The use of phosphoric acid in dentin bonding presents several disadvantages, even though dentin etching is an essential step in adhesive procedures performed with etch-and-rinse systems [1]. This type of bonding agent requires the etched dentin to be kept moist before applying the adhesive resin [2]. Therefore, the collapse of the exposed collagen fibrils, which can hinder the penetration of monomers and the formation of the hybrid layer [3], is avoided.

However, the contact of water with adhesive monomers causes phase alterations in their molecules [4], making it difficult for the adhesive resin to diffuse into the interfibrillar spaces [5]. As a result, a discrepancy between the depth of etching and the depth of adhesive infiltration can be observed, leaving collagen fibrils exposed and unprotected at the bottom of the hybrid layer [6]. These fibrils become susceptible to degradation by endogenous collagenolytic enzymes (dentin matrix metalloproteinases - MMPs - and cysteine cathepsins) [7]. Furthermore, phosphoric acid was shown to cause changes in the molecular arrangement of collagen that favor the action of dentin MMPs [8], while also activating such enzymes [9].

The combination of these factors herein described leads to the hydrolytic degradation of collagen fibrils within the hybrid layer over time, which can decrease the bond strength between dentin and adhesive [10]. Thus, milder dentin conditioners that could replace phosphoric acid have been investigated [11], including citric, maleic, oxalic, ethylenediaminetetraacetic acids, and, more recently, metal salt-based zirconium oxynitrate [53], [54]. One of these studies suggested the use of an acid solution composed of 6.8% ferric oxalate hexahydrate (Fe2(C2O4) • 6 H2O) as a “cleaning/fixation” agent to be applied on dentin before the adhesive resin [12]. According to the authors, this solution could dissolve the smear layer and interact with the organic components of dentin, promoting condensation and crosslinking, while also decreasing the solubility of the etched substrate [12].

In the presence of iron-containing compounds, dentin polyelectrolytes bond to Fe+3 ions, becoming insoluble and therefore allowing the maintenance of interfibrillar spaces even in the absence of water [13]. In addition, Fe+3 is also capable of inhibiting the activity of MMP-2, MMP-9, and cathepsin-K [14], [15]. Hence, the use of a dentin conditioner containing Fe+3 ions could allow effective bonding on dry etched dentin, which would simplify adhesive procedures that follow the etch-and-rinse technique. Besides, the inhibitory effect of iron on the proteolytic activity of dentin could also result in long-lasting dentin bonding.

Despite the abovementioned advantages of iron-containing etchants, Blosser & Bowen pointed out that the removal of the smear layer by these agents was probably due to the action of varying concentrations (1–10%) of nitric acid in the composition of ferric oxalate rather than the action of the compound itself [16]. Thus, different types of acids can be incorporated into the ferric oxalate solution so that it can remove the smear layer and promote adequate bonding between dentin and adhesive. Taking this observation into account, citric acid is a potential candidate to enhance the effects of ferric oxalate-based etchants because of its ability to substantially increase dentin permeability [13]. Another drawback of etchants containing ferric oxalate is the yellowish color of the compound, which can cause staining and discoloration of dentin upon application [17], limiting its use in regard to esthetic concerns. In order to solve this problem, Fe+3 was replaced by Al+3 in the 1990 s [18]. Similarly to iron ions, Al+3 can inhibit elastin degradation by MMP-2 and MMP-9 [19].

Considering all the problems and limitations arising from the current approach to dentin bonding with phosphoric acid, it became necessary to study milder etching agents that could enable dry-bonding techniques and potentially inhibit dentin proteolytic activity. Thus, this study aimed to evaluate the microtensile dentin bond strength (at 24 h and after one year of artificial aging), the bonding interface and surface morphology (at 24 h), the in situ enzymatic activity of the hybrid layer (at 24 h), and the antibiofilm activity of alternative dentin etchants. The null hypotheses were: (1) there would be no significant differences in dentin bond strength among the evaluated groups, regardless of etching strategy or aging; (2) the percentages of occurrence of the studied failure modes would not differ among groups at 24 h or one year; (3) the bonding interface morphology would not differ among the tested etchants; (4) no differences among dentin etchants regarding enzymatic activity would be observed; and (5) no differences among dentin etchants regarding antibacterial activity against S. Mutans biofilm would be observed.