The Inﬂuence of Application Protocol of Simpliﬁed and Universal Adhesives on the Dentin Bonding Performance

: Contemporary adhesives use etch-and-rinse, self-etch, and multimode adhesive strategies. Simpliﬁed and universal adhesives present lower bond strength to dentin than conventional, two-bottle etch-and-rinse adhesives. The purpose of this study was to evaluate bonding performance of simpliﬁed and universal adhesives to dentin after modifying their application protocol (multiplying applications and extending application time). Adhesive layer thickness (ALT) and shear bond strength (SBS), as well as the correlation between these parameters was calculated. Two universal (Single Bond Universal and Prime & Bond One Select) and two self-etch adhesives (Adper Easy One and Xeno V) were tested. Signiﬁcant di ﬀ erences in ALT were detected between the adhesives, as well as within the same adhesive between study groups. Tested adhesives presented the thinnest adhesive layer when applied 2 times in 20 s. Single Bond Universal obtained the highest SBS results of all adhesives. Most adhesives (except for Prime & Bond One Select) obtained the highest SBS, when applied two or three times in 40 or 60 s, respectively. No correlation between the ALT and SBS was found. The study showed that increasing the number of applications and extending the application time of self-etch and universal adhesives can be recommended to improve their performance.


Introduction
The origin of contemporary restorative dentistry dates back to 1955, when Buonocore proposed the acid etching of enamel as a method to increase the adhesion of acrylic restorations to dental tissues [1]. Since then, eight different generations of adhesives, which constitute the intermediate layer between dental fillings and enamel or dentin, have been introduced. The first three generations appeared to provide insufficient bond strength and were withdrawn from clinical use. The fourth and fifth generation is based on the total-etch technique, which consists of phosphoric acid etching of both enamel and dentin in order to create porosities, which increase the adhesion surface of dental tissues.
Micromechanical adhesion is proved to be the prime mechanism in bonding of resin materials to dental tissues [2]. It is achieved in a process, where resin monomers replace inorganic dental component, which leads to the monomers being interlocked in dental porosities, and polymerized [3]. When According to recent studies, the application of SE and MM adhesives in SEE mode is the most effective protocol to ensure a durable bond to dental tissues [3,23]. When bonding to etched enamel, these adhesives obtain satisfactory results [24]. Yet, when bonding to dentin, alterations of the application procedure were suggested in order to further improve the bond strength of both SE and MM adhesives [25]. These alterations included extending the application time [26] and multiplying the number of consecutive applications (no light-curing in between) [27] or layers (each layer light-cured) [27].
The aim of this study was to determine the effect of multiple applications and extended application time of self-etch and universal adhesives on dentin bond strength and on thickness of adhesive layer as well as the correlation between these parameters. The null hypotheses tested are: (1) There is no effect of multiple applications and extended application time of self-etch and universal adhesives on dentin bond strength. (2) There is no effect of multiple applications and extended application time of self-etch and universal adhesives on thickness of adhesive layer. (3) There is no correlation between adhesive layer thickness and dentin bond strength.

Sample Preparation
Intact human molars (n = 140), stored in 0.5% chloramine solution to inhibit microbial growth, were used in this study. The teeth were examined for lack of caries and restorations under 10× magnification optical microscope and used within 2 months of extraction. After being separated from roots, crowns were sectioned mesio-distally in precision micro-cutting machine (Mecatome T210 Prezi, France). Two samples, buccal and lingual, were obtained from each tooth. The enamel of each sample was removed and dentin was exposed and ground (Mecatome T330, Prezi, France) under water lubrication. Silicon carbide papers, 180-grit followed by 600-grit were used to create standardized smear layer on dentin [25]. The dentinal surface was examined for residual enamel under 10x magnification optical microscope. Institutional Ethical Committee approval was obtained for this study (RNN/435/18/KE).
Dentin samples (n = 280) were randomly assigned to 4 groups (n = 70) according to the dental adhesive used. The study used two self-etch and two universal dental adhesives (Table 1.). For each tested adhesive, five study groups (n = 14) were designed. The tested adhesives were applied on the dentin surface according to the research model presented in Table 2. The adhesives were light-cured with LED curing lamp (Elipar™ S10 LED Curing Light, 3M ESPE, Germany). Next, silicone ring with the inner diameter of 3 mm and height of 4 mm was placed on the sample surfaces. Flowable composite (Flow-Art, Arkona, Poland) was applied inside the ring and light-cured for 40 s in 2 equal increments, to ensure complete polymerization. Within each study group, 11 samples were tested for SBS, while 3 samples underwent SEM-EDS analysis of dentin-resin interface. Adhesive was applied according to the manufacturers' instructions: Apply adhesive and rub into the entire dentin surface with disposable microbrush for 20 s. Gently air-dry the surface for 5 s at a distance of 10 cm to form a slightly shiny adhesive film. Light-cure for 20 s.

2A20
Adhesive was applied two times in the abovementioned manner during 20 s and light-cured for 20 s afterwards.

3A20
Adhesive was applied three times in the abovementioned manner during 20 s and light-cured for 20 s afterwards.

2A40
Each of the 2 applications of adhesive lasted 20 s, which made 40 s in total. The adhesive was light-cured for 20 s.

3A60
Each of the 3 applications of adhesive lasted 20 s, which made 60 s in total. The adhesive was light-cured for 20 s.

SEM and EDS Analysis
In order to evaluate the interfacial morphology, for each tested adhesive, 3 dentin samples with adhesives applied according to the research model (Table 2) were prepared as described in 2.1. Using 180, 600 and 1000-grit silicon carbide paper under running water, the resin-dentin interface was exposed and polished. The dental etchant, 37% orthophosphoric acid solution, was applied for 1 min on the samples in order to demineralize dentin, which was later rinsed with distilled water for 1 min. Next, to remove the organic debris, the specimens were treated by soaking in 5% NaOCl for 5 min. The solution was replaced after each minute to prevent it from deactivation. After another distilled water-wash and air-drying, the specimens were dehydrated in ascending ethanol concentration (50%, 70%, 90% and 95% for 20 min each and 100% for 1 h), and then transferred to a critical point dryer for 30 min. All specimens were mounted on aluminium stubs, so that the specimens' surface was parallel to the stubs' surface, and sputter-coated with gold layer. The adhesive/dentin interfaces were then examined using a scanning electron microscope (SEM, FEI Nova NanoSEM 450, FEI, Hillsboro, OR, USA) under 1000× and 2000× magnification, at 90 • tilt angle. In order to measure adhesive layer thickness (ALT), three different points of adhesive interface: the middle part and both extremities of the interface were evaluated [30]. At each point, four measurements were taken using software that enabled vertical positioning of the adhesive layer and marking parallel measuring lines (perpendicular to the adhesive interface). In total, 12 measurements were taken along the interface of each sample based on SEM images at 2000 x magnification. Further, the dentin-resin interface was analyzed from a chemical aspect, using energy dispersive spectrometry (EDS, EDAX/AMETEK, Materials Analysis Division, Model Octane Super, Mahwah, NJ, USA).

Shear Bond Strength
For bond strength testing, the dentin samples (n = 220) were mounted in round PCV molds, using auto polymerizing acrylic resin, with their prepared surfaces exposed to the environment. As soon as the chemical reaction began, the specimens were placed in cold water to prevent the temperature rise during polymerization of acrylic resin that might negatively affect dental substrate. Next, the samples' surfaces were ground under water lubrication with 600-grit silicon carbide paper to achieve a flat surface and standardized smear layer. The samples were randomly assigned to twenty groups (n = 11), depending on the adhesive used and the method of application. The tested adhesives (Table 1) were applied to dentin samples according to the research model presented above (Table 2). Samples were stored in saline for 24 h. Shear bond strength (SBS) was tested with universal testing machine of 20 kN maximum load cell capacity (Zwick-Roell Z005, Zwick-Roell, Germany), running at crosshead speed of 2 mm/min. Blunt knife-edged chisel was used parallel to dentin-composite adhesive interface with less than 1 mm gap distance between the crosshead and substrate. The SBS was calculated by corresponding software program and expressed in MPa.

Statistical Analysis
For each tested group, SBS and ALT values were obtained. Minimal and maximal values, mean values, median, standard deviation and coefficient of variation were calculated. Normal distribution was tested with Shapiro-Wilk test. Due to the fact that the results were not normally distributed, nonparametric tests were employed. Multiple comparisons of the SBS of different bonding systems with control were performed with Mann-Whitney's ANOVA. In order to compare each SBS and ALT group with other respective groups, Kruskal-Wallis ANOVA was employed. As a post hoc, Dunnett's test was used, where p ≤ 0.05 was considered statistically significant. To determine correlation between SBS and ALT, mixed-effects linear regression models with robust standard errors were fitted. All the regression equations were controlled for the modality of application. All the statistical procedures were carried out using STATISTICA 10 (version 10, Publisher: StatSoft, Poland).

SEM and EDS Analysis
SEM images of dentin-resin interface obtained after treatment of dentin with tested adhesives, according to the research model were presented in Figures 1-4.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 5 of 17 temperature rise during polymerization of acrylic resin that might negatively affect dental substrate. Next, the samples' surfaces were ground under water lubrication with 600-grit silicon carbide paper to achieve a flat surface and standardized smear layer. The samples were randomly assigned to twenty groups (n = 11), depending on the adhesive used and the method of application. The tested adhesives (Table 1) were applied to dentin samples according to the research model presented above (Table 2). Samples were stored in saline for 24 h. Shear bond strength (SBS) was tested with universal testing machine of 20 kN maximum load cell capacity (Zwick-Roell Z005, Zwick-Roell, Germany), running at crosshead speed of 2 mm/min. Blunt knife-edged chisel was used parallel to dentincomposite adhesive interface with less than 1 mm gap distance between the crosshead and substrate. The SBS was calculated by corresponding software program and expressed in MPa.

Statistical Analysis
For each tested group, SBS and ALT values were obtained. Minimal and maximal values, mean values, median, standard deviation and coefficient of variation were calculated. Normal distribution was tested with Shapiro-Wilk test. Due to the fact that the results were not normally distributed, nonparametric tests were employed. Multiple comparisons of the SBS of different bonding systems with control were performed with Mann-Whitney's ANOVA. In order to compare each SBS and ALT group with other respective groups, Kruskal-Wallis ANOVA was employed. As a post hoc, Dunnett's test was used, where p ≤ 0.05 was considered statistically significant. To determine correlation between SBS and ALT, mixed-effects linear regression models with robust standard errors were fitted. All the regression equations were controlled for the modality of application. All the statistical procedures were carried out using STATISTICA 10 (version 10, Publisher: StatSoft, Poland).

SEM and EDS Analysis
SEM images of dentin-resin interface obtained after treatment of dentin with tested adhesives, according to the research model were presented in Figures 1     Appl. Sci. 2020, 10, x FOR PEER REVIEW 6 of 17    The mean values of ALT (µm) obtained for the tested adhesives were presented in Table 3. For tested adhesives, except for Xeno V, double application in 20 s resulted in the thinnest adhesive layer among all study groups. The thinnest adhesive layer was obtained for Prime&Bond One Select (6.31 ± 0.91). Also, there was no statistically significant difference in adhesive layer thickness between groups within Prime & Bond One Select. Single Bond Universal, on the other hand, showed the thinnest adhesive layer in 2A20 group and it was significantly lower value than in the control group or in 3A20 group. Adper Easy One reached the lowest mean value in 2A20 group, which was significantly lower than in 3A60 group. The thickest adhesive layer presented Xeno V applied three times in 60 s (14.15 ± 0.67) and it was significantly higher, almost double, when compared to the control group. Within tested adhesives, means followed by the same lowercase letters in row and superscript letters in column indicate significant statistical differences.
The results of EDS analyses of the resin-dentin interface ( Figures 5-8) confirmed that the concentrations of different elements corresponded with specific layers of that interface. In dentin, mainly calcium and phosphorus were detected. The adhesive layer contained carbon, which is the main element present in resin polymer. Silicon was detected in adhesive layers of Adper Easy One and Single Bond Universal, which can be attributed to the presence of nanofiller in the composition of these adhesives. All tested specimens showed increased content of silicon and aluminum, derived from inorganic filler, in composite material. The mean values of ALT (µm) obtained for the tested adhesives were presented in Table 3. For tested adhesives, except for Xeno V, double application in 20 s resulted in the thinnest adhesive layer among all study groups. The thinnest adhesive layer was obtained for Prime&Bond One Select (6.31 ± 0.91). Also, there was no statistically significant difference in adhesive layer thickness between groups within Prime & Bond One Select. Single Bond Universal, on the other hand, showed the thinnest adhesive layer in 2A20 group and it was significantly lower value than in the control group or in 3A20 group. Adper Easy One reached the lowest mean value in 2A20 group, which was significantly lower than in 3A60 group. The thickest adhesive layer presented Xeno V applied three times in 60 s (14.15 ± 0.67) and it was significantly higher, almost double, when compared to the control group. Within tested adhesives, means followed by the same lowercase letters in row and superscript letters in column indicate significant statistical differences.
The results of EDS analyses of the resin-dentin interface ( Figures 5-8) confirmed that the concentrations of different elements corresponded with specific layers of that interface. In dentin, mainly calcium and phosphorus were detected. The adhesive layer contained carbon, which is the main element present in resin polymer. Silicon was detected in adhesive layers of Adper Easy One and Single Bond Universal, which can be attributed to the presence of nanofiller in the composition of these adhesives. All tested specimens showed increased content of silicon and aluminum, derived from inorganic filler, in composite material.

Shear Bond Strength
Among all tested adhesives, the highest SBS results were obtained for Single Bond Universal, a universal adhesive ( Table 4). The increase in number of applications up to three along with 60 s application time resulted in higher SBS for Single Bond Universal adhesive than the control group (from 16.34 ± 4.62 up to 31.40 ± 2.83 MPa).

Shear Bond Strength
Among all tested adhesives, the highest SBS results were obtained for Single Bond Universal, a universal adhesive ( Table 4). The increase in number of applications up to three along with 60 s application time resulted in higher SBS for Single Bond Universal adhesive than the control group (from 16.34 ± 4.62 up to 31.40 ± 2.83 MPa).

Shear Bond Strength
Among all tested adhesives, the highest SBS results were obtained for Single Bond Universal, a universal adhesive ( Table 4). The increase in number of applications up to three along with 60 s application time resulted in higher SBS for Single Bond Universal adhesive than the control group (from 16.34 ± 4.62 up to 31.40 ± 2.83 Mpa). Within tested adhesives, means followed by the same lowercase letters in row and superscript letters in column indicate significant statistical differences.
When multi-mode adhesives were considered, Prime & Bond One Select presented lower bond strength values in all study groups compared with Single Bond Universal (Figure 9). None of the modifications of application procedures contributed to a significant increase of SBS of Prime&Bond One Select. In case of Single Bond Universal doubling or tripling the number of applications along with prolonged application time in groups 2A40 and 3A60 resulted in a statistically significant increase in SBS. Within tested adhesives, means followed by the same lowercase letters in row and superscript letters in column indicate significant statistical differences.
When multi-mode adhesives were considered, Prime & Bond One Select presented lower bond strength values in all study groups compared with Single Bond Universal (Figure 9). None of the modifications of application procedures contributed to a significant increase of SBS of Prime&Bond One Select. In case of Single Bond Universal doubling or tripling the number of applications along with prolonged application time in groups 2A40 and 3A60 resulted in a statistically significant increase in SBS.   The lowest values of SBS were observed for SE adhesive, Xeno V applied once (2.24 ± 1.0 MPa), however the bond strength significantly increased when application methods were modified. The highest mean SBS was observed in group 3A60 (7.68 ± 4.04 MPa) (Table 4, Figure 10b).  Within tested adhesives, means followed by the same lowercase letters in row and superscript letters in column indicate significant statistical differences.
When multi-mode adhesives were considered, Prime & Bond One Select presented lower bond strength values in all study groups compared with Single Bond Universal (Figure 9). None of the modifications of application procedures contributed to a significant increase of SBS of Prime&Bond One Select. In case of Single Bond Universal doubling or tripling the number of applications along with prolonged application time in groups 2A40 and 3A60 resulted in a statistically significant increase in SBS. With regard to self-etch adhesives, SBS of Adper Easy One was higher than of Xeno V adhesive in all respective study groups. Bond strength values of both Adper Easy One and Xeno V increased in groups, which involved doubling or tripling application time: 2A40, 3A60 ( Figure 10). A significant interaction between factors (type of adhesive, modification of application procedure) was revealed using Kruskal-Wallis ANOVA. For most adhesives, except Prime & Bond One Select, each modification contributed to an increase in SBS. However, Dunnett's post hoc test showed that the difference in bond strength was statistically significantly higher than the control only in 2A40 and 3A60 groups for Adper Easy One, Single Bond Universal and Xeno V (Figures 9 and 10).

Discussion
A positive correlation between prolonged application time and higher resin-dentin bond strength values of simplified adhesives has been demonstrated [31,32]. This could be attributed to both higher penetration of monomers into decalcified dentin and higher water and solvent evaporation before light curing. The conversion degree and mechanical properties of the adhesive polymer are compromised if solvent content within the adhesive solution is high [33,34]. When water is present during the conversion from monomer to polymer, the formation of a cross-link polymer within the collagen fiber network is prevented [35][36][37]. Therefore, if the application procedure is prolonged, a thicker hybrid layer of higher conversion degree and better quality can be obtained [26,31,38].
Multiple application of SE and MM adhesives without light-curing each layer causes a minimal increase in the thickness of the adhesive layer (a physical factor), but it may have effects of chemical nature [8]. During the first application, the adhesive etches dentin, but is likely to be quickly buffered. Providing more acidic monomers during the next application may increase the ability of the adhesive to etch dentin. High-pressure or prolonged air-drying enhances solvent evaporation between successive applications of the adhesive. Therefore, the concentration of acidic monomers increases. This may contribute to a more thorough infiltration of the collagen fibers and more uniform polymerization of the adhesive [27]. Thus, without increasing the thickness, the adhesive layer of higher quality is obtained, which ensures a stronger bond between the adhesive and dentin [27].
The primary reasons for failure of composite restorations are secondary caries, marginal defects and restoration fracture [39]. Therefore, achieving a stable and durable bonding interface plays a significant role in terms of dental composite performance. There is abundant literature confirming the advantage of the etch-and-rinse over the self-etch approach when using enamel [40]. However, when dentin is regarded, the bonding strategy depends on the pH of the adhesive and its composition. Simplified adhesives are classified according to their acidity as strong (pH ≤ 1), moderate (1 < pH < 2) or mild (pH = 2) [40]. Ermis et al. [41] added the fourth group to this classification, the so-called ultra-mild adhesives (pH ≥ 2.5). Low-pH adhesives present decreased dentin bond strength values when compared to mild ones [25]. The etching pattern they produce is similar to that of 37% phosphoric acid, but the dissolved calcium phosphates are not rinsed away and become embedded in hybrid layer. Since calcium phosphates are unstable in aqueous environments and considerably weaken the interfacial integrity, the use of strong SE adhesives is not recommended [3]. Self-etch or universal, mild adhesives are proven to achieve best dentin bond strength results when phosphoric acid etching is not involved [42]. When these adhesives are applied directly on unetched dentin, the tissue is partially demineralized. A substantial amount of hydroxyapatite crystals is left around the collagen fibrils. The micromechanical interaction takes place due to in situ polymerization of the monomers that infiltrate into the collagen network. Moreover, some of the universal adhesives interact chemically with hydroxyapatite. They contain functional monomers in their composition, such as MDP. These monomers interact chemically with calcium in the residual hydroxyapatite through ionic bonding, creating a non-soluble stable nano-layer [18,24]. It contributes to increasing mechanical strength and preventing the degradation over time of the bond interface.
When bonding to dentin is regarded, SE and universal adhesives still cannot compete with the excellent performance of conventional three step etch-and-rinse adhesives. In this study the efficacy of several methods to enhance their dentin bonding performance was evaluated, as well as their impact on the adhesive layer thickness. The methods included increasing the number of applications and extending the application time.
In the present study each modification of application procedure of Single Bond Universal contributed to an increase of SBS to dentin. The difference was not statistically significant in case of sole increased number of applications (2A20 and 3A20). However, in 2A40 and 3A60 groups, mean SBS results were significantly higher and almost doubled when compared to the control. It can be concluded that the chemical action of MDP monomer provided with subsequent applications in 2A20 and 3A20 groups could not fully develop in 20 s. Only after the application procedure was multiplicated within the prolonged time, did it allow Single Bond Universal to achieve higher SBS values. The considerable improvement in bond durability is attributed to an additional chemical reaction between MDP and dentin [3]. According to the adhesion-decalcification (AD) concept, the intensity and stability of molecular adhesion of calcium salts of acidic monomers to HAp is inversely proportional to their solubility [43]. The MDP possesses relatively strong decalcification ability, which enables it to form stable monomer-calcium salts, manifested as 'nano-layering' [18]. A more water-stable interface is produced, therefore contributing to bond durability. The monomer consists of a methylmethacrylate group, which allows for light-polymerization, and a phosphoric-acid group, enabling ionic interaction with calcium of HAp. Both groups are separated by a 10-fold carbon chain, which is responsible for wetting, solubility, flexibility, and the hydrophobicity-hydrophilicity balance of the monomer [44]. All of the abovementioned features of MDP may also be the reason why Single Bond Universal had the highest mean SBS compared to all other adhesives used in this study. Especially in 2A40 and 3A60 groups, where additional portions of MDP were supplied and sufficient time was ensured, the results of SBS to dentin were the highest.
In the case of Prime & Bond One Select, another universal adhesive used in this study, no study group obtained significantly higher results than the control. Mean SBS values were significantly lower than those of Single Bond Universal in each respective group. This was probably due to several mechanisms operating simultaneously. One factor is the adhesive layer thickness. Previous studies showed that increased adhesive layer thickness enhanced its elasticity [45]. Resin deformation suppressed polymerization, mastication or temperature-caused stress [46]. In this study it was measured that Prime & Bond One Select obtained thinner adhesive layer than Single Bond Universal in each respective group. Prime & Bond One Select applied in separately light-cured layers was shown to present both thicker adhesive layer and improved SBS [47]. Another factor contributing to the difference in bond strength between both universal adhesives is their acidity. The pH of Single Bond Universal is higher than of Prime&Bond One Select (2.7 vs. 1.6, respectively), which makes them an ultra-mild and a moderate adhesive. As mentioned above, previous studies proved low-pH adhesives, such as Prime & Bond One Select in this study, present lower bond strength than mild ones [25]. Another factor contributing to the higher SBS of Single Bond Universal in this study is the presence of inorganic nanofiller in its composition. Prime & Bond One Select, on the other hand, is an unfilled adhesive. Studies confirmed, that adhesives created reinforced, thicker hybrid layers, due to increased viscosity, when filler was included in their composition [48]. "Inverse" functionalized phosphoric acid esters are components of Prime & Bond One Select. As reported by the manufacturer, they are responsible for interacting chemically with dentin [49]. However, insufficient research is available to support the actual occurrence of this phenomenon. It should also be mentioned that, as described in the brochure provided by the manufacturer, all etching techniques: ER, SE and SEE can be utilized with Prime & Bond One Select. The composition of this adhesive and of Xeno V (Table 1) is comparable, however, which makes Prime & Bond One Select rather a SE product with broadened indications for use. The present study showed that Prime & Bond One Select produced statistically higher SBS than Xeno V in the control group, whereas the modification of application mode resulted in comparable bonding performance of both adhesives.
In the case of Xeno V, multiplying the number of applications led to an increase of bond strength, but the differences were not statistically significant (2A20, 3A20). These results are in line with other studies, which concluded that multiple applications of all-in-one adhesives did not significantly affect their bond strength to dentin [50]. The main factor responsible for the increase of SBS to dentin of Xeno V in this study was prolonged application time. Mean bond strength results in 2A40 and 3A60 groups were significantly higher than control, but not statistically different from each other. The adhesive layer was also the thickest in these groups for Xeno V, although the differences were not statistically significant. Other studies confirm that prolonging the application time of all-in-one adhesives contributes to an increase of bond strength [26,31]. Thicker hybrid layer of higher conversion degree and better quality is obtained, which allows higher SBS results [26,31].
When Xeno V was compared to Adper Easy One, another SE adhesive used in this study, it obtained lower SBS results in all respective study groups. The analysis of the composition and the pH of both adhesives may provide an explanation. Adper Easy One contains nanofiller which, as explained above, contributes to enhancing the thickness and quality of the adhesive layer. In this study, Adper Easy One obtained almost double the results of Xeno V in terms of adhesive layer thickness (13.31 ± 0.43 vs. 7.40 ± 0.82 µm) in control group. In other groups the differences were not statistically significant. Also, Vitrebond™ Copolymer and MHP (methacrylohexyl phosphate) functional monomer are present in the composition of Adper Easy One. These chemical compounds are reported to bond chemically with hydroxyapatite by the formation of a complex with calcium ions [43], therefore enhancing SBS results. Moreover, Xeno V is a moderate adhesive in terms of acidity and Adper Easy One is a mild one. As mentioned above, mild adhesives generally show higher dentin bond strength than the ones of lower pH [25].
In the case of Adper Easy One, similarly to Xeno V, the main factor responsible for improving bond strength results was prolonged application time. SBS results in 2A20 and 3A20 groups were higher than the control, but the differences were not statistically significant. Multiplying the number of applications within a prolonged time resulted in the enhancement of bond strength (from 6.10 ± 2.39 MPa in control group to 16.86 ± 2.29 MPa in 3A60 group). It can be concluded that when an additional amount of MHP was delivered with subsequent applications, the etching ability increased, more complete solvent evaporation was enabled, and an adhesive layer of better quality was obtained. Previous studies confirmed, that the quality of the adhesive layer is of paramount importance for dentin bond strength [26].
In terms of adhesive layer thickness results, Xeno V presented significantly higher, almost double mean thickness values after triple application in prolonged time of 60 s when compared to control. Most of the tested adhesives, Adper Easy One, Single Bond Universal, and Prime&Bond One Select, obtained lower thickness results in 2A20 group than in control or other study groups. However, for Prime & Bond One Select, the comparison of all the study groups with the control in terms of adhesive layer thickness proved that there was no statistically significant difference between them. The present study reported the great variability of the adhesive layer thickness measurements, that may result from a low number of specimens and few measurement points usually used in SEM observations [30]. Also, faulty positioning of the samples or undefined reference points could influence the measurements of adhesive layer thickness [51].
The present study showed that the correlation between dentin shear bond strength and adhesive layer thickness produced by tested all-in-one adhesives was statistically insignificant. Previous study reported similar findings on the lack of positive correlation between these parameters [52,53], depicting that it is the quality of the adhesive layer that might play a key role in dentin bond strength [51]. The increase in SBS of self-etch and universal adhesives observed in the present study could be related to an improvement of the resin-dentin interface quality, or the quality of the adhesive layer rather than to an increase in its thickness. This is in line with the findings of previous studies, which showed that some alterations of application procedures of all-in-one adhesives improved the quality of hybrid and/or adhesive layer and consequently led to an increase of SBS [25,27,50,54]. For all tested all-in-one adhesives, except for Prime&Bond One Select, increased number of applications along with extended application time (groups 2A40 and 3A60) resulted in significant increase in SBS when compared to the controls.
The limitations of this in vitro study should be taken into consideration. Bond strength tests are generally a valuable tool aiming to rank dental adhesives according to their bond strength performance. However, they do not examine crucial factors influencing bond strength, which take place in clinical situation, such as pH, temperature changes, moisture or masticatory stress [55]. Some authors regard bond strength tests to be invalid in order to conclude the clinical results of dental adhesives [56]. Other studies proved a relationship between bond strength values and clinical performance of these adhesives [16]. Therefore, as long as the validity of this study to predict the behavior of tested adhesives in the oral cavity is questionable, it can provide valuable data to identify substrate variables and establish the most effective application protocol for each adhesive tested. This information is relevant especially in clinically demanding situations where maximum bond strength is required, e.g., fiber post cementation, direct pulp capping, or class V cavity.
Considering the abovementioned results, the first part of the null hypothesis can be partially rejected; it is supported in case of Prime & Bond One Select and rejected in the case of Single Bond Universal, Xeno V, and Adper Easy One. Increasing the number of applications to two or three and extending the application time (40-60 s) significantly improved bond strength of the latter adhesives to dentin. Prime & Bond One Select showed no enhancement of bond strength to dentin in study groups. The second part of the null hypothesis, considering adhesive layer thickness after modified application, is rejected. The tested adhesives did not present increased thickness of the adhesive layer in the study groups. Furthermore, the third part of the null hypothesis is refuted. Improved bond strength of tested adhesives did not correspond with an increase of adhesive layer thickness.

Conclusions
Within the limitations of this study, modifying the application protocol of self-etch and universal adhesives by increasing the number of applications (two or three) and extending the application time (40-60 s) can be recommended to improve their performance. No correlation between adhesive layer thickness and bond strength of the tested adhesives was determined. The exact protocol of application of each product should be determined to ensure maximum performance due to the diverse compositions of simplified all-in-one adhesives.