BioMed Research International

BioMed Research International / 2016 / Article
Special Issue

Orthodontics: Bracket Materials, Adhesives Systems, and Their Bond Strength

View this Special Issue

Research Article | Open Access

Volume 2016 |Article ID 1818939 | 7 pages |

Effect of Removal of Enamel on Rebonding Strength of Resin Composite to Enamel

Academic Editor: Salvatore Sauro
Received19 May 2016
Revised22 Aug 2016
Accepted25 Aug 2016
Published20 Sep 2016


Objective. To examine the effect of removing the surface layer of enamel on the rebonding strength of resin composite. Methods. Teeth in four groups () were etched, a small amount of resin composite was bonded and debonded, then specimens in three groups were ground for different lengths of time (10 s, 20 s, 30 s) to remove an increasing amount of enamel, one group was left untouched. The teeth were bonded again and the bond strengths of 1st and 2nd bonding were compared and analysed against the amount of enamel loss in different groups (7 µm (±2); 12 µm (±1); 16 µm (±3)). Specimens were examined with SEM and by noncontacting optical profilometer. Results. Although results indicated higher rebonding strength with increasing enamel removal ANOVA showed low statistical differences between the groups (). However, values between first bonding and rebonding strengths differed significantly () in the group that was not ground. SEM revealed that enamel-surfaces that were ground after debonding etched well, compared to the surfaces that still contained adhesive remnants. Conclusions. Removal of small amount of enamel refreshed the surface for rebonding. Rebonding strengths without grinding the surface before bonding were lower than bond strength to intact enamel.

1. Introduction

The etching of enamel is the basis of bonding resins and resin composites to enamel, as it increases the surface area and the surface energy of enamel [1]. Etching produces microporosities into which the adhesive can flow and form a structure of tags, and this micromechanical retention leads to a stronger bond [2, 3]. The depth of the resin tag formation is reported to vary between 5 and 50 µm [36]. It has been demonstrated that increasing the length of the tags contributes little to the bond strength [5]. Etching produces a preferential etching pattern depending on the direction of the enamel rods. The two most common patterns are type one where the center of the prisms is dissolved and type two where the prism periphery is dissolved. A third pattern also exists where no prism structure is evident [79].

Phosphoric acid etching with an acid concentration of 10–30% produces the highest bond strengths [10]. These acid concentrations generate maximum enamel dissolution and deposits removable by thorough water-rinsing. In clinical work, concentrations of 30–40% are commonly used. Bond strengths measured by so-called shear bond method to etched enamel with different etching times are as an average 20 MPa, which is considered sufficient and represents the highest bonding values which are available with dental tissues and resin composites [11, 12]. A large variety of dental restorative and active orthodontic treatments are based on enamel bonding. However, sometimes debondings occur and there is need for rebonding of the restoration or orthodontic device. Typically, some amount of enamel is removed for the rebonding procedure.

There can be a difference in the generated etching pattern depending on whether the enamel is intact or instrumented, for example, by grinding. In mature enamel, there is a mineralization gradient, increasing from inner to outer enamel, and there can also be differences in the composition of enamel due to patient’s age, properties of saliva, fluoride concentration of drinking water, and so forth. In a young intact tooth, the outermost layer of enamel is aprismatic, that is, more condensed. Also, salivary calcium ions can mineralize the enamel and fluorine ions can transform the hydroxyapatite to fluoroapatite. Therefore the superficial enamel layer is usually harder than the inner enamel [9, 13, 14]. The etching result of intact enamel depends on the characteristics of the specific area of enamel and may not be uniform over the whole enamel area [15, 16]. In some studies, it has been reported that when etching with 32–35% phosphoric acid, the intact and ground or otherwise instrumented enamel surfaces both develop a porous surface and the bond strength is similar for both surfaces [17].

When rebonding orthodontic brackets, or when recementing loose adhesive restorations, the properties of the underlying layer of previously treated enamel can affect the rebonding strength. The surface of the enamel can contain adhesive remains even after removing all visible adhesive with a scaler [18]. The remaining adhesive can decrease the roughness of the enamel surface and therefore diminish also the rebonding strength [19, 20]. On the other hand, it has also been suggested that the residual adhesive provides a surface for the new adhesive to bond to, chemically or mechanically [21]. It has been reported that reetching does not remove this residual adhesive [19, 20], and therefore a method that does remove the surface enamel should be employed. The bond strength values in rebonding can be inconsistent. Some studies have found the rebonding strength to be lower than the initial bond strength [19, 20, 22, 23], whereas other studies have found the rebonding strength to be higher than the initial bond strength and attribute this to, for example, an increase in enamel roughness caused by the residual adhesive removal [24].

The objective of this study was to examine systematically the effect of removing the surface layer of enamel in rebonding procedures and to analyse microstructural changes at the teeth surfaces.

2. Materials and Methods

The teeth used in the study were extracted molars acquired from the teaching clinic of Institute of Dentistry, University of Turku, Turku, Finland. The roots of the teeth were cut off with a histological saw (Secotom-50, Struers A/S, Ballerup, Denmark) and 40 teeth were horizontally embedded in acrylic resin inside plastic cylinders. A circular area (minimum diameter 3.6 mm) of enamel was exposed and polished with a polishing machine (LaboPol-21, Struers A/S, DK-2750 Ballerup, Denmark), using first a 180-grit (FEPA) and then a 2400-grit SiC paper. The coarser paper was used to remove a bulk of enamel to get a wide enough surface for the bonding, and the finer paper to finish and smooth the rough surface. A control group of intact enamel () was also prepared, where the teeth were half-embedded in acrylic resin.

The enamel substrates were etched using a 32% phosphoric acid etching gel (Table 2) and a small cylindrical amount (height 2-3 mm, Ø 3.6 mm) of Transbond XT orthodontic adhesive (Table 2) was bonded on the enamel. The etching and bonding proceeded as follows: enamel was etched for 15 seconds, rinsed with water, and air-dried according to the manufacturer’s instructions. A cylindrical (height 5 mm, Ø  3.6 mm) mold cut from a plastic tube was placed on the enamel, the adhesive was dispensed into the mold and light cured for 60 s (20 s from above and 20 s from two sides) with a handheld light-curing unit (light-emitting diode, Elipar S10, 3M ESPE, St. Paul, MN, USA), with the intensity of 1834.8 mW/cm2 and the wavelength peak of 455 nm ± 10 nm according to the manufacturer. The plastic mold was removed and the specimens were stored in distilled water in 37°C for four hours. After that the specimens were tested for initial bond strength with a testing machine (LLOYD Instruments, AMETEK Lloyd Instruments Ltd., West Sussex UK) with so-called shear bond strength test with cross-head speed of 1 mm/min. Load-displacement curves were recorded. Testing was made in air at room temperature. Then the specimens were stored overnight in distilled water in 37°C. The teeth in the control group of intact enamel were cleaned with pumice, rinsed with water, air-dried, and then etched, bonded, and tested with the same procedures as the rest of the specimens.

The next day the specimens were divided into four groups (): one group was left untouched and in three groups a small amount of enamel was ground off with an automatic polishing machine (RotoPol-11, Struers A/S, Pederstrupvej 84, DK-2750 Ballerup, Denmark) using a 4000-grit SiC paper (Struers A/SDK-2750 Ballerup, Denmark). The groups underwent grinding with the same settings (4000-grit SiC paper, 150 RPM, 5 N) but with different grinding times: 10 s, 20 s, and 30 s. The groups and their treatments are presented in Table 1. The roughness of the SiC papers was chosen so that with the other grinding factors they removed a desired amount of enamel. Then all the enamel substrates were etched and bonded again with the previously described procedures, stored for four hours, and tested for bond strength with LLOYD testing machine. The amount of enamel that was ground off was determined with additional measuring-samples: five substrates were each ground for 10, 20, and 30 seconds and measured. The samples were measured with a micrometer (Coolant Proof Micrometer, Mitutoyo Corporation, Japan). Every sample was measured five times to avoid error, and an average thickness was calculated for every sample. The average amount of enamel that was removed was 7 µm (±2) for 10 s, 12 µm (±1) for 20 s, and 16 µm (±3) for 30 s. All procedures were performed by the same operator.

Group Treatment

EIntact enamel, etched, and bonded
G0Prepared flat surface, etched, bonded and debonded, then reetched, and rebonded
G1Prepared flat surface, etched, bonded and debonded, ground for 10 s, then reetched, and rebonded
G2Prepared flat surface, etched, bonded and debonded, ground for 20 s, then reetched, and rebonded
G3Prepared flat surface, etched, bonded and debonded, ground for 30 s, then reetched, and rebonded

MaterialsManufacturer Lot no.ContentsWt%

Transbond XT3M Unitek (Monrovia, CA, USA)N568393Silane treated quartz70–80
Silane treated silica<2

Etching gel3M Unitek (Monrovia, CA, USA)576331Water55–65
Phosphoric acid30–40
Amorphous silica5–10

Bis-GMA indicates bisphenol-A-diglycidyl ether dimethacrylate and EBPADMA bisphenol-A-bis(2-hydroxyethyl ether) dimethacrylate.
The specific chemical identity and/or exact percentage (concentration) of this composition has been withheld as a trade secret.

The enamel surfaces were also imaged for visual analysis using a scanning electron microscope (SEM, JSM-5500, Jeol USA, Inc., Peabody, MA). The substrates were gold-sputtered and imaged. A few substrates of interest of different treatments were selected for examining from the prepared groups. One sample was cut and imaged in transverse section, to observe the depth of the resin tags.

Surface roughness of one specimen from each group was determined by optical noncontacting profiler (Contour-GT-K1, Bruker, Billerica, MA, USA) and analysed with a Bruker Vision 64 software (version 5.41, update 4, Bruker, Billerica, MA, USA), to see how it would correspond to the view of the SEM micrographs. Microroughness of the surface was reported as average surface roughness ().

Statistical analysis was performed with SPSS Statistics version 22.0. The data was tested for normality and a one-way ANOVA was performed with a Tukey’s post hoc test. Regression analysis was used to demonstrate correlation between grinding time, that is, removal of enamel layer before rebonding and the rebonding strength.

3. Results

3.1. Bond Strength Measurement

The increase in the grinding time, that is, removal of the enamel layer, showed trend for higher rebonding strengths as demonstrated by the regression analysis (Figure 1). The mean bond strength value for intact enamel was 18.3 MPa (±3.4), and the bond strengths of the 1st bonding were 19.4 MPa (±5.2) for G0 group, 15.7 MPa (±5.2) for G1, 17.5 MPa (±4.9) for G2, and 20.3 MPa (±5.9) for G3. The rebonding strength for the G0 group was 14.3 MPa (±3.6), and the rebonding strengths for the ground substrate groups were 16.1 MPa (±3.3) for G1, 16.3 MPa (±4.8) for G2, and 18.0 MPa (±4.3) for G3 (Table 3). Although there was trend between the increase of rebonding strength and longer grinding time before rebonding (Figure 1), ANOVA did not show statistical differences between groups (). Within the groups, the values between first bonding and rebonding strengths differed significantly in G0 group (). Figure 2 shows typical load-displacement curves demonstrating brittle type of debonding failure for the first bonding and minor ductility for the early stage of loading of the rebonded specimens. Grinding times, amounts of enamel loss, surface roughness parameters, and bond strengths are presented in Table 3. Average surface roughness () for the intact enamel was 0.954 μm, for the etched enamel (group E) 2.307 μm, for group G0 0.301 μm, for group G1 1.945 μm, for group G2 0.857, and for group G3 0.343 (Table 4).


Grinding time0102030
Thickness of removed enamel 7 (±2)12 (±1)16 (±3)
Surface roughness (one specimen)0.3011.9450.8570.343
Bond strength (1st bonding)19.4 (±5.2)15.7 (±5.2)17.5 (±4.9)20.3 (±5.9)
Bond strength (rebonding)14.3 (±3.6)16.1 (±3.3)16.3 (±4.8)18.0 (±4.3)

An asterisk indicates statistical difference () between values.

Intact enamel Etched intact enamel (E)Etched enamel before 1st bonding Reetched after debonding (G0)Ground 10 s, etched (G1)Ground 20 s, etched (G2)Ground 30 s, etched (G3)


3.2. SEM Examination

Figure 4 shows representative SEM images and noncontacting profilometer images of the enamel surface of the study groups. The intact enamel surface exhibited signs of wear, that is, pits and grooves. The ground enamel surface was rather smooth, and grinding the enamel surface after the first bonding resulted in a similar surface as in the initial substrate surface, indicating that the adhesive was removed by the procedure. The reetched substrate surfaces showed clear etching patterns, though with different pattern types. The G0 substrate that was not ground before reetching and rebonding contained remnants of adhesive resin after reetching, whereas the ground substrates showed clearly etched enamel surfaces. It can be seen that the reetching of the not-ground substrate did not remove the remaining adhesive resin layer, but turned it into “adhesive resin-mash.” The most common fracture pattern, presented in almost all the specimens, was adhesive failure. The enamel fractured in three specimens. SEM of the cross section of the adhesive interface showed depth of the resin tags to be 5–10 µm into the enamel (Figure 5).

4. Discussion

This study aimed to examine differences between removal of different amounts of enamel before rebonding. There was a trend indicating that higher rebonding strength was obtained with increasing enamel removal, although statistical analysis did not show strong relationship between the variables. It is likely that bigger number of specimens would have increased the statistical significance between the variables. It is also possible that the testing method to measure the bond strength with predominantly shear type of stress could have also contained microlocations of tensile stress which could have increased the standard deviations and lowered the statistical differences between the groups. The fracture type was brittle fracture as demonstrated by the load-extension curve (Figure 2). Brittleness of the fracture type is due to the cross-linked polymer matrix of the resin composite. Interestingly, the load-extension curve of the rebonded group showed at the early stage of the loading ductility of some degree which may relate to the presence of partly loose remnants of the resin composite and adhesive resin on the enamel surface on to which the rebonding was made. In the loading event, partly loose particles debond with lower level of stress which can be seen in early stage on the load-extension curve.

It was observed that the G0 group showed significantly higher initial bonding strength (19.4 MPa) than corresponding rebonding strength (14.3 MPa) which suggests that remnants of adhesive resin harm etching the surface for rebonding. Therefore, it is beneficial to reveal a fresh enamel surface before rebonding by grinding the surface. The intact enamel surface etched relatively well, despite presence of aprismatic layer or hypermineralization of the enamel surface.

The results of this study revealed that the differences between rebonding directly on the debonded surface or on a slightly ground surface were minor. If the rebonding strength is desired to be as high as the initial bonding strength, then removal of enamel can be recommended. A thickness of a maximum 17 µm of enamel was removed during the grinding process resulting in a similar bond strength as with initial bonding of intact enamel. According to the literature, residual adhesive resin clean-up removes approximately 5–30 µm of enamel [2527], depending on the grinding system that is used. Even a loss of 40–60 µm of enamel has been reported to occur in the entire debonding and clean-up procedure [6, 28]. In the present study, a 7 µm grinding created a reetchable enamel surface which was also confirmed by the surface roughness measurement for and . Interestingly, surface roughness parameters lowered when the enamel was ground further.

Grinding of enamel can be made clinically by rotating silicon carbide finishing bur and damage to the enamel is only minor [16, 29, 30]. On the other hand, there are reports stating that even with a clean-up with a silicon carbide finishing bur small remnants of adhesive can remain on enamel [19, 20]. Although it was found in the present and other studies [29] that the reetching produced a regular etching pattern to ground enamel surface after debonding, it has also been suggested that the reetching step could be omitted, to avoid risk of enamel fracture due to rebonding strengths that are too high [31].

Clinically, the desired bond strength is different in different areas of dentistry: in restorative dentistry the highest possible bond strength is desirable, whereas in orthodontics the bond must be strong enough to keep the appliances in place for the duration of the treatment but at the same time allow easy detachment of the appliance once the treatment is over. If the bond strength is too high, it can result in enamel damage or discomfort for the patient at the removal. Further research is needed to investigate the amount of enamel loss in debonding and rebonding process with brackets of different types and materials.

5. Conclusions

(i)Rebonding strengths without grinding the surface before bonding were lower than bond strength to intact enamel.(ii)Removal of small amount of enamel refreshed the surface for rebonding.

Competing Interests

The authors declare that they have no competing interests.


  1. W. O'Brien, “Surface phenomena and adhesion to tooth structure,” in Dental Materials and Their Selection, W. O'Brien, Ed., chapter 5, p. 68, Quintessence Publishing, 4th edition, 2008. View at: Google Scholar
  2. A. J. Gwinnett and A. Matsui, “A study of enamel adhesives. The physical relationship between enamel and adhesive,” Archives of Oral Biology, vol. 12, no. 12, pp. 1615–1620, 1967. View at: Publisher Site | Google Scholar
  3. M. G. Buonocore, A. Matsui, and A. J. Gwinnett, “Penetration of resin dental materials into enamel surfaces with reference to bonding,” Archives of Oral Biology, vol. 13, no. 1, pp. 61–70, 1968. View at: Publisher Site | Google Scholar
  4. A. J. Gwinnett, “Histologic changes in human enamel following treatment with acidic adhesive conditioning agents,” Archives of Oral Biology, vol. 16, no. 7, pp. 731–738, 1971. View at: Publisher Site | Google Scholar
  5. M. J. Shinchi, K. Soma, and N. Nakabayashi, “The effect of phosphoric acid concentration on resin tag length and bond strength of a photo-cured resin to acid-etched enamel,” Dental Materials, vol. 16, no. 5, pp. 324–329, 2000. View at: Publisher Site | Google Scholar
  6. D. A. Fitzpatrick and D. C. Way, “The effects of wear, acid etching, and bond removal on human enamel,” American Journal of Orthodontics, vol. 72, no. 6, pp. 671–681, 1977. View at: Publisher Site | Google Scholar
  7. P. Lambrechts, B. van Meerbeek, J. Perdigão, and G. Vanherle, “Adhesives: dos and don'ts,” in Adhesion—The Silent Revolution in Dentistry, J.-F. Roulet and M. Degrange, Eds., chapter 4, p. 46, Quintessence Publishing, 2000. View at: Google Scholar
  8. L. M. Silverstone, C. A. Saxton, I. L. Dogon, and O. Fejerskov, “Variation in the pattern of acid etching of human dental enamel examined by scanning electron microscopy,” Caries Research, vol. 9, no. 5, pp. 373–387, 1975. View at: Publisher Site | Google Scholar
  9. G. C. Lopes, D. G. Thys, P. Klaus, G. M. S. Oliveira, and N. Widmer, “Enamel acid etching: a review,” Compendium of Continuing Education in Dentistry, vol. 28, no. 1, pp. 18–42, 2007. View at: Google Scholar
  10. Soetopo, D. R. Beech, and J. L. Hardwick, “Mechanism of adhesion of polymers to acid-etched enamel,” Journal of Oral Rehabilitation, vol. 5, no. 1, pp. 69–79, 1978. View at: Publisher Site | Google Scholar
  11. R. O. Gilpatrick, J. A. Ross, and R. J. Simonsen, “Resin-to-enamel bond strengths with various etching times,” Quintessence International, vol. 22, no. 1, pp. 47–49, 1991. View at: Google Scholar
  12. W. W. Barkmeier, S. E. Shaffer, and A. J. Gwinnett, “Effects of 15 vs 60 second enamel acid conditioning on adhesion and morphology,” Operative Dentistry, vol. 11, no. 3, pp. 111–116, 1986. View at: Google Scholar
  13. A. J. Gwinnett, “Structure and composition of enamel,” Operative Dentistry, supplement 5, pp. 10–17, 1992. View at: Google Scholar
  14. A. J. Gwinnett, “The ultrastructure of the ‘prismless’ enamel of permanent human teeth,” Archives of Oral Biology, vol. 12, no. 3, pp. 381–387, 1967. View at: Publisher Site | Google Scholar
  15. M. S. Shinohara, M. T. De Oliveira, V. Di Hipólito, M. Giannini, and M. F. De Goes, “SEM analysis of the acid-etched enamel patterns promoted by acidic monomers and phosphoric acids,” Journal of Applied Oral Science, vol. 14, no. 6, pp. 427–435, 2006. View at: Publisher Site | Google Scholar
  16. M. Aksu and I. Kocadereli, “Influence of two different bracket base cleaning procedures on shear bond strength reliability,” Journal of Contemporary Dental Practice, vol. 14, no. 2, pp. 250–254, 2013. View at: Publisher Site | Google Scholar
  17. N. Kanemura, H. Sano, and J. Tagami, “Tensile bond strength to and SEM evaluation of ground and intact enamel surfaces,” Journal of Dentistry, vol. 27, no. 7, pp. 523–530, 1999. View at: Publisher Site | Google Scholar
  18. M. A. Montasser, J. L. Drummond, J. R. Roth, L. Al-Turki, and C. A. Evans, “Rebonding of orthodontic brackets. Part II, an XPS and SEM study,” The Angle Orthodontist, vol. 78, no. 3, pp. 537–544, 2008. View at: Publisher Site | Google Scholar
  19. S. E. Bishara, L. VonWald, J. F. Laffoon, and J. J. Warren, “The effect of repeated bonding on the shear bond strength of a composite resin orthodontic adhesive,” The Angle Orthodontist, vol. 70, no. 6, pp. 435–441, 2000. View at: Google Scholar
  20. S. E. Bishara, J. F. Laffoon, L. VonWald, and J. J. Warren, “The effect of repeated bonding on the shear bond strength of different orthodontic adhesives,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 121, no. 5, pp. 521–525, 2002. View at: Publisher Site | Google Scholar
  21. M. A. Montasser, J. L. Drummond, and C. A. Evans, “Rebonding of orthodontic brackets. Part I, a laboratory and clinical study,” Angle Orthodontist, vol. 78, no. 3, pp. 531–536, 2008. View at: Publisher Site | Google Scholar
  22. M. Oshagh, H. R. Pakshir, H. Z. Najafi, M. M. Naseri, N. I. Nasrabadi, and S. Torkan, “Comparison of the shear bond strength of orthodontic brackets in bonding and rebonding: preparation with laser versus conventional acid etch technique,” Photomedicine and Laser Surgery, vol. 31, no. 8, pp. 360–364, 2013. View at: Publisher Site | Google Scholar
  23. A. P. Kinch, H. Taylor, R. Warltler, R. G. Oliver, and R. G. Newcombe, “A clinical trial comparing the failure rates of directly bonded brackets using etch times of 15 or 60 seconds,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 94, no. 6, pp. 476–483, 1988. View at: Publisher Site | Google Scholar
  24. N. Eminkahyagil, A. Arman, A. Çetinşahin, and E. Karabulut, “Effect of resin-removal methods on enamel and shear bond strength of rebonded brackets,” Angle Orthodontist, vol. 76, no. 2, pp. 314–321, 2006. View at: Google Scholar
  25. S. N. Suliman, T. M. Trojan, D. Tantbirojn, and A. Versluis, “Enamel loss following ceramic bracket debonding: a quantitative analysis in vitro,” The Angle Orthodontist, vol. 85, no. 4, pp. 651–656, 2015. View at: Publisher Site | Google Scholar
  26. R. E. Thompson and D. C. Way, “Enamel loss due to prophylaxis and multiple bonding/debonding of orthodontic attachments,” American Journal of Orthodontics, vol. 79, no. 3, pp. 282–295, 1981. View at: Publisher Site | Google Scholar
  27. B. Øgaard and M. Fjeld, “The enamel surface and bonding in orthodontics,” Seminars in Orthodontics, vol. 16, no. 1, pp. 37–48, 2010. View at: Publisher Site | Google Scholar
  28. C. R. L. Brown and D. C. Way, “Enamel loss during orthodontic bonding and subsequent loss during removal of filled and unfilled adhesives,” American Journal of Orthodontics, vol. 74, no. 6, pp. 663–671, 1978. View at: Publisher Site | Google Scholar
  29. B. Mui, P. E. Rossouw, and G. V. Kulkarni, “Optimization of a procedure for rebonding dislodged orthodontic brackets,” Angle Orthodontist, vol. 69, no. 3, pp. 276–281, 1999. View at: Google Scholar
  30. H. B. Pont, M. Özcan, B. Bagis, and Y. Ren, “Loss of surface enamel after bracket debonding: an in-vivo and ex-vivo evaluation,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 138, no. 4, pp. 387.e1–387.e9, 2010. View at: Publisher Site | Google Scholar
  31. Q.-F. Zhang, H. Yao, Z.-Y. Li, L. Jin, and H.-M. Wang, “Optimal enamel conditioning strategy for rebonding orthodontic brackets: a laboratory study,” International Journal of Clinical and Experimental Medicine, vol. 7, no. 9, pp. 2705–2711, 2014. View at: Google Scholar

Copyright © 2016 L. Kilponen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

947 Views | 325 Downloads | 3 Citations
 PDF  Download Citation  Citation
 Download other formatsMore
 Order printed copiesOrder
 Sign up for content alertsSign up