), P1000 = 14.6(BC), and UP400 = 51.6(A); and after 12 months are: HA = 31.2(A), B10 = 15.2(B), B30 = 9.0(B), P400 = 9.1(B), P1000 = 13.3(B), and UP400 = 35.7(A). Between the HEMA-free groups, the adhesive system formulated with PEG 400 UDMA produced similar mu TBS to the HEMA-containing group. Also, the storage of specimens decreased the mu TBS (p < 0.05). Replacing HEMA for PEG 400 UDMA in an adhesive system formulation generated a satisfactory mu TBS to dentin. Surfactant dimethacrylates have a potential use in the development of HEMA-free self-etching adhesive systems, which are more chemically stable.