| Sign In to gain access to subscriptions and/or personal tools. |
Contraction Stress in Dentin Adhesives Bonded to Dentin
1 Division of Pediatric Dentistry, Hokkaido University, Graduate School of Dental Medicine, Kita 13, Nishi 7, Kitaku, Sapporo 060–8586, Hokkaido, Japan; and Correspondence: * corresponding author, masanori-h{at}mue.biglobe.ne.jp
Adhesives cured under constrained conditions develop contraction stresses. We hypothesized that, with dentin as a bonding substrate, the stress would reach a maximum, followed by a continuous decline. Stress development was determined with a tensilometer for two total-etch systems and two systems with self-etching primers. The adhesives were placed in a thin layer between a glass plate and a flat dentin surface pretreated with phosphoric acid or self-etching primer. After an initial maximum shortly after light-curing, the stress decreased dramatically for the total-etch systems (70%) and, to a lesser extent, for the adhesives with self-etching primers (30%). The greater stress decrease for the total-etch systems was ascribed to water and/or solvents released into the adhesives from the fully opened dentinal tubules by the pulling/sucking action of the contraction stress. This happened less with the adhesives with self-etching primers, where the tubules remained mainly closed.
Key Words: polymerization • contraction stress • dentin adhesive • resin composite
During setting, with a resin composite restoration, which is adhesively bonded to the cavity walls, a tensile stress will develop (Feilzeret al., 1987; Ferracane, 2005), which is exerted on the cavity walls (Causton et al., 1985; Ferracane and Mitchem, 2003). The presence of an adhesive layer between the dentin walls and the resin composite restoration may provide stress relief during light-curing by acting as an elastic buffer (Kemp-Scholte and Davidson, 1990; Choi et al., 2000; Ausiello et al., 2002; Braga et al., 2003), but may also suppress further stress build-up in the period after light-curing. The latter has been observed recently with experiments in a universal testing machine, where a light-cured resin composite, placed between glass and dentin with various adhesives, consistently showed declining stress curves shortly after light-curing (Bolhuis et al., 2006). A similar decrease was not observed when the composite was placed between glass and steel. At this point, it is not clear which factors are involved in the observed decline of stress, but analysis of the data suggests (Bolhuis et al., 2006) that the morphology of the dentin surface plays an important role. The surface morphology may allow fluids contained in the dentin to be released into the adhesive layer (Hashimoto et al., 2005) by the pulling action of the polymerization stress generated by the composite, thereby relieving some of the stress. To what extent this will occur may depend on how "open" the dentin surface is, which will be determined by the surface treatment required for a particular bonding system. For example, total-etch systems, which utilize phosphoric acid, remove the smear layer and completely open the dentinal tubules, while systems with self-etching primers dissolve and mix with the smear layer, closing off most of the tubules. The observed decline in stress with time in the above-mentioned experiments (Bolhuis et al., 2006) may become more pronounced, and could reveal more details, when the adhesive is placed between glass and dentin without a composite. In this situation, the polymerizing adhesive pulls at the dentin surface, but with composite included, it is the pre-cured adhesive that is pulled by the contraction stress of the polymerizing composite. The aim of this study was to investigate the course of the polymerization contraction stress development of thin adhesive layers between glass and dentin for total-etch systems and systems with self-etching primers. The hypothesis tested was that a greater decline of stress after light-curing would be seen for total-etch systems than for systems with self-etching primers. To monitor the effect on stress development when the adhesives are pre-cured, we also conducted measurements on these adhesives combined with composite. We performed experiments with the adhesives between glass and steel, to see the effect on the course of stress development when a bonding substrate other than dentin was used.
Adhesives and Resin Composite Two total-etch systems and two systems with self-etching primers were used in this study (Table  ). The composite used was Filtek Z250/A2 (3M/ESPE, St. Paul, MN, USA).
Preparation of Dentin and Steel Specimens Ninety-six cylindrical cores of dentin, 7.0 mm in diameter, were cut from the roots of central bovine incisors, normal to the flat ground mesial or distal root surface, by means of a hollow diamond drill with copious water cooling (Diamant Boart Inc., Vianen, The Netherlands). Each dentin core was mounted at the free end of a rotatable rod with the central axes aligned. A 1.0-mm length of the free end of each dentin core was trimmed to a diameter of 6 mm, by means of a diamond bur in an air-rotor with water spray (Fig. 1 ). The flat end of each core was finally wet-ground with SiC paper up to grit 600, creating a smear layer at the bonding site, and stored in water until testing. Similarly shaped steel specimens were machined, and the flat surfaces were sandblasted.
Dentin Surface Treatment Prior to Bonding The surfaces of the dentin cores used with the Excite and One-Step Plus adhesive systems were acid-etched with 37% phosphoric acid for 15 sec, thoroughly washed, and blot-dried with an absorbent pellet, leaving the dentin surface visibly moist (wet-bonding). With the Clearfil Protect Bond and Clearfil Liner Bond 2V adhesive systems, the dentin surfaces were covered with the self-etching primers, left undisturbed for 20 sec, and then air-dried for 3 sec to remove the volatile solvents.
Contraction Stress Measurements The adhesives were applied to the pre-treated dentin or sandblasted steel surface (for Excite and One-Step Plus in two coats and gently air-dried for 3 sec to allow the solvent to evaporate), and then the crosshead was lowered toward the glass plate and adjusted to a position to form an adhesive layer of 15 µm. To avoid forcing adhesive into the opened dentin tubules, which could influence the stress measurements, we lowered the crosshead at a very low speed, to keep the force on the adhesive at 0 N. The adhesive layer was then light-cured from underneath the glass plate for 20 sec (600 mW/cm2). The contraction stress development was recorded continuously, from the start of light-curing up to 30 min. With dentin, the number of experiments was n = 12, and with steel, n = 3. For the adhesive/composite combinations, the applied adhesives were first light-cured for 20 sec, then the composite was applied and the crosshead lowered to obtain a layer thickness of 1 mm of the composite. Excess of expressed composite was removed, and the layer was then light-cured for 40 sec. The number of experiments was n = 12.
Statistical Analysis
The course of stress development showed a maximum at approximately 1 min after the start of light-curing, followed by a clear decline (Figs. 2A, 2B ), except for the adhesives cured between glass and steel (Fig. 2C).
Paired t tests showed a significant difference between the maximum and 30-minute values for each adhesive (Fig. 3A ) or adhesive/composite combination (Fig. 3B).
One-way ANOVA and Tukey’s post hoc tests showed differences for Excite and One Step Plus between the maximum stress values and between the 30-minute values, respectively, but for Liner Bond 2V and Protect Bond (Fig. 3A  ), and for the adhesive/composite combinations (Fig. 3B), the differences between adhesives were not significant.
The course of stress development of polymerizing resins, as measured in a universal testing machine, commonly shows a continuous increase of stress, at least during the first 30 min after light-curing (Dauvillier et al., 2003; Watts et al., 2003), as was seen in the present experiments with the adhesives between glass and steel. However, between glass and dentin, the stress curves of the adhesives first reached a maximum approximately 1 min after the start of light-curing, and then decreased steadily. For the two total-etch systems, the decrease from maximum to 30 min was greater than 70%, and for the two systems with the self-etching primers, the decrease was approximately 30%. So the hypothesis was accepted, that a greater decline of stress after light-curing is seen for total-etch systems than for systems with self-etching primers. Clearly, dentin plays an important role in the stress decline, but the extent depends on the adhesive system used. With the total-etch systems, phosphoric acid demineralized the smear layer and the underlying intact dentin to expose collagen fibrils and the dentinal tubules. Since the adhesives were applied according to the wet-bonding technique on a moist surface, visibly wet, the monomers diffused, along with their solvents, among the collagen fibrils and penetrated the tubules. Upon light-curing of the thin adhesive layer between the dentin and glass, as was done in the present study, the developing contraction stress pulled at the underlying "open" tissue, which released residual water and/or solvent into the adhesive film and interfered with setting. This retarded not only stress development, but also ultimate stress. An interesting observation was that the ultimate stress (peak value) for One Step Plus (1.6 MPa) was significantly lower than that for Excite (4.4 MPa). This may be explained by the different solvents used in the two systems, acetone in One Step Plus and ethanol in Excite. Micromorphological studies have demonstrated that acetone-based adhesives are most effective in hybridization with the collagen fibrils and in tag formation (Gregoire et al., 2002; Mohan and Kandaswamy, 2005). The deep penetration of acetone may be the result of efficiently replacing water from the collagen network and dentinal tubules better than any other solvent carrier. But this could be a disadvantage in the evaporation step, just prior to light-curing, if the acetone evaporates completely. Residual acetone in the deeper dentinal structure, when "sucked up" by the contraction stress into the growing polymer network, may keep the polymer network dissolved and weak in the very first seconds of light-curing. With the ethanol-based Excite, this may not occur to the same extent, due to more complete evaporation (in the evaporation step) and a lesser solubility of the growing polymer in ethanol, which would allow for faster stiffening of the polymer network and thereby higher stress development. The steep decline of stress, starting approximately 1 min after the start of light-curing, for both One Step Plus and Excite, must be explained by further infiltration of water and/or solvents into the set polymerized resin, accelerated by the contraction stress. This strongly supports previous research on single-bottle adhesives, showing that the bonding resin layer behaves as a permeable membrane (Tay et al., 2004; Hashimoto et al., 2005). The transudation of fluids from the underlying dentinal structure into the resin layer (accelerated by the contraction stress) eventually reversed the contraction stress of One Step Plus into an expansion stress within the 30-minute measuring period, presumably by swelling of the poly-HEMA component in the adhesives. The stress curve for Excite did not cross the x-axis within 30 min, but from its slope, one could expect that this may occur later on, or that the stress will at least level to zero. With the systems using self-etching primers, the smear layer and smear plugs were not removed. The milder etching by the primers resulted in hybrid structures being formed with the smear layer and smear plugs and with intact dentin just beyond the smear layer (Tay et al., 2000). After application of the adhesive and light-curing under restrained conditions, the contraction stress pulled at a "closed" dentinal structure, in contrast to the situation for the total-etch systems, where the stress pulled at an "open" structure. As a result, the stress curves of both adhesives, Liner Bond 2V and Protect Bond, showed only a moderate decline in stress after reaching a maximum value. The course of the stress development of the two self-etching systems was quite similar, which may be due to similar chemical formulations of their primers and adhesives. Since no solvents other than water were present in the two adhesive systems, the decline could be caused only by water, which was pulled into the adhesive layer from the hybridized dentinal regions and/or from intact dentin below it. The moderate decline in stress supports earlier findings that adhesive systems based on self-etching primers produce good dentin sealing (Hashimoto et al., 2004), but are also permeable (Tay and Pashley, 2003; Hashimoto et al., 2005). In the clinical situation, adhesive layers are never cured under the constrained conditions as used in this study, but polymerization contraction stresses are also produced in free resin layers applied to the cavity walls. The stresses are generally shearing in nature, but at the outer borders of a layer, the stresses are normal to the surface (Feilzer et al., 1990), and probably also at regions where the adhesive layer changes in thickness. Whether these stresses are strong enough to pull fluids out of the underlying dentin surface cannot be concluded at this point, but in this study, the stresses produced when composite was applied to the adhesive and cured under constrained conditions were strong enough. All curves in the adhesive/composite combinations showed a maximum, followed by a decline. Constrained conditions can also occur in the clinical situation, and one may expect that fluids can be pulled from the underlying dentin surface into the adhesive layer, which is supported by previous research that adhesives are permeable (Hashimoto et al., 2004, 2005; Tay et al., 2004). Limitation of stress development and decline of stress following light-curing would have obvious clinical benefits. However, the withdrawal of fluids from the hybridized structures, leaving channels (which will fill with water), and the uptake of these fluids by the adhesive layer may have a negative effect on the durability of the bonded joint (Sano et al., 1999; Shono et al., 1999; Hashimoto et al., 2000, 2002; Armstrong et al., 2001; Li et al., 2001; Tay et al., 2003; Reis et al., 2005). It should be noted that the observed course of stress development, where a maximum is reached followed by a decline, may be caused partly by the viscoelastic properties of demineralized dentin (Pashley et al., 2003). However, its contribution will be small as it involves demineralized dentin of only a few micrometers. Another aspect that could have influenced the results was that there was no control of the amount of wetness and evaporation of solvents in the bonding procedures. These factors, which influence dentin shrinkage (Eddleston et al., 2003; Ito et al., 2005), may have contributed to the relatively large standard deviations of the data.
This study was financially supported by the Netherlands Institute for Dental Research (IOT). The authors thank Dr. R.L. Erickson, Professor, Creighton University School of Dentistry (NE), for his valuable suggestions and language editing of the manuscript. This paper is based on an abstract presented at the IADR CED and NOF meeting, Amsterdam, Sept. 14–17, 2005. Received for publication August 18, 2005. Revision received March 28, 2006. Accepted for publication April 21, 2006.
Journal of Dental Research, Vol. 85, No. 8,
728-732 (2006)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
