- Email: [email protected]
Preliminary study on chitosan modified glass ionomer restoratives
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Preliminary study on chitosan modified glass ionomer restoratives Denise F.S. Petri a,∗ , Juliana Donega´ a , Andr´e M. Benassi a , Jorge A.J.S. Bocangel b a b
˜ Paulo, P.O. Box 26077, 05513-970 Sao ˜ Paulo, SP, Brazil Instituto de Qu´ımica, Universidade de Sao Faculdade de Odontologia, Universidade Federal de Pelotas, RS, Brazil
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. The aim of this study was to investigate the effect of chitosan (CH), a biocompat-
Received 10 February 2006
ible polysaccharide, on the flexural strength of glass ionomer restoratives (GIR) and on the
Accepted 20 June 2006
release of fluoride ions from GIR. Methods. Commercial GIR (Vidrion, SS White) has been modified by adding chitosan (CH, Fluka). Samples containing 0.0044, 0.012, 0.025 and 0.045 wt% CH were prepared, molded
Keywords:
and weighed. For flexural strength, sets of 10 specimens 10 mm × 2 mm × 2 mm of com-
Glass ionomers restoratives
mercial and CH modified GIR were prepared. Differences were analyzed by the one-way
Chitosan
analysis of variances (ANOVA) test, at significance level 0.05. The data were also analyzed
Fluoride release
by post hoc Tukey’s HSD for unequal n (Spjotvoll/Stoline) test. Scanning electron microscopy
Flexural strength
analyses were performed on the composites cryo-fracture surfaces. For the fluoride release tests and medium pH determination, discs with 10 mm diameter and 2 mm height were prepared in a PTFE mold placed between two glass slides. Samples were weighed in order to normalize each material test group. At least 10 samples of each material were prepared. Approximately 5 min after preparation the discs were transferred into individual glass flasks containing 50 mL of distilled water. The concentration of released fluoride was determined as a function of time by means of a fluoride ion selective electrode Orion 94-09 SC connected to an Ionanalyser (Orion Research Inc., USA). The medium pH was monitored as a function of time at (25 ± 1◦ C), using a Digimed DM20 potentiometer (Digicrom Instrumentos, Brazil) equipped with a combined glass electrode. Ellipsometric measurements were performed to quantify the thickness of adsorbed polymer (poly(acrylic acid) or the mixture of poly(acrylic acid) with CH). Results. The addition of 0.0044 wt% of CH led to a significant increase in the flexural resistance. CH contents higher than 0.022 wt% led to poor performance. For the same period of time the amount of fluoride ions released from CH modified GIR was much larger than that released from commercial GIR. CH catalyzed the fluoride release from GIR to the medium, especially from those with 0.0044 wt% of CH. As a consequence, the medium pH increased from 5.0 to 6.3. A model based on the formation of a polymeric network around the inorganic particles was proposed to explain the experimental findings. The adsorption of CH and poly(acrylic acid) onto planar Si/SiO2 substrates was quantified and supported the proposed model. Significance. The results presented here showed that the flexural strength of a commercial GIR can be considerably improved by the addition of a tiny amount of CH. Moreover, in the presence of CH, the release of fluoride ions from GIR is catalyzed. © 2006 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
∗
Corresponding author. Tel.: +55 11 30913831; fax: +55 11 38155579. E-mail address: [email protected] (D.F.S. Petri). 0109-5641/$ – see front matter © 2006 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2006.06.038
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
1.
1005
Introduction
Glass ionomer restoratives (GIR), frequently composed of calcium aluminosilicate, fluoride and poly(acrylic acid), are interesting materials because of their good adhesion to the calcified tissues, their ability to release fluoride and relatively low cost. However, they are brittle materials with relatively poor mechanical performance [1]. One way to improve the mechanical performance would be the introduction of a rubbery phase into the mixture. For instance, adding polybutadiene, an elastomer, to polystyrene was a successful solution found to improve the impact properties of polystyrene, a brittle polymer [2]. However, from a practical point of view it is not so trivial, because the rubbery phase must present good compatibility with the other phases. Therefore, ‘compatibilizers’ must be added to the system in order to provide good interfacial adhesion. Actually, the poor mechanical performance presented by GIR materials is not only due to their brittle characteristic. GIR are multicomponent materials. If the adhesion between each component is weak, or in other words, if the interfacial tension between each component is high, the mechanical properties are poor. Therefore, an additive able to reduce the interfacial tension or to increase the adhesion among the components might lead to a better mechanical performance. On the other hand, the anticariogenic effect of GIR is well known and considered a result of fluoride release. GIR are generally recommended where caries protection is needed. Chitosan, a natural linear biopolyaminosaccharide is obtained by alkaline deacetylation of chitin, which is the second most abundant polysaccharide next to cellulose [3–5]. Chitin is the principal component of protective cuticles of crustaceans such as crabs, shrimps, prawns, lobsters and cell walls of some fungi such as aspergillus and mucor. Chitin is a straight homopolymer composed of (1,4)-linked N-acetylglucosamine units, while chitosan is comprised of copolymers of glucosamine and N-acetyl-glucosamine [3–5]. Chitosan has at least one primary amino and two free hydroxyl groups for each C6 building unit. Chitosan is a weak base and is insoluble in water and organic solvents, however, it is soluble in dilute aqueous acidic solution (pH < 6.5), which can convert the glucosamine units into a soluble form, R–NH3 + . Chitosan presents properties such as biocompatibility and biodegradability, mucoadhesion [6–11], which can be advantageous for biomedical applications. Moreover, chitosan can be used in paint formulations to act as an antimicrobial agent and colloidal stabilizer [12]. Latex particles were prevented from Enterobacter cloaceae and Pseudomonas aeruginosa contamination upon addition of minute chitosan and dioctadecyldimethylammonium bromide doses [12]. However, few applications have been reported in dentistry [13,14]. Linden and co-workers [13] prepared polymeric hydrogels based on poly(acrylic acid) and metal salts, and chitosan, which were formed directly in the microchannels of dental hard tissues in order to tighten them. Pawlowska [14] observed that CH modified dental primers applied in rat dental pulp caused slight, reversible pathological changes in the pulp. In this work, the authors studied (i) the flexural resistance of CH modified GIR as a function of CH content and (ii) the amount of fluoride ions released from CH modified GIR as a function of time.
Fig. 1 – Chemical structures of PAA (a), chitin (b) and chitosan (c).
2.
Materials and methods
Vidrion R powder (batch number 03183) and liquid (batch number 03193) produced in Germany were purchase from SS White, Brazil. The powder is composed of 41.3% SiO2 , 28.4% Al2 O3 , 1.6% AlF3 , 15.6% CaF2 , 9.3% NaF and 3.8% AlPO4 , while the commercial liquid is composed of 50% poly(acrylic acid) (PAA), represented in Fig. 1a, and 50% of a mixture of maleic acid, tartaric acid and phosphoric acid [15]. Chitosan (CH) purchased from Fluka was purified by dissolving in 0.1 mol/L acetic acid and precipitating three times in 0.1 mol/L NaOH. Afterwards the CH flocs were freeze dried. CH used in this work is actually a copolymer composed of 20 wt% of chitin and 80 wt% of chitosan, the corresponding structures are shown in Fig. 1b and c, respectively. The average molecular weight, Mv , of 218,500 g/mol has been determined by capillary viscometry in 0.1 mol/L acetic acid and 0.2 mol/L NaCl, considering the Mark–Howink–Sakurada constants as a = 0.93 and K = 1.81 × 10−3 [16].
2.1.
Flexural strength
Solutions of CH were prepared in acetic acid, 0.1 mol/L. Aliquots of CH solution were added to the commercial liquid so that the CH concentration in the liquid varied from 0 (original liquid) to 2 g/L (pure chitosan solution). Regardless the CH content in the liquid the pH was always ∼1.0. The liquid was then mixed with the powder in order to prepare the samples. Table 1 shows the formulations used in the present study. All GIR samples were prepared according to manufacturer’s instructions using the scoops provided. The recommended quantities in weight corresponded to 0.22 g of Vidrion powder and 0.05 g of liquid, so that the content of CH in the GIR in wt% could be also calculated (Table 1).
1006
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
Table 1 – Mean flexural strength values, Mr (MPa), measured for original GIR and chitosan modified GIR Material code
CH concentration in the liquid (g/L)
GIR-Control GIR-CH10 GIR-CH25 GIR-CH50 GIR-CH100
Content of CH in the GIR (wt%)
0 (0 v/v, %) 0.2 (10 v/v, %) 0.5 (25 v/v, %) 1.0 (50 v/v, %) 2.0 (100 v/v, %)
0 0.0044 0.012 0.025 0.045
Mr (MPa) 14.27 18.14 17.00 15.07 6.88
± ± ± ± ±
2.60 3.26 3.98 4.34 1.63
The numbers in the brackets represent the content (% v/v) of solution 2 g/L CH.
For the flexural strength tests the specimens were made by inserting the composite material into a poly(tetrafluoro ethylene), PTFE, mold of 10 mm × 2 mm × 2 mm. After 5 min curing, the specimens were removed, coated with silicone wax and stored in distilled water at 37.5 ◦ C for 20 h. At least 10 samples of each material were prepared and weighed. Threepoint bending tests were performed using Instron equipment at (24 ± 1◦ C) at a cross-head speed of 0.5 mm/min. The flexural strength Mr (MPa) values were calculated by [15]: Mr =
3Qmax L 2bh2
(1)
where Qmax is the maximum load (N), L the distance between two points (6 mm), b the specimen width (2 mm) and h is the specimen height (2 mm). Statistical analysis was performed using the one-way analysis of variances (ANOVA) test, at significance level 0.05. The data were also analyzed by post hoc Tukey’s HSD for unequal n (Spjotvoll/Stoline) test. Scanning electron microscopy (SEM) analyses on the composites cryo-fracture surfaces were obtained in Phillips XL30 equipment. In order to avoid artifacts due to plastic deformations the samples were fractured under liquid N2 prior to the analysis.
2.2.
Fluoride release
For the fluoride release determination tests, discs with 10 mm diameter and 2 mm height were prepared in a PTFE mold placed between two glass slides. Samples were weighed in order to normalize each material test group. At least 10 samples of each material were prepared. Approximately 5 min after preparation the discs were transferred into individual glass flasks containing 50 mL of distilled water. The concentration of released fluoride [F− ] was determined as a function of time by means of a fluoride ion selective electrode Orion 9409 SC connected to an Ionanalyser (Orion Research Inc., USA). First of all, a calibration curve of potential E (mV) as a function of log fluoride concentration, [F− ], in mol/L, was obtained (Fig. 2). The linear coefficient of −63.4, obtained in the concentration range of 0.00006 to 0.5 mol/L, indicated that ionic strength effects can be neglected and that the electrode presented the ideal Nerst behavior. The dependence of E with log[F− ] was described by E (mV) = −244.16 − 63.4 log [F− ]
fluoride ions released is expressed as g/cm2 . For this reason mol/L was converted into g/cm2 , considering the volume of 50 mL, the molar mass of fluoride as 18.99 g/mol and the disc area (both sides) as 1.6 cm2 . Thus, the unit conversion mol/L to g/cm2 can be obtained by multiplying the concentration (mol/L) by 0.6.
2.3.
Medium pH determination
Tests discs with 10 mm diameter and 2 mm height were prepared in a PTFE mold placed between two glass slides. Samples were weighed in order to normalize each material test group. At least 10 samples of each material were prepared. Approximately 5 min after preparation the discs were transferred into individual glass flasks containing 50 mL of distilled water. The medium pH was monitored as a function of time at (25 ± 1◦ C), using a Digimed DM20 potentiometer (Digicrom Instrumentos, Brazil) equipped with a combined glass electrode. Distilled water was always used as blank.
2.4.
Ellipsometry
Ellipsometric measurements were performed with a DREX02C Ellipsometer (Ratzeburg, Germany), equipped with a He–Ne laser (632.8 nm) and angle of incidence set to 70◦ . The index of refraction n for each layer was considered: Si, 3.88–0.018i; for SiO2 , 1.462; for the polymer layer, 1.500 and for the air 1.000 [17–19]. The mean thickness (d) of the native oxide layer on the silicon wafers and adsorbed polymer was calculated from the ellipsometric angles and , using a
(2)
Therefore, the amount of fluoride ions (mol/L) released by the GIR samples was calculated by substituting the experimental E value in Eq. (2). Conventionally the amount of
Fig. 2 – Calibration curve of potential E (mV) as a function of log fluoride concentration, [F− ].
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
multilayer model composed of the substrate, the unknown layer, and the surrounding medium with the fundamental ellipsometric equation and iterative calculations with Jones matrixes [17]. Si/SiO2 wafers are commonly used as substrates for polymer adsorption studies [18,19]. They are advantageous because they can be easily characterized by ellipsometry, an optical technique, which allows the determination of very thin layers [17]. The Si/SiO2 wafers were immersed for 30 min in the commercial fluid (PAA) or in the modified fluid (PAA + CH 0.0044 wt%). Then the wafers were washed with distilled water, dried and analyzed by means of ellipsometry.
3.
Results
The mean flexural strength (Mr ) values measured for original GIR and chitosan modified GIR are presented in Table 1. Comparing the Mr values obtained for GIR and GIR-CH10 a significant increase in the Mr values with p = 0.010699 was observed, indicating that the addition of the tiny amount of 0.0044 wt% of CH in the GIR caused a relevant improvement in the mechanical performance. Upon increasing the CH content to 0.012 wt% no effect in the mechanical behavior could be observed, since GIR and GIR-CH25 presented no statistical dif-
1007
ferences when Mr values where compared with a p of 0.074906. CH contents larger than 0.022 wt% caused antagonistic effects on the Mr values. Exchanging the original liquid (poly(acrylic acid) and a mixture of maleic acid, tartaric acid and phosphoric acid) for CH solution (prepared in acetic acid) led to the lowest Mr values. This finding indicates that the presence of poly(acrylic acid) is mandatory for satisfactory mechanical performance. Although the results on the flexural strength showed that the tiny amount of 0.0044 wt% CH improved the interfacial adhesion between the GIR components, scanning electron micrographies of the cryo-fracture surface of the GIR-Control, GIR-CH10 and GIR-CH25 revealed similar morphological features, as shown in Fig. 3a–c, respectively. Neither an improved dispersion nor a possible segregation of CH chains could be observed by means of SEM images. The experiments on the amount of released fluoride ions were performed for GIR-CH10 and GIR-CH25 samples because they presented the highest Mr values (Table 1) and for original GIR samples, as control experiment. Fig. 4 shows the amount of fluoride ions (g/cm2 released from the GIR samples as a function of time. For up to 2 h of immersion in distilled water the amount of fluoride released increased modestly as a function of time, regardless of the composition of GIR. However,
Fig. 3 – Scanning electron micrographies of cryo-fracture surfaces of (a) GIR-Control, the bar corresponds to 50 m, (b) GIR-CH10 the bar corresponds to 20 m and (c) GIR-CH20, the bar corresponds to 20 m.
1008
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
Fig. 4 – Fluoride ions (g/cm2 ) released from GIR-Control (open circle), GIR-CH10 (open square) and GIR-CH25 (cross) as a function of time.
Fig. 5 – Variation of medium pH as a function of time for GIR-Control (open circle), GIR-CH10 (open square) and GIR-CH25 (cross).
after 21 h or 1 month, huge amounts of fluoride had been released. Especially in the case of GIR-CH10, the amounts of fluoride released after 21 h and 1 month were 1497 and 3740 g/cm2 , respectively. For comparison, the amounts of fluoride released from commercial resin modified glass ionomers cements after 1 month were close to 500 g/cm2 [20]. Fig. 5 shows the pH variation as a function of time for GIRControl, GIR-CH10 and GIR-CH25. A general tendency of pH increase with time was observed. This trend agrees with the increase of fluoride release with the time observed in Fig. 4. Hydrogen fluoride is a weak acid with pH 3.45. Since water is amphiprotic, released fluoride ions tend to form HF with the protons transferred from water, increasing the medium pH. In other words, the hydrolysis of weak acid salt always increases the medium pH.
4.
Discussion
In order to get an insight into the GIR properties, each component should be mentioned. In commercial GIR the fillers, SiO2 and Al2 O3 , present an isoelectric point at pH 6.8 [21] and 9.1 [22], respectively. Since the fluid pH is 1, during the GIR prepa-
Fig. 6 – Schematic representation of (a) PAA interacting with the surface of an inorganic particle and (b) PAA adsorbing onto CH bound to the particle surface.
ration the inorganic particles surfaces are uncharged, instead they carry hydroxyl groups on the surface. The commercial fluid is mainly composed of poly(acrylic acid), PAA. Its pH was estimated as 4.75 [23], so that at pH 1.0 the carboxylic groups along the PAA chains are mostly protonated. Therefore, the adhesion between the particles and PAA is mainly driven by hydrogen bonding between the hydroxyl and carboxylic acid groups. Moreover, strong association of carboxylate functions by hydrogen bonding is commonly observed in PAA solutions [23]. Fig. 6a represents this situation schematically. In the case of CH modified GIR, the addition of CH under acid conditions is mandatory to guarantee its solubility. At pH 1.0 protonated CH chains are not able to interact with the particle surface or with PAA chains by electrostatic interactions, because there is little negative charge on them. On the other hand, CH chains carry many hydroxyl groups and acetamide groups, which are able to bind to the particles hydroxyl groups and to PAA carboxylic groups by hydrogen bonding. The network formed by CH and PAA around the inorganic particles (Fig. 6b) might reduce the interfacial tension among the GIR components, improving mechanical performance. In the presence of inorganic fillers, reinforced complexes can be built with high mechanical performance. This situation might correspond to the GIR-CH10 samples (Table 1),
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
where the CH content was 0.0044 wt%. Upon increasing the CH content to 0.012 wt%, the mechanical performance worsened and reached a level similar to that observed for commercial GIR. This effect can be explained considering that some CH chains segregate, interacting with each other, and no longer with PAA or the particle surface. Similar behavior has been observed for composites composed of CH, PAA and hydroxyapatite intended for bone substitute applications. Optimum compressive strength was observed for 70 wt% hydroxyapatite [24]. One should note that the advantages of adding CH to commercial GIR are the high biocompatibility and hydrophilicity, which are important features for intracorporal applications. For instance, CH has been shown to possess mucoadhesive properties [6–10] due to molecular attractive forces formed by electrostatic interaction between positively charged CH and negatively charged mucosal surfaces. Chitosan–alginate coacervate capsules were prepared to control the release characteristics and physicochemical properties of drugs [25]. CH–carboxymethylcellulose complex microparticles have also been used to immobilize cell culture [26]. Fluoride release and uptake by glass-ionomers and related materials and its clinical effect is well reported in the literature [27,28]. The anticaries action of fluoride is rather complex because it involves different mechanisms: direct binding of fluoride, binding of metal–fluoride complex and action as a transmembrane proton carrier [27]. The most direct mode of action for fluoride involves binding of F or HF to specific sites of enzymes or to the heme portion of a variety of enzymes [27]. For instance, enolase inhibition is considered to be important in fluoride inhibition of glycolysis by intact cells, and this can occur at low fluoride concentrations down to the micromolar level in acidified environments [27]. Fluoride inhibition of heme catalase, heme peroxidase and phosphatases takes place at millimolar level [27]. A generality is that fluoride binding is favored by acidification [27]. Phan et al. [29] monitored the NADH oxidase activity as function of pH in the presence of fluoride and weak organic acids. Respiration inhibition for oral streptococci was observed for pH 4 and 1 mmol/L of fluoride. The results shown in Figs. 4 and 5 indicate that fluoride release and pH, increase as a function of time. After 21 h of incubation ∼2.5 mmol/L of fluoride was released from GIR-CH10, which is considered adequate for anticaries action. However, the pH was close to 5.7, which might be slightly above the optimal pH. Since HF is a weak acid, pH increase is always associated with a large release of fluoride ions. These are important results, but the most relevant finding is that only 0.0044 wt% of chitosan in the GIR has a catalytic effect on the fluoride release, or in other words, it makes the diffusion of fluoride through the GIR towards the medium faster. The catalytic effect became weaker when CH content in the GIR was 0.012 wt%. These trends corroborate those observed for the mechanical performance. The release of fluoride ions from the inorganic matrix seems to be favored when reinforced complexes have been formed (Fig. 6). Even in the case of segregation of some CH chains, the release is still much faster than that observed for commercial GIR. The catalytic effect can be related to the entropic gain associated with fluoride release. Generally, the adsorption of a polymer, which causes the displacement of small ions always drives the system to a more favorable ener-
1009
getic situation, because the entropy associated with small ions diffusing into a medium is always larger than that associated with macromolecules free in solution. The first attempt to check the validity of the model proposed in Fig. 6 to explain GIR properties, was measuring zeta potential of powder particles dispersed at pH 1.0 in (i) commercial liquid and (ii) commercial liquid with 0.0044 wt% of CH. The particle number density was in the order of 1 × 1010 particles/mL to avoid strong scattering. The zeta potential value corresponds to the potential at the shear plane of a charged surface and is involved in electrophoresis and other electrokinetic phenomena [30]. In the presence of commercial liquid at pH 1.0 the surfaces are uncharged, as already discussed. Therefore, one would expect zeta potential values close to zero. Upon adding CH to the medium, the zeta potential value should shift to a positive value. Unfortunately, the dispersions presented very high conductivity values, which prevented measurement. The second experimental approach to check off the model suggested in Fig. 6 involved planar Si/SiO2 wafers, simulating the GIR particles. The mean thicknesses of the adsorbed polymeric layers were quantified by ellipsometry and amounted to 0.28 ± 0.05 and 0.5 ± 0.1 nm, in the case of PAA and PAA + CH, respectively. The addition of 0.0044 wt% of CH led to a relative thickness increase of 0.22 nm, indicating that PAA and CH adsorbed onto Si/SiO2 wafers. These findings support the model based on network formation by CH and PAA around the inorganic particles (Fig. 6b).
5.
Conclusions
The addition of 0.0044 wt% of CH to a commercial GIR improved the mechanical performance and catalyzed the release of fluoride ions. This effect was explained based on a model where a polymeric network binds strongly around the inorganic filler. From a practical point of view, this study shows that CH, a biocompatible and low cost additive, can be easily inserted in the traditional formulations bringing about synergic effects.
Acknowledgements The authors acknowledge FAPESP and CNPq for financial support. The authors are grateful to Prof. Jorge Masini and Prof. ˜ Paulo, for allowing us to Paulo T. Sumodjo, University of Sao use the fluoride ion selective electrode Orion 94-09 SC connected to the Ionanalyser (Orion Research Inc., USA).
references
[1] Prosser HJ, Powis DR, Brant P, Wilson AD. Characterisation of glass-ionomer cements. 7. The physical properties of current cements. J Dent 1984;12:231–40. [2] Utracki LA. Polymer alloys and blends: thermodynamics and rheology. New York: Carl Hanser Verlag; 1989. [3] Muzzarelli RAA. Chitin. New York: Pergamon Press; 1977. [4] Roberts GAF. Chitin chemistry. Houndmills: Mac Millan Press; 1992.
1010
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
´ [5] Rinaudo M, Domard A. In: Skjak-Braek G, Anthonsen T, Sandford P, editors. Chitin and chitosan: sources, chemistry, biochemistry, physical properties and applications. London: Elsevier Applied Sci.; 1989. p. 1–37. [6] Lehr CM, Bouwstra JA, Schacht EH, Junginger HE. In vitro evaluation of mucoadhesive properties of chitosan and some other natural polymers. Int J Pharm 1992;78:43–8. [7] Needleman IG, Smales FC. In vitro assessment of bioadhesion for periodontal and buccal drug-delivery. Biomaterials 1995;6:617–24. [8] Rillosi M, Buckton G. Modeling mucoadhesion by use of surface-energy terms obtained from the Lewis-acid Lewis base approach. 2. Studies on anionic, cationic, and unionisable polymers. Pharm Res 1995;12:669–75. [9] Shimoda J, Onishi H, Machida Y. Bioadhesive characteristics of chitosan microspheres to the mucosa of rat small intestine. Drug Dev Ind Pharm 2001;27:567–76. [10] Kockisch S, Rees GD, Young SA, Tsibouklis J, Smart JD. Polymeric microspheres for drug delivery to the oral cavity: an in vitro evaluation of mucoadhesive potential. J Pharm Sci 2003;92:1614–23. [11] Sinha VR, Singla AK, Wadhawan S, Kaushik R, Kumria R, Bansal K, et al. Chitosan microspheres as a potential carrier for drugs. Int J Pharm 2004;274:1–33. [12] Vieira DB, Lincopan N, Mamizuka EN, Petri DFS, Carmona-Ribeiro AM. Competitive adsorption of cationic bilayers and chitosan on latex: optimal biocidal action. Langmuir 2003;19:924–32. [13] Linden LANG, Rabek JF, Adamczak E, Morge S, Kaczmarek H, Wrzyszczynski AA. Polymer networks in dentistry. Macromol Symp 1995;93:337–50. [14] Pawlowska E. The assessment of influence of chitosan on the dental pulp in rats. Adv Chitin Sci 1997;2:705–10. ´ [15] Anusavice, Kenneth J. Phillips materiais dentarios. 10th ed. Guanabara Koogan: Rio de Janeiro; 1998. p. 412. [16] Roberts GAF, Domszy JG. Determination of the viscometric constants for chitosan. Int J Biol Macromol 1982;4:374–7. [17] Azzam RMA, Bashara NM. Ellipsometry and polarized light. North Holland Publication: Amsterdam; 1987. [18] Fujimoto J, Petri DFS. Adsorption behavior of carboxymethylcellulose on amino-terminated surfaces. Langmuir 2001;17:56.
[19] Castro LBR, Petri DFS. Assemblies of concanavalin A onto carboxymethylcellulose. J Nanosci Nanotechnol 2005;5:2063–9. [20] Marks LAM, Verbeeck RMH, De Maeyer EAP, Martens LC. Effect of maturation on the fluoride release of resin-modified glass ionomer and polyacid-modified composite resin cements. Biomaterials 2000;21:1373–8. [21] Loidl-Stahlhofen A, Schmitt J, Noller J, Hartmann T, Brodowsky H, Schmitt W, et al. Solid-supported biomolecules on modified silica surfaces—a tool for fast physicochemical characterization and high-throughput screening. Adv Mater 2001;13:1829–34. [22] Singh BP, Menchavez R, Takai C, Fuji M, Takahashi M. Stability of dispersions of colloidal alumina particles in aqueous suspensions. J Coll Interface Sci 2005;291:181–6. [23] Nemec JW, Bauer Jr W. In: Mark HF, editor. Encyclopedia of polymer science and technology, vol. 1. New York: John Wiley & Sons; 1985. p. 211–34. [24] Sailaja GS, Velayudhan S, Sunny MC, Sreenivasan K, Varma HK, Ramesh P. Hydroxyapatite filled chitosan-polyacrylic acid polyelectrolyte complexes. J Mater Sci 2003;38:3653–62. ´ [25] Kim SK, Rha C. In: Skjak-Braek G, Anthonsen T, Sandford P, editors. In chitin and chitosan: sources, chemistry, biochemistry, physical properties and applications. London: Elsevier Applied Sci.; 1989. p. p. 365. ´ [26] Shioya T, Rha C. In: Skjak-Braek G, Anthonsen T, Sandford P, editors. In chitin and chitosan: sources, chemistry, biochemistry, physical properties and applications. London: Elsevier Applied Sci.; 1989. p. p. 627. [27] Marquis RE, Clock SA, Mota-Meira M. Fluoride and organic weak acids as modulators of microbial physiology. FEMS Microbiol Rev 2003;26:493–510. [28] Forsten L. Fluoride release and uptake by glass-ionomers and related materials and its clinical effect. Biomaterials 1998;19:503–8. [29] Phan TN, Nguyen PTM, Abranches J, Marquis RE. Fluoride and organic acids as respiration inhibitors for oral streptococci in acidified environments. Oral Microbiol Immunol 2002;17:119–24. [30] Adamson WA. Physical chemistry of surfaces. 5th ed. Toronto: John Wiley & Sons; 1990.
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Preliminary study on chitosan modified glass ionomer restoratives Denise F.S. Petri a,∗ , Juliana Donega´ a , Andr´e M. Benassi a , Jorge A.J.S. Bocangel b a b
˜ Paulo, P.O. Box 26077, 05513-970 Sao ˜ Paulo, SP, Brazil Instituto de Qu´ımica, Universidade de Sao Faculdade de Odontologia, Universidade Federal de Pelotas, RS, Brazil
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. The aim of this study was to investigate the effect of chitosan (CH), a biocompat-
Received 10 February 2006
ible polysaccharide, on the flexural strength of glass ionomer restoratives (GIR) and on the
Accepted 20 June 2006
release of fluoride ions from GIR. Methods. Commercial GIR (Vidrion, SS White) has been modified by adding chitosan (CH, Fluka). Samples containing 0.0044, 0.012, 0.025 and 0.045 wt% CH were prepared, molded
Keywords:
and weighed. For flexural strength, sets of 10 specimens 10 mm × 2 mm × 2 mm of com-
Glass ionomers restoratives
mercial and CH modified GIR were prepared. Differences were analyzed by the one-way
Chitosan
analysis of variances (ANOVA) test, at significance level 0.05. The data were also analyzed
Fluoride release
by post hoc Tukey’s HSD for unequal n (Spjotvoll/Stoline) test. Scanning electron microscopy
Flexural strength
analyses were performed on the composites cryo-fracture surfaces. For the fluoride release tests and medium pH determination, discs with 10 mm diameter and 2 mm height were prepared in a PTFE mold placed between two glass slides. Samples were weighed in order to normalize each material test group. At least 10 samples of each material were prepared. Approximately 5 min after preparation the discs were transferred into individual glass flasks containing 50 mL of distilled water. The concentration of released fluoride was determined as a function of time by means of a fluoride ion selective electrode Orion 94-09 SC connected to an Ionanalyser (Orion Research Inc., USA). The medium pH was monitored as a function of time at (25 ± 1◦ C), using a Digimed DM20 potentiometer (Digicrom Instrumentos, Brazil) equipped with a combined glass electrode. Ellipsometric measurements were performed to quantify the thickness of adsorbed polymer (poly(acrylic acid) or the mixture of poly(acrylic acid) with CH). Results. The addition of 0.0044 wt% of CH led to a significant increase in the flexural resistance. CH contents higher than 0.022 wt% led to poor performance. For the same period of time the amount of fluoride ions released from CH modified GIR was much larger than that released from commercial GIR. CH catalyzed the fluoride release from GIR to the medium, especially from those with 0.0044 wt% of CH. As a consequence, the medium pH increased from 5.0 to 6.3. A model based on the formation of a polymeric network around the inorganic particles was proposed to explain the experimental findings. The adsorption of CH and poly(acrylic acid) onto planar Si/SiO2 substrates was quantified and supported the proposed model. Significance. The results presented here showed that the flexural strength of a commercial GIR can be considerably improved by the addition of a tiny amount of CH. Moreover, in the presence of CH, the release of fluoride ions from GIR is catalyzed. © 2006 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
∗
Corresponding author. Tel.: +55 11 30913831; fax: +55 11 38155579. E-mail address: [email protected] (D.F.S. Petri). 0109-5641/$ – see front matter © 2006 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2006.06.038
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
1.
1005
Introduction
Glass ionomer restoratives (GIR), frequently composed of calcium aluminosilicate, fluoride and poly(acrylic acid), are interesting materials because of their good adhesion to the calcified tissues, their ability to release fluoride and relatively low cost. However, they are brittle materials with relatively poor mechanical performance [1]. One way to improve the mechanical performance would be the introduction of a rubbery phase into the mixture. For instance, adding polybutadiene, an elastomer, to polystyrene was a successful solution found to improve the impact properties of polystyrene, a brittle polymer [2]. However, from a practical point of view it is not so trivial, because the rubbery phase must present good compatibility with the other phases. Therefore, ‘compatibilizers’ must be added to the system in order to provide good interfacial adhesion. Actually, the poor mechanical performance presented by GIR materials is not only due to their brittle characteristic. GIR are multicomponent materials. If the adhesion between each component is weak, or in other words, if the interfacial tension between each component is high, the mechanical properties are poor. Therefore, an additive able to reduce the interfacial tension or to increase the adhesion among the components might lead to a better mechanical performance. On the other hand, the anticariogenic effect of GIR is well known and considered a result of fluoride release. GIR are generally recommended where caries protection is needed. Chitosan, a natural linear biopolyaminosaccharide is obtained by alkaline deacetylation of chitin, which is the second most abundant polysaccharide next to cellulose [3–5]. Chitin is the principal component of protective cuticles of crustaceans such as crabs, shrimps, prawns, lobsters and cell walls of some fungi such as aspergillus and mucor. Chitin is a straight homopolymer composed of (1,4)-linked N-acetylglucosamine units, while chitosan is comprised of copolymers of glucosamine and N-acetyl-glucosamine [3–5]. Chitosan has at least one primary amino and two free hydroxyl groups for each C6 building unit. Chitosan is a weak base and is insoluble in water and organic solvents, however, it is soluble in dilute aqueous acidic solution (pH < 6.5), which can convert the glucosamine units into a soluble form, R–NH3 + . Chitosan presents properties such as biocompatibility and biodegradability, mucoadhesion [6–11], which can be advantageous for biomedical applications. Moreover, chitosan can be used in paint formulations to act as an antimicrobial agent and colloidal stabilizer [12]. Latex particles were prevented from Enterobacter cloaceae and Pseudomonas aeruginosa contamination upon addition of minute chitosan and dioctadecyldimethylammonium bromide doses [12]. However, few applications have been reported in dentistry [13,14]. Linden and co-workers [13] prepared polymeric hydrogels based on poly(acrylic acid) and metal salts, and chitosan, which were formed directly in the microchannels of dental hard tissues in order to tighten them. Pawlowska [14] observed that CH modified dental primers applied in rat dental pulp caused slight, reversible pathological changes in the pulp. In this work, the authors studied (i) the flexural resistance of CH modified GIR as a function of CH content and (ii) the amount of fluoride ions released from CH modified GIR as a function of time.
Fig. 1 – Chemical structures of PAA (a), chitin (b) and chitosan (c).
2.
Materials and methods
Vidrion R powder (batch number 03183) and liquid (batch number 03193) produced in Germany were purchase from SS White, Brazil. The powder is composed of 41.3% SiO2 , 28.4% Al2 O3 , 1.6% AlF3 , 15.6% CaF2 , 9.3% NaF and 3.8% AlPO4 , while the commercial liquid is composed of 50% poly(acrylic acid) (PAA), represented in Fig. 1a, and 50% of a mixture of maleic acid, tartaric acid and phosphoric acid [15]. Chitosan (CH) purchased from Fluka was purified by dissolving in 0.1 mol/L acetic acid and precipitating three times in 0.1 mol/L NaOH. Afterwards the CH flocs were freeze dried. CH used in this work is actually a copolymer composed of 20 wt% of chitin and 80 wt% of chitosan, the corresponding structures are shown in Fig. 1b and c, respectively. The average molecular weight, Mv , of 218,500 g/mol has been determined by capillary viscometry in 0.1 mol/L acetic acid and 0.2 mol/L NaCl, considering the Mark–Howink–Sakurada constants as a = 0.93 and K = 1.81 × 10−3 [16].
2.1.
Flexural strength
Solutions of CH were prepared in acetic acid, 0.1 mol/L. Aliquots of CH solution were added to the commercial liquid so that the CH concentration in the liquid varied from 0 (original liquid) to 2 g/L (pure chitosan solution). Regardless the CH content in the liquid the pH was always ∼1.0. The liquid was then mixed with the powder in order to prepare the samples. Table 1 shows the formulations used in the present study. All GIR samples were prepared according to manufacturer’s instructions using the scoops provided. The recommended quantities in weight corresponded to 0.22 g of Vidrion powder and 0.05 g of liquid, so that the content of CH in the GIR in wt% could be also calculated (Table 1).
1006
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
Table 1 – Mean flexural strength values, Mr (MPa), measured for original GIR and chitosan modified GIR Material code
CH concentration in the liquid (g/L)
GIR-Control GIR-CH10 GIR-CH25 GIR-CH50 GIR-CH100
Content of CH in the GIR (wt%)
0 (0 v/v, %) 0.2 (10 v/v, %) 0.5 (25 v/v, %) 1.0 (50 v/v, %) 2.0 (100 v/v, %)
0 0.0044 0.012 0.025 0.045
Mr (MPa) 14.27 18.14 17.00 15.07 6.88
± ± ± ± ±
2.60 3.26 3.98 4.34 1.63
The numbers in the brackets represent the content (% v/v) of solution 2 g/L CH.
For the flexural strength tests the specimens were made by inserting the composite material into a poly(tetrafluoro ethylene), PTFE, mold of 10 mm × 2 mm × 2 mm. After 5 min curing, the specimens were removed, coated with silicone wax and stored in distilled water at 37.5 ◦ C for 20 h. At least 10 samples of each material were prepared and weighed. Threepoint bending tests were performed using Instron equipment at (24 ± 1◦ C) at a cross-head speed of 0.5 mm/min. The flexural strength Mr (MPa) values were calculated by [15]: Mr =
3Qmax L 2bh2
(1)
where Qmax is the maximum load (N), L the distance between two points (6 mm), b the specimen width (2 mm) and h is the specimen height (2 mm). Statistical analysis was performed using the one-way analysis of variances (ANOVA) test, at significance level 0.05. The data were also analyzed by post hoc Tukey’s HSD for unequal n (Spjotvoll/Stoline) test. Scanning electron microscopy (SEM) analyses on the composites cryo-fracture surfaces were obtained in Phillips XL30 equipment. In order to avoid artifacts due to plastic deformations the samples were fractured under liquid N2 prior to the analysis.
2.2.
Fluoride release
For the fluoride release determination tests, discs with 10 mm diameter and 2 mm height were prepared in a PTFE mold placed between two glass slides. Samples were weighed in order to normalize each material test group. At least 10 samples of each material were prepared. Approximately 5 min after preparation the discs were transferred into individual glass flasks containing 50 mL of distilled water. The concentration of released fluoride [F− ] was determined as a function of time by means of a fluoride ion selective electrode Orion 9409 SC connected to an Ionanalyser (Orion Research Inc., USA). First of all, a calibration curve of potential E (mV) as a function of log fluoride concentration, [F− ], in mol/L, was obtained (Fig. 2). The linear coefficient of −63.4, obtained in the concentration range of 0.00006 to 0.5 mol/L, indicated that ionic strength effects can be neglected and that the electrode presented the ideal Nerst behavior. The dependence of E with log[F− ] was described by E (mV) = −244.16 − 63.4 log [F− ]
fluoride ions released is expressed as g/cm2 . For this reason mol/L was converted into g/cm2 , considering the volume of 50 mL, the molar mass of fluoride as 18.99 g/mol and the disc area (both sides) as 1.6 cm2 . Thus, the unit conversion mol/L to g/cm2 can be obtained by multiplying the concentration (mol/L) by 0.6.
2.3.
Medium pH determination
Tests discs with 10 mm diameter and 2 mm height were prepared in a PTFE mold placed between two glass slides. Samples were weighed in order to normalize each material test group. At least 10 samples of each material were prepared. Approximately 5 min after preparation the discs were transferred into individual glass flasks containing 50 mL of distilled water. The medium pH was monitored as a function of time at (25 ± 1◦ C), using a Digimed DM20 potentiometer (Digicrom Instrumentos, Brazil) equipped with a combined glass electrode. Distilled water was always used as blank.
2.4.
Ellipsometry
Ellipsometric measurements were performed with a DREX02C Ellipsometer (Ratzeburg, Germany), equipped with a He–Ne laser (632.8 nm) and angle of incidence set to 70◦ . The index of refraction n for each layer was considered: Si, 3.88–0.018i; for SiO2 , 1.462; for the polymer layer, 1.500 and for the air 1.000 [17–19]. The mean thickness (d) of the native oxide layer on the silicon wafers and adsorbed polymer was calculated from the ellipsometric angles and , using a
(2)
Therefore, the amount of fluoride ions (mol/L) released by the GIR samples was calculated by substituting the experimental E value in Eq. (2). Conventionally the amount of
Fig. 2 – Calibration curve of potential E (mV) as a function of log fluoride concentration, [F− ].
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
multilayer model composed of the substrate, the unknown layer, and the surrounding medium with the fundamental ellipsometric equation and iterative calculations with Jones matrixes [17]. Si/SiO2 wafers are commonly used as substrates for polymer adsorption studies [18,19]. They are advantageous because they can be easily characterized by ellipsometry, an optical technique, which allows the determination of very thin layers [17]. The Si/SiO2 wafers were immersed for 30 min in the commercial fluid (PAA) or in the modified fluid (PAA + CH 0.0044 wt%). Then the wafers were washed with distilled water, dried and analyzed by means of ellipsometry.
3.
Results
The mean flexural strength (Mr ) values measured for original GIR and chitosan modified GIR are presented in Table 1. Comparing the Mr values obtained for GIR and GIR-CH10 a significant increase in the Mr values with p = 0.010699 was observed, indicating that the addition of the tiny amount of 0.0044 wt% of CH in the GIR caused a relevant improvement in the mechanical performance. Upon increasing the CH content to 0.012 wt% no effect in the mechanical behavior could be observed, since GIR and GIR-CH25 presented no statistical dif-
1007
ferences when Mr values where compared with a p of 0.074906. CH contents larger than 0.022 wt% caused antagonistic effects on the Mr values. Exchanging the original liquid (poly(acrylic acid) and a mixture of maleic acid, tartaric acid and phosphoric acid) for CH solution (prepared in acetic acid) led to the lowest Mr values. This finding indicates that the presence of poly(acrylic acid) is mandatory for satisfactory mechanical performance. Although the results on the flexural strength showed that the tiny amount of 0.0044 wt% CH improved the interfacial adhesion between the GIR components, scanning electron micrographies of the cryo-fracture surface of the GIR-Control, GIR-CH10 and GIR-CH25 revealed similar morphological features, as shown in Fig. 3a–c, respectively. Neither an improved dispersion nor a possible segregation of CH chains could be observed by means of SEM images. The experiments on the amount of released fluoride ions were performed for GIR-CH10 and GIR-CH25 samples because they presented the highest Mr values (Table 1) and for original GIR samples, as control experiment. Fig. 4 shows the amount of fluoride ions (g/cm2 released from the GIR samples as a function of time. For up to 2 h of immersion in distilled water the amount of fluoride released increased modestly as a function of time, regardless of the composition of GIR. However,
Fig. 3 – Scanning electron micrographies of cryo-fracture surfaces of (a) GIR-Control, the bar corresponds to 50 m, (b) GIR-CH10 the bar corresponds to 20 m and (c) GIR-CH20, the bar corresponds to 20 m.
1008
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
Fig. 4 – Fluoride ions (g/cm2 ) released from GIR-Control (open circle), GIR-CH10 (open square) and GIR-CH25 (cross) as a function of time.
Fig. 5 – Variation of medium pH as a function of time for GIR-Control (open circle), GIR-CH10 (open square) and GIR-CH25 (cross).
after 21 h or 1 month, huge amounts of fluoride had been released. Especially in the case of GIR-CH10, the amounts of fluoride released after 21 h and 1 month were 1497 and 3740 g/cm2 , respectively. For comparison, the amounts of fluoride released from commercial resin modified glass ionomers cements after 1 month were close to 500 g/cm2 [20]. Fig. 5 shows the pH variation as a function of time for GIRControl, GIR-CH10 and GIR-CH25. A general tendency of pH increase with time was observed. This trend agrees with the increase of fluoride release with the time observed in Fig. 4. Hydrogen fluoride is a weak acid with pH 3.45. Since water is amphiprotic, released fluoride ions tend to form HF with the protons transferred from water, increasing the medium pH. In other words, the hydrolysis of weak acid salt always increases the medium pH.
4.
Discussion
In order to get an insight into the GIR properties, each component should be mentioned. In commercial GIR the fillers, SiO2 and Al2 O3 , present an isoelectric point at pH 6.8 [21] and 9.1 [22], respectively. Since the fluid pH is 1, during the GIR prepa-
Fig. 6 – Schematic representation of (a) PAA interacting with the surface of an inorganic particle and (b) PAA adsorbing onto CH bound to the particle surface.
ration the inorganic particles surfaces are uncharged, instead they carry hydroxyl groups on the surface. The commercial fluid is mainly composed of poly(acrylic acid), PAA. Its pH was estimated as 4.75 [23], so that at pH 1.0 the carboxylic groups along the PAA chains are mostly protonated. Therefore, the adhesion between the particles and PAA is mainly driven by hydrogen bonding between the hydroxyl and carboxylic acid groups. Moreover, strong association of carboxylate functions by hydrogen bonding is commonly observed in PAA solutions [23]. Fig. 6a represents this situation schematically. In the case of CH modified GIR, the addition of CH under acid conditions is mandatory to guarantee its solubility. At pH 1.0 protonated CH chains are not able to interact with the particle surface or with PAA chains by electrostatic interactions, because there is little negative charge on them. On the other hand, CH chains carry many hydroxyl groups and acetamide groups, which are able to bind to the particles hydroxyl groups and to PAA carboxylic groups by hydrogen bonding. The network formed by CH and PAA around the inorganic particles (Fig. 6b) might reduce the interfacial tension among the GIR components, improving mechanical performance. In the presence of inorganic fillers, reinforced complexes can be built with high mechanical performance. This situation might correspond to the GIR-CH10 samples (Table 1),
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
where the CH content was 0.0044 wt%. Upon increasing the CH content to 0.012 wt%, the mechanical performance worsened and reached a level similar to that observed for commercial GIR. This effect can be explained considering that some CH chains segregate, interacting with each other, and no longer with PAA or the particle surface. Similar behavior has been observed for composites composed of CH, PAA and hydroxyapatite intended for bone substitute applications. Optimum compressive strength was observed for 70 wt% hydroxyapatite [24]. One should note that the advantages of adding CH to commercial GIR are the high biocompatibility and hydrophilicity, which are important features for intracorporal applications. For instance, CH has been shown to possess mucoadhesive properties [6–10] due to molecular attractive forces formed by electrostatic interaction between positively charged CH and negatively charged mucosal surfaces. Chitosan–alginate coacervate capsules were prepared to control the release characteristics and physicochemical properties of drugs [25]. CH–carboxymethylcellulose complex microparticles have also been used to immobilize cell culture [26]. Fluoride release and uptake by glass-ionomers and related materials and its clinical effect is well reported in the literature [27,28]. The anticaries action of fluoride is rather complex because it involves different mechanisms: direct binding of fluoride, binding of metal–fluoride complex and action as a transmembrane proton carrier [27]. The most direct mode of action for fluoride involves binding of F or HF to specific sites of enzymes or to the heme portion of a variety of enzymes [27]. For instance, enolase inhibition is considered to be important in fluoride inhibition of glycolysis by intact cells, and this can occur at low fluoride concentrations down to the micromolar level in acidified environments [27]. Fluoride inhibition of heme catalase, heme peroxidase and phosphatases takes place at millimolar level [27]. A generality is that fluoride binding is favored by acidification [27]. Phan et al. [29] monitored the NADH oxidase activity as function of pH in the presence of fluoride and weak organic acids. Respiration inhibition for oral streptococci was observed for pH 4 and 1 mmol/L of fluoride. The results shown in Figs. 4 and 5 indicate that fluoride release and pH, increase as a function of time. After 21 h of incubation ∼2.5 mmol/L of fluoride was released from GIR-CH10, which is considered adequate for anticaries action. However, the pH was close to 5.7, which might be slightly above the optimal pH. Since HF is a weak acid, pH increase is always associated with a large release of fluoride ions. These are important results, but the most relevant finding is that only 0.0044 wt% of chitosan in the GIR has a catalytic effect on the fluoride release, or in other words, it makes the diffusion of fluoride through the GIR towards the medium faster. The catalytic effect became weaker when CH content in the GIR was 0.012 wt%. These trends corroborate those observed for the mechanical performance. The release of fluoride ions from the inorganic matrix seems to be favored when reinforced complexes have been formed (Fig. 6). Even in the case of segregation of some CH chains, the release is still much faster than that observed for commercial GIR. The catalytic effect can be related to the entropic gain associated with fluoride release. Generally, the adsorption of a polymer, which causes the displacement of small ions always drives the system to a more favorable ener-
1009
getic situation, because the entropy associated with small ions diffusing into a medium is always larger than that associated with macromolecules free in solution. The first attempt to check the validity of the model proposed in Fig. 6 to explain GIR properties, was measuring zeta potential of powder particles dispersed at pH 1.0 in (i) commercial liquid and (ii) commercial liquid with 0.0044 wt% of CH. The particle number density was in the order of 1 × 1010 particles/mL to avoid strong scattering. The zeta potential value corresponds to the potential at the shear plane of a charged surface and is involved in electrophoresis and other electrokinetic phenomena [30]. In the presence of commercial liquid at pH 1.0 the surfaces are uncharged, as already discussed. Therefore, one would expect zeta potential values close to zero. Upon adding CH to the medium, the zeta potential value should shift to a positive value. Unfortunately, the dispersions presented very high conductivity values, which prevented measurement. The second experimental approach to check off the model suggested in Fig. 6 involved planar Si/SiO2 wafers, simulating the GIR particles. The mean thicknesses of the adsorbed polymeric layers were quantified by ellipsometry and amounted to 0.28 ± 0.05 and 0.5 ± 0.1 nm, in the case of PAA and PAA + CH, respectively. The addition of 0.0044 wt% of CH led to a relative thickness increase of 0.22 nm, indicating that PAA and CH adsorbed onto Si/SiO2 wafers. These findings support the model based on network formation by CH and PAA around the inorganic particles (Fig. 6b).
5.
Conclusions
The addition of 0.0044 wt% of CH to a commercial GIR improved the mechanical performance and catalyzed the release of fluoride ions. This effect was explained based on a model where a polymeric network binds strongly around the inorganic filler. From a practical point of view, this study shows that CH, a biocompatible and low cost additive, can be easily inserted in the traditional formulations bringing about synergic effects.
Acknowledgements The authors acknowledge FAPESP and CNPq for financial support. The authors are grateful to Prof. Jorge Masini and Prof. ˜ Paulo, for allowing us to Paulo T. Sumodjo, University of Sao use the fluoride ion selective electrode Orion 94-09 SC connected to the Ionanalyser (Orion Research Inc., USA).
references
[1] Prosser HJ, Powis DR, Brant P, Wilson AD. Characterisation of glass-ionomer cements. 7. The physical properties of current cements. J Dent 1984;12:231–40. [2] Utracki LA. Polymer alloys and blends: thermodynamics and rheology. New York: Carl Hanser Verlag; 1989. [3] Muzzarelli RAA. Chitin. New York: Pergamon Press; 1977. [4] Roberts GAF. Chitin chemistry. Houndmills: Mac Millan Press; 1992.
1010
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1004–1010
´ [5] Rinaudo M, Domard A. In: Skjak-Braek G, Anthonsen T, Sandford P, editors. Chitin and chitosan: sources, chemistry, biochemistry, physical properties and applications. London: Elsevier Applied Sci.; 1989. p. 1–37. [6] Lehr CM, Bouwstra JA, Schacht EH, Junginger HE. In vitro evaluation of mucoadhesive properties of chitosan and some other natural polymers. Int J Pharm 1992;78:43–8. [7] Needleman IG, Smales FC. In vitro assessment of bioadhesion for periodontal and buccal drug-delivery. Biomaterials 1995;6:617–24. [8] Rillosi M, Buckton G. Modeling mucoadhesion by use of surface-energy terms obtained from the Lewis-acid Lewis base approach. 2. Studies on anionic, cationic, and unionisable polymers. Pharm Res 1995;12:669–75. [9] Shimoda J, Onishi H, Machida Y. Bioadhesive characteristics of chitosan microspheres to the mucosa of rat small intestine. Drug Dev Ind Pharm 2001;27:567–76. [10] Kockisch S, Rees GD, Young SA, Tsibouklis J, Smart JD. Polymeric microspheres for drug delivery to the oral cavity: an in vitro evaluation of mucoadhesive potential. J Pharm Sci 2003;92:1614–23. [11] Sinha VR, Singla AK, Wadhawan S, Kaushik R, Kumria R, Bansal K, et al. Chitosan microspheres as a potential carrier for drugs. Int J Pharm 2004;274:1–33. [12] Vieira DB, Lincopan N, Mamizuka EN, Petri DFS, Carmona-Ribeiro AM. Competitive adsorption of cationic bilayers and chitosan on latex: optimal biocidal action. Langmuir 2003;19:924–32. [13] Linden LANG, Rabek JF, Adamczak E, Morge S, Kaczmarek H, Wrzyszczynski AA. Polymer networks in dentistry. Macromol Symp 1995;93:337–50. [14] Pawlowska E. The assessment of influence of chitosan on the dental pulp in rats. Adv Chitin Sci 1997;2:705–10. ´ [15] Anusavice, Kenneth J. Phillips materiais dentarios. 10th ed. Guanabara Koogan: Rio de Janeiro; 1998. p. 412. [16] Roberts GAF, Domszy JG. Determination of the viscometric constants for chitosan. Int J Biol Macromol 1982;4:374–7. [17] Azzam RMA, Bashara NM. Ellipsometry and polarized light. North Holland Publication: Amsterdam; 1987. [18] Fujimoto J, Petri DFS. Adsorption behavior of carboxymethylcellulose on amino-terminated surfaces. Langmuir 2001;17:56.
[19] Castro LBR, Petri DFS. Assemblies of concanavalin A onto carboxymethylcellulose. J Nanosci Nanotechnol 2005;5:2063–9. [20] Marks LAM, Verbeeck RMH, De Maeyer EAP, Martens LC. Effect of maturation on the fluoride release of resin-modified glass ionomer and polyacid-modified composite resin cements. Biomaterials 2000;21:1373–8. [21] Loidl-Stahlhofen A, Schmitt J, Noller J, Hartmann T, Brodowsky H, Schmitt W, et al. Solid-supported biomolecules on modified silica surfaces—a tool for fast physicochemical characterization and high-throughput screening. Adv Mater 2001;13:1829–34. [22] Singh BP, Menchavez R, Takai C, Fuji M, Takahashi M. Stability of dispersions of colloidal alumina particles in aqueous suspensions. J Coll Interface Sci 2005;291:181–6. [23] Nemec JW, Bauer Jr W. In: Mark HF, editor. Encyclopedia of polymer science and technology, vol. 1. New York: John Wiley & Sons; 1985. p. 211–34. [24] Sailaja GS, Velayudhan S, Sunny MC, Sreenivasan K, Varma HK, Ramesh P. Hydroxyapatite filled chitosan-polyacrylic acid polyelectrolyte complexes. J Mater Sci 2003;38:3653–62. ´ [25] Kim SK, Rha C. In: Skjak-Braek G, Anthonsen T, Sandford P, editors. In chitin and chitosan: sources, chemistry, biochemistry, physical properties and applications. London: Elsevier Applied Sci.; 1989. p. p. 365. ´ [26] Shioya T, Rha C. In: Skjak-Braek G, Anthonsen T, Sandford P, editors. In chitin and chitosan: sources, chemistry, biochemistry, physical properties and applications. London: Elsevier Applied Sci.; 1989. p. p. 627. [27] Marquis RE, Clock SA, Mota-Meira M. Fluoride and organic weak acids as modulators of microbial physiology. FEMS Microbiol Rev 2003;26:493–510. [28] Forsten L. Fluoride release and uptake by glass-ionomers and related materials and its clinical effect. Biomaterials 1998;19:503–8. [29] Phan TN, Nguyen PTM, Abranches J, Marquis RE. Fluoride and organic acids as respiration inhibitors for oral streptococci in acidified environments. Oral Microbiol Immunol 2002;17:119–24. [30] Adamson WA. Physical chemistry of surfaces. 5th ed. Toronto: John Wiley & Sons; 1990.