| Sign In to gain access to subscriptions and/or personal tools. |
Interaction of Glass-ionomer Cements with Moist Dentin
1 Paediatric Dentistry and Orthodontics, Faculty of Dentistry, University of Hong Kong, Prince Philip Dental Hospital, 34 Hospital Road, Hong Kong SAR, China; Correspondence: * corresponding author, kfctay{at}hknet.com
Glass-ionomer cements (GICs) are regarded as aqueous gels made up of polyalkenoic acid salts containing ion-leachable glass fillers. The consequence of water permeation across the GIC-dentin interface is unknown. This study used SEM, field-emission/environmental SEM (FE-ESEM), and TEM to examine the ultrastructure of GIC-bonded moist dentin. Dentin surfaces bonded with 6 auto-cured GICs were examined along the fractured GIC-dentin interfaces. Additional specimens fractured 3 mm away from the interfaces were used as controls. SEM revealed spherical bodies along GIC-dentin interfaces that resembled hollow eggshells. FE-SEM depicted similar bodies with additional solid cores. Energy-dispersive x-ray analysis and TEM showed that the spherical bodies consisted of a silicon-rich GIC phase that was absent from the air-voids in the controls. The GIC inclusions near dentin surfaces result from a continuation of the GI reaction, within air-voids of the original polyalkenoate matrix, that occurred upon water diffusion from moist dentin.
Key Words: glass ionomer • spherical bodies • water • diffusion
Glass-ionomer cements (GICs) exhibit several clinical advantages compared with other restorative materials. They include physicochemical bonding to tooth structures (Glasspoole et al., 2002), long-term fluoride release, and low coefficients of thermal expansion (Nassan and Watson, 1998). However, their low mechanical strength compromises their use in high-stress-bearing restorations (Kerby and Knobloch, 1992). The acid-base setting reaction in GICs is initiated by the depletion of metallic ions from the fluoro-aluminosilicate glass (FASG) fillers by polyalkenoic acid, leaving siliceous hydrogel layers on the surface of the glass particles (Hatton and Brook, 1992; Nicholson, 1998). Cross-linking of these ions with polyalkenoic acid results in the formation of a primary polysalt matrix within the set cement (Maeda et al., 1999). A secondary reaction that leads to the deposition of a silicate phase contributes to the gradual increase in compressive strength with time (Wasson and Nicholson, 1993; Matsuya et al., 1996). To date, both ultrastructural and analytical evidence is available on the existence of an intermediate layer along the GIC-dentin/enamel interfaces that is caused by ion exchange between the material and tooth substrate (Geiger and Weiner, 1993; Ngo et al., 1997; Ferrari and Davidson, 1997; Sennou et al., 1999; Tay et al., 2001a; Yip et al., 2001). Water permeation from dentin into GICs (Watson et al., 1991, 1998) or resin-modified GICs (Sidhu and Watson, 1998) has been observed by dynamic laser confocal microscopy. This phenomenon of localized water sorption has not been previously observed in GIC-bonded dentin. The consequence of such water permeation across the GIC-dentin interface is unknown. This study used scanning electron microscopy (SEM), field-emission/environmental SEM (FE-ESEM), and transmission electron microscopy (TEM) to examine the regional differences in the ultrastructure of GIC bonded onto moist human dentin. Dentin surfaces bonded with 6 auto-cured GICs were examined along the fractured GIC-dentin interfaces. Additional GIC specimens were fractured 3 mm away from the interfaces, and fractured unbonded GIC samples were used as controls. The null hypothesis tested was that there is no difference among the ultrastructural features of GICs fractured at different locations from the bonded dentin, and fractured unbonded GICs.
Bonding was performed on the occlusal surfaces of deep coronal dentin of extracted human third molars. The teeth were collected after the patients’ informed consent had been obtained under a protocol reviewed and approved by the institutional review board at the Medical College of Georgia, Augusta, USA. The teeth were stored in 0.9% NaCl at 4°C and were used within 1 mo of extraction. The occlusal enamel was removed by means of a slow-speed saw (Isomet, Buehler Ltd., Lake Bluff, IL, USA) under water cooling, while the roots of these teeth were left intact.
Experimental Design Additional GICs were mixed and applied to the surfaces of mylar strips, which were then allowed to cure without being bonded to tooth substrates. After initial setting, both the bonded and unbonded GICs were covered with a thin layer of light-cured, oxygen-inhibition-layer-free, glaze resin (BisCover, Bisco Inc., Schaumburg, IL, USA), to prevent them from losing or gaining water via external sources. After storage at 37°C and 100% relative humidity for 48 hrs, the bonded specimens were cut occlusogingivally into 0.9 x 0.9 mm beams, following the method used for evaluation of microtensile bond strengths of GICs (Tay et al., 2001a). With this technique, none of the specimens fractured during preparation into beams, confirming our previous experience. Similar beams were prepared from the unbonded GICs. These beams were sealed in bottles at 100% humidity for 48 hrs before analyses. For SEM examination, 5 beams were randomly selected from the teeth bonded with each GIC, and fractured close to the GIC-dentin interface by three-point bending (razor blade at interface of end-supported beams). We fractured 5 additional bonded beams, at a location that was approximately 3 mm away from the bonded interfaces, by moving the supports. Both the dentin and GIC sides of these beams were air-dried, sputter-coated with gold/palladium, and examined with a SEM (Cambridge Stereoscan 440, Cambridge, UK) operating at 10 kV. A similar number of beams from each group was used for FE-ESEM examination. The beams were fractured on-site, inserted immediately into the Peltier (cooling) stage of the microscope (Philips XL-30, Eindhoven, The Netherlands), and covered with water to avoid desiccation of the fractured GICs. They were examined at 15 kV, wet and without being coated, with the temperature of the Peltier stage fixed at 4°C and the vapor pressure of the specimen chamber varying between 5.4 and 5.9 Torr to generate a relative humidity of 96–99.9%. We used energy-dispersive x-ray (EDX) analysis to determine the composition of various structural phases identified by FE-ESEM along the fractured GIC-dentin interfaces. To determine the extension of spherical bodies from bonded interfaces, we bonded 2 teeth with Fuji VII to form 6-mm-high cores over the surface of moist dentin, with the surface sealed with glaze resin. Bonded GIC-dentin slabs 2 mm wide were sectioned occluso-gingivally from the bonded teeth. They were polished under wet conditions with 1200-grit SiC paper, followed by 1 µm alpha alumina (Buehler Ltd.). These surfaces were briefly etched for 15 sec with 10% phosphoric acid (Bisco Inc.) to remove the smear layer and to bring the interfaces into relief. Two additional teeth were similarly bonded with Fuji VII, but without the surfaces being sealed with glaze resin. After setting of the GIC, these 2 teeth were immersed in distilled water for 10 min before storage at 100% relative humidity for 48 hrs. GIC-dentin slabs 2 mm wide were similarly prepared and examined by FE-ESEM. TEM examination was performed only on bonded GIC-dentin interfaces. Two beams were examined for each GIC. Intact beams (i.e., without being fractured) were processed and resin-embedded according to the TEM protocol for GIC examination described by Tay et al.(2001a). Examination of 90-nm-thick sections was performed with a transmission electron microscope (Philips EM208S), operating at 80 kV.
Since the results were highly reproducible in all 6 GICs, only representative Figs. are shown. As expected, artifactual cracks were ubiquitously observed within the polyalkenoate matrix when the specimens were examined by conventional SEM. The corresponding dentin (Fig. 1A  ) and GIC (Fig. 1B) sides of a representative beam that was fractured along the GIC-dentin interface demonstrated the presence of numerous spherical bodies within the air-voids that were trapped inside the polyalkenoate matrix. The majority of the air-voids were filled with spherical bodies along the interface. These spherical bodies were easily discerned from the angular FASG fillers (Fig. 1C). Partially fractured spherical bodies revealed their hollow eggshell-like internal structure (Fig. 1D). Spherical bodies were absent from both the bonded specimens that were fractured at sites remote from the bonded interfaces (Fig. 1E), and in fractured unbonded GICs, where only air-voids could be identified. These spherical bodies ranged in size from 14 to 30 µm. Their wall thickness varied within the same spherical body and ranged from 0.4 to 3 µm.
Spherical bodies could also be seen under FE-ESEM when bonded GIC specimens were fractured along the GIC-dentin interface (Fig. 2A ). In the absence of desiccation (Cowan et al., 1996), no artifactual cracks were observed, and the spherical bodies were closely adapted to the GIC matrix. Dehydration cracks began to form in the polyalkenoate matrix when the pressure of the microscope chamber was reduced to 5.4 Torr (Fig. 2B). Fractured spherical bodies appeared solid under FE-ESEM examination (Fig. 2C). Similar to SEM observations, these spherical bodies were absent in unbonded GICs (Fig. 2D), as well as in bonded GICs that were fractured at sites remote from the bonded interfaces. Representative EDX spectra taken from the FASG fillers, polyalkenoate matrix, and spherical bodies are shown in Fig. 2E. As expected, Ca and P were predominantly absent from the FASG fillers. The Si content in the spherical bodies was approximately three times higher than that of the polyalkenoate matrices.
FE-ESEM examination of polished slabs revealed that, for resin-sealed GICs, spherical bodies were observed only along the GIC-dentin interfaces (Fig. 3A ) and not along the surfaces of the resin-sealed GICs. The location of the spherical bodies that were farthest from the GIC-dentin interface was about 350 µm. No spherical bodies could be identified from the bonded GIC-enamel interfaces (not shown). For the unsealed GICs, spherical bodies could also be seen from the unsealed GIC surface (Fig. 3B). These bodies could be identified at about 180 µm from the GIC surface. Beyond this depth, only air-voids were exclusively observed.
TEM of GIC-dentin interfaces revealed the presence of a 2- to 3-µm-thick intermediate layer between the dentin and all of the GICs that were fractured during ultramicrotomy (Fig. 4A ). A siliceous hydrogel layer could also be identified along the periphery of the FASG fillers (Fig. 4B). Spherical bodies that were occasionally sectioned were found to be hollow, completely devoid of FASG fillers, and with an ultrastructural structure that resembled that of the adjacent polyalkenoate matrix phase more than the siliceous hydrogel layers of the FASG fillers (Fig. 4C).
The spherical bodies were consistently identified in all 6 auto-cured GICs during ultrastructural examination of the bonded interfaces by SEM, FE-ESEM, and TEM. The presence of these spherical bodies in the matrices of GICs at the GIC-dentin interface, but their absence 0.5 mm from the dentin, suggests that they are formed from water that slowly permeates from dentin into the GIC matrix. The presence of moisture in the specimen chamber of FE-ESEM allows for examination of GIC in a hydrated status. The close association of these spherical bodies with the GIC-dentin interface under FE-ESEM examination suggested that the mechanism of spherical body formation depends on the permeability of the GIC-dentin interface and the availability of water from the underlying moist dentin. EDX and TEM results further confirmed that the shells of these spherical bodies were not simply siliceous hydrogel layers of the FASG fillers, since the latter contained distinct fluoride-rich, electron-dense, "seed-like" phases that are characteristic of the original glass composition (Barry et al., 1979; Tay et al, 2001b). Rather, the eggshell-like structure represents a distinct Si-rich phase that has a composition slightly different from that of the original polyalkenoate matrix. The Si-rich phase is probably formed as a result of water diffusion across the bonded interface, initiating the delayed GI reaction that occurred during the maturation phase of GIC. Since permeation of water into GIC is a diffusion-controlled process, we speculate that a longer storage time will allow more spherical bodies to form. The presence of this secondary phase resembles the silica phase that was observed following the aging of GICs (Wasson and Nicholson, 1993; Matsuya et al., 1996). Undeniably, the EDX used during FE-ESEM examination has a relatively large depth of compositional analysis that may affect the accuracy of our results. However, recent examination of elemental compositions of spherical bodies by TEM/EDX analysis resulted in similar conclusions (Tay, unpublished results). The difference between the hollow appearance of the spherical bodies, as observed with SEM and TEM, and their solid appearance, under FE-ESEM, may be attributed to the presence of a pool of water within the eggshell-like structure when specimens were examined by the latter technique. When the specimens were examined by conventional SEM, the water probably evaporated from the fractured shells, revealing the original eggshell-like characteristics.
Since the GICs were protected with resin-coatings in the present study, water for the additional GIC reaction could be provided only by the underlying moist dentin. Such a phenomenon parallels what has been recently reported for single-step, self-etch dentin adhesives (Tay et al., 2002), in which water can diffuse from moist dentin through the adhesive layer to form water droplets along the adhesive-composite interface, with the adhesive layer acting as a semi-permeable membrane. Since an intermediate layer (Tanumiharja et al., 2001) is present between dentin and GIC (Fig. 4A The additional Si-rich phase represented by these spherical bodies probably played an important role in strengthening GIC along the bonded interface. Leiskar et al.(2003) reported a continuous increase in shear punch strength for up to 2 wks when Fuji IX GP, that was not initially protected by a resin coating, was exposed to water. Similar results were reported by Miyazaki et al.(1996). The clinical implication is that bonded GICs will continue to draw water from the underlying moist dentin for additional acid-base reaction—a process that does not occur in GIC-bonded enamel, due to its lower water content (ten Bosch et al. 2000). The clinical disadvantage is that the GIC may draw water from dentin fast enough to cause the occasional post-operative sensitivity (Christensen, 1990) that is associated with their use. In conclusion, in the absence of water-sorption from external sources, the existence of a distinct Si-rich phase within the air-voids of GIC-bonded dentin is probably caused by the permeation of water across the GIC-dentin interface, as previously reported by Watson et al.(1998). Since similar water movement across bonded resin-modified GICs has resulted in the formation of an amorphous "absorption layer" (Sidhu and Watson, 1998; Sidhu et al., 1999, 2002), further research should be performed to examine if similar spherical bodies could be seen in this class of restorative material.
This study was based on the work partially performed by Cynthia Yiu for the fulfillment of the degree of Doctor of Philosophy, the University of Hong Kong. We thank W.S. Lee of the Electron Microscopy Unit, the University of Hong Kong, for technical assistance. The glass-ionomer cements examined were generous gifts from the respective manufacturers. This study was supported, in part, by grant 20003755/90800/08004/400/01, the Faculty of Dentistry, the University of Hong Kong, and by grant DE014911 from the NIDCR, USA. The authors are grateful to Zinnia Pang and Michelle Barnes for secretarial support. Received for publication February 15, 2003. Revision received December 5, 2003. Accepted for publication January 22, 2004.
Journal of Dental Research, Vol. 83, No. 4,
283-289 (2004)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
