Materials Letters 65 (2011) 22542256

Contents lists available at ScienceDirect

Materials Letters
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a t l e t

Effects of cola drinks on the morphology and elastic modulus of dentin

Physics University of the Witwatersrand, PO WITS 2050, Johannesburg, South Africa Lab of Mechanics and Materials, Aristotle University of Thessaloniki 54124, Greece Physics, Michigan Technological University, Houghton, MI 49931, USA

a r t i c l e

i n f o

a b s t r a c t
The mechanical and morphological degradation that human dentin undergoes after prolonged exposure in cola drinks was examined using nanoindentation and AFM. The elastic modulus of the dentin prior cola exposure was 18.28 1.7 GPa, while after exposure it had decreased to 4.25 1.4 GPa on the surface that was in direct contact with the drink, and to 13.38 2.9 GPa on the surface that was in indirect contact with the drink. AFM documented the corrosion effects of the cola through topography images and histograms of the RMS roughness. 2011 Elsevier B.V. All rights reserved.

Article history: Received 14 December 2010 Accepted 15 April 2011 Available online 20 April 2011 Keywords: Dentin Elastic modulus Phosphoric acid Cola

1. Introduction Understanding the effect that fruit juices have on teeth has become of particular interest during the past years, since complete remineralization by the saliva is not possible after exposure to citric acid. In [1] prolometry was used to capture the mass loss that took place after enamel immersion in orange juice, and atomic force microscopy (AFM) images documented that the surface became rougher, and hence more prone to bacteria adhesion and plaque formation [2], while nanoindentation experiments recorded a 85% hardness decrease after juice exposure [3]. However, it is not only fruit juices that are highly corrosive. Soft drinks, particularly those containing phosphoric acid, are also damaging, but have not been studied to a similar extent. Initial in situ studies concerning the effect that colas have on teeth were performed in [4,5]. Prolometry and indentation were used to indicate mass reduction and the percentage reduction of the surface microhardness. However, hardness or elastic modulus values were not given and microscopy was not performed to illustrate the morphological damage. The present study is concerned with the degradation that the elastic modulus of dentin undergoes after prolonged exposure in a cola drink, whose main acidic ingredient is phosphoric acid.1 Dentin is a highly mineralized tissue, that consists of approximately 50% mineral, 30% organic components (mostly type I collagen) and 20%
Corresponding author at: Physics, Michigan Technological University, Houghton, MI 49931, USA. Tel.: + 30 2310 995921. E-mail address: kaifanti@[Link] (K.E. Aifantis).
1 Phosphoric acid (H3PO4) has a pH of 1.1 and is a main ingredient in colas. It is stronger than citric acid (5% citric acid has a pH of 1.87), which is also used in carbonated sodas.

uids. Its distinct microstructure is characterized by tubules (12 m in diameter) that run from the dentinenamel junction towards the pulp [6,7]. Micropores and channels created at the dentinenamel junction [8], in addition, to microcracks on the enamel surface allow for acidic solutions to permeate into the dentin and corrode it, especially since remineralization is more difcult, as saliva do not reach the dentin easily. It is anticipated that once phosphoric acid penetrates into the dentin, it is not removed easily, but trapped there. Hence the dentin sample was exposed to the cola for 16 h. The mechanical and morphological degradation were captured through nanoindentation and AFM.

2. Materials and methods The premolar tooth tested was obtained from an adult incision and was sliced using a microtome along the labiolingual plane. The resulting specimen had a thickness of 2 mm and an area of 1 cm 1 cm; as they were cross sectional they consisted mainly of dentin. Both surfaces (upper and bottom) of the sample were polished using abrasive paper from 800 to 1200 grit, followed by 0.25 m diamond paste polishing. AFM and nanoindentation were performed prior to immersing the sample in the cola. AFM was carried out using the Veeco Multimode, under tapping mode, with a Si3N4 tip with a tip radius of 12 nm. Following AFM, 50 nanoindentations were performed using a CSM instrument with a Berkovich diamond tip (tip radius was 150 nm). The indentations were load controlled, and a linear loading trapezoidal sequence was used (maximum load was 200 mN, and a hold time of 20 s was allowed). The spacing between the indents was 50 m, so as to ensure that each indentation was not inuenced by that

0167-577X/$ see front matter 2011 Elsevier B.V. All rights reserved. doi:10.1016/[Link].2011.04.046

S. Shrivastava, K.E. Aifantis / Materials Letters 65 (2011) 22542256

2255

Fig. 1. 3D images of (a) as polished dentin and (b) bottom surface after cola immersion.

Fig. 2. (a) Line prole across dentin tubules. (b) Normalized roughness histogram (color image online).

preceding it. It should be noted that the experiment was repeated on two samples obtained from the same premolar tooth. Once the microstructure and elastic modulus of the as-polished surface were recorded, the samples were placed in a beaker, which was then lled with 30 ml of a freshly opened cola drink can. The entire sample was covered by the drink; particularly, the upper surface was directly exposed to the cola, while the bottom surface was indirectly exposed, as it was in contact with the beaker. After 16 h the samples were rinsed with methanol by agitation and placed directly under the nanoindenter without removing the lubricant that the drink had formed on the upper surface. Fifty indentations were performed using the same parameters as before immersion on both surfaces of the samples. Once the elastic modulus was recorded, AFM was performed on the bottom surface. Thereafter, the samples were stored in a dry environment to allow for evaporation of the cola. After 1 week 50 nanoindentations were again performed on both surfaces; however, on the upper surface the lubricant that had formed by the cola was removed with a methanol soaked cotton swab. Ultrasonication was not used to remove this lubricant, as it introduces visiblecracking. 3. Results and discussion 3.1. Microstructure of dentin before and after cola immersion In Fig. 1a the microstructure of the as polished dentin surfaces is illustrated (both top and bottom surfaces had the same microstructure at the start of the experiment). In order to have a visual comparison of the effect that the cola had, AFM was performed on the bottom surface after the 16 h immersion (Fig. 1b). The microtubular (porous) structure

of the dentin is seen and the RMS roughness was found to be 97 13.5 nm prior to cola immersion, while it was 80 4 nm2 after immersion. By qualitatively comparing Fig. 1a and b it can be seen that signicant corrosion resulted from the cola immersion. This corrosion effect is better illustrated by quantitatively comparing the line proles across a dentin microtubule. In Fig. 2a, therefore, the average line proles across the tubules obtained from Fig. 1a and b are plotted together, along with that obtained 1 week after the tooth was exposed to cola. It is seen that after cola exposure both the depth of the tubule and the slope of the line prole decreased. This indicates that in addition to corroding the surface of the dentin, the lip and inner wall of the tubules were also corroded, during cola exposure, giving a granularlike structure; as a result the tubule walls became slopped and the corroded sub-micron particles lled in the inner tubules giving them a lower height. Fig. 2b compares the normalized roughness histogram for topography images with bin (0255) for three cases: as-polished, after immersion, and 1-week after immersion in the cola. The negative skew of the distribution for the after immersion cases produces a lower roughness value, while the distribution for the as-polished is almost Gaussian. The negative skewness could have occurred only due to corrosion. The as polished surface appears to have a higher RMS roughness since the tubule depth was greater before corrosion. It should be noted that Fig. 2 was derived from the AFM images of the bottom surface after immersion. AFM could not be performed on the upper surface as it had undergone such corrosion that the AFM tip broke due to the roughness.

RMS roughness as measured using the software WSxM v.5.0.

2256

S. Shrivastava, K.E. Aifantis / Materials Letters 65 (2011) 22542256

dry in a desiccator. The elastic modulus of the top surface appears to have increased, due to the fact that the cola lubricant layer was removed prior to this second run of indentations, whereas in the rst set of measurements, right after immersion, it was still present. Hence, it can be seen that the lubricant layer is soft and protects the underlying dentin. After 1 week the modulus of the bottom surface had further decreased due to the fact that phosphoric acid nanoparticles attach on the dentin and, therefore, as the cola evaporated high stresses developed on the dentin tubules that damaged them, and could even fracture them, reducing hence the elastic modulus even more. A more detailed analysis on stress developments during the dehydration process can be found in [10]. 4. Conclusions The present study examined the effect that cola drinks have on dentin. AFM images illustrated the morphological degradation of the dentin surface and microtubules after prolonged cola exposure, while nanoindentation showed that the elastic modulus was reduced from 18.28 1.7 GPa to 4.25 1.4 GPa on the surface that was in direct contact with the drink, and to 13.38 2.9 GPa on the surface that was in indirect contact. Furthermore it was shown that a soft protective lubricant layer forms after prolonged cola exposure, while during evaporation of the cola further mechanical degradation takes place. Acknowledgements
9.19 1.1 7.46 2.1

Fig. 3. Representative load vs. depth nanoindentation curves.

Table 1 Elastic modulus before and after cola immersion. Before immersion After immersion bottom surface 13.38 2.9 After immersion upper surface 4.25 1.4 1 week after immersion bottom surface 1 week after immersion upper surface

E (GPa)

18.28 1.7

The authors are grateful to KEA's European Research Council Starting Grant 211166 for funding this work and to Dr. E. Karnezi for sample preparation and helpful discussions. 3.2. Elastic modulus In Fig. 3 typical load versus depth plots are shown from the indentation tests. The CSM software automatically performed the OliverPharr method [9] to compute the elastic modulus of the indented dentin area for each set of experiments and the results are summarized in Table 1. The maximum indentation depth was several microns (Fig. 3) and therefore the nanoscale roughness (Fig. 1b) did not affect the measurements. The initial elastic modulus of the as-polished dentin (18.281.7 GPa) is in exact agreement with previous studies [6,7]. After cola exposure the elastic modulus decrease of the top surface was about 76%, while that of the bottom surface was 27%, which indicates that the porosity of dentin is not such that allows for fast diffusion of corrosive uids. This decrease in the elastic modulus resulted from the corrosion that took place and was indicated in the AFM images (Fig. 2) which show that the tubule diameter increased, while their depth decreased after immersion. Such degradation in the elastic modulus of dentin after cola immersion is consistent with previous studies [4,5] that have reported a decrease in the microhardness of enamel and dentin after exposure to cola drinks; elastic modulus values, however, were not stated [4,5]. It is interesting to note the differences in the elastic modulus obtained 1 week after the sample was taken out of the cola and left to References
[1] Barbour ME, Rees JS. The laboratory assessment of enamel erosion: a review. J Dent 2004;32:591602. [2] Quirynen M. The clinical meaning of the surface roughness and the surface free energy of intra-oral hard substrata on the microbiology of the supra- and subgingival plaque: results of in vitro and in vivo experiments. J Dent 1994;22 (Supplement 1):S136. [3] Ren YF, Amin A, Malmstrom H. Effects of tooth whitening and orange juice on surface properties of dental enamel. J Dent 2009;37:42431. [4] Sales-Peres SHC, Pessan JP, Buzalaf MAR. Effect of an iron mouthrinse on enamel and dentine erosion subjected or not to abrasion: an in situ/ex vivo study. Arch Oral Biol 2007;52:12832. [5] Rios D, Honorio HM, Francisconi LF, Magalhaes AC, de Andrade Moreira MA, Buzalaf MAR. In situ effect of an erosive challenge on different restorative materials and on enamel adjacent to these materials. J Dent 2008;36:1528. [6] Marshall GW, Marshall SJ, Kinney JH, Balooch M. The dentin substrate: structure and properties related to bonding. J Dent 1997;25:44158. [7] Kinney JH, Balooch M, Marshall SJ, Marshall GW, Weihs TP. Hardness and Young's modulus of human peritubular and intertubular dentine. Arch Oral Biol 1996;41: 913. [8] Imbeni V, Kruzic JJ, Marshall GW, Marshall SJ, Ritchie RO. The dentinenamel junction and the fracture of human teeth. Nat Mater 2005:22932. [9] Oliver WC, Pharr GM. An Improved Technique for Determining Hardness and Elastic-Modulus using Load and Displacement Sensing Indentation Experiments. J Mat Res 1992;7:156483. [10] Zhang D, Mao S, Lu C, Romberg E, Arola D. Dehydration and the Dynamic Dimensional Changes within Dentin and Enamel. Dent Mater 2009;25:93745.