SciELO - Scientific Electronic Library Online

Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista de Odontologia da Universidade de São Paulo

Print version ISSN 0103-0663

Rev Odontol Univ São Paulo vol. 11 no. 3 São Paulo July/Sept. 1997 




Flávio Fernando DEMARCO*
Miriam Lacalle TURBINO**
Edmir MATSON***



DEMARCO, F. F. et al. Cohesive strength of dentin. Rev Odontol Univ São Paulo, v. 11, n. 3, p. 189-194, jul./set. 1997.

The bond strength of dentin adhesives to dentin has increased after each generation. Although dentin substratum is part of the bonding process, little importance has been given to measure dentin cohesive strength. The aim of this study was to evaluate the cohesive strength of dentin in human canines. Seventeen non carious canines were selected. All of them had been extracted for more than one year. The teeth were ground until dentin square samples with approximately 2 X 2 mm were obtained. They were embedded in acrylic resin and subjected to shear stress, in a Wolpert Machine, at a crosshead speed of 0.5 mm/min. The mean cohesive strength of dentin in shear mode was 33.95 (± 9.72) MPa. The fracture surfaces were observed under a X40 magnification. A finite element analysis was performed to observe the stress distribution as related to the shear test. The failure pattern was compatible with the shear test and also with the stress distribution in the finite element analysis.

UNITERMS: Dentin; Dentin adhesives; Bond strength.




An efficient procedure for the adhesion between resin and enamel came out when acid etching of enamel was introduced in Dentistry. Nevertheless, adhesion to dentin is more difficult to obtain, mainly due to the differences in structure and composition of this substratum16,25.

The development achieved in dentin bonding agents, with modifications in chemical composition and mechanism of action, has improved the bond strength between dentin and composite resin1,8. To fully understand dentin properties is an important objective of Adhesive Dentistry and, to this effect, the cohesive strength of dentin has been studied10,14,21,24,28.

In recent reports, GWINNETT10 (1994); SANO et al.21 (1994); WATANABE et al.28 (1996), using different methods, found different values of dentin cohesive strength. In addition, some recent studies3,22, using fourth generation dentin bonding agents in microtensile tests, have found bond strength values higher than some values measured as cohesive strength of dentin10,24. The concern about bond strength overcoming dentin cohesive strength is present in the literature10.

Several aspects could influence the mechanical properties of dentin. Tooth position18, the time elapsed after extraction4,5, and the testing method6,9,27 are some of these aspects. Tubule orientation and intratooth location have been reported to influence dentin shear strength29.

The finite element analysis is becoming a useful tool in dental research since it is well suited for modeling the stress pattern in complex composite structures such as teeth20,30. It has also been useful to understand bond strength data6,28.

The purpose of this study was to evaluate the cohesive strength of dentin, using long term extracted human canines. Finite Element Analysis (FEA) was also used to investigate the stress distribution of the testing mode. The failure pattern from fractured samples was observed and compared with the stress analysis.



Sample preparation for shear test

In this study, seventeen non carious human canines, were used. The teeth had been extracted for more than one year, from old patients because of periodontal problems. After extraction they were kept in a neutral-buffered formalin until one week before the test, when they stored in distilled water.

Before the preparation, radiographs were taken from all teeth to measure the distance from the pulp chamber to external surface. The preparation of the specimens were performed with diamond burs in high speed under air/water cooled spray. Dentin square preparations (2 mm X 2 mm) were obtained (Figure 1). Each tooth was measured with a paquimeter for calculating the respective area.


Image672.GIF (8881 bytes)

FIGURE 1 - Tooth after the preparation of the sample.


The samples were embedded in acrylic resin in a polymethyl methacrilate resin (PVC) matrix. The base of preparation was in the same level of the matrix edges. New radiographs were taken. If the sample had less than 0.5 mm of remaining dentin to pulp chamber it would be removed from the study.

Shear test

The prepared samples were stored in distilled water for 24 hours at 37ºC and then subjected to shear test using a knife edge applied as close as possible to the base of the preparation (Figure 2) in a Universal Wolpert Machine at a crosshead speed of 0,5 mm/min, according to a previously reported methodology7. The force recorded at the moment of failure was divided by the original area and the results expressed in MPa. After the test, the teeth were removed from acrylic resin matrix and sectioned through the center of the samples to observe the fracture pattern. The fracture surface of each sample was examined under 40X magnification.


Image673.GIF (7916 bytes)

FIGURE 2 - Sample mounted in a jig. Sheaer test is being performed.


Finite element stress analysis (FEA)

The stress distributions for the shear test was determined from a two-dimensional plane-strain computer model of a central section with dimensions identical to those of the experimental samples (ANSYS 5.2). The nodes at the sides and bottom of the base material were constrained in both the x and y directions. Poisson's ratio (m) and elastic modulus (E) for dentin was m = 0.33 and E = 18.6 GPa. A shear load of 10 N was applied at the interface of the base and sample.

The stress distributions for the tensile test was performed using the same parameters used for the shear test. The same tensile load was applied in both sides of the interface.



Resin adhesion to dentin has improved with the development of new generations of dentin bonding agents1,8. In some studies15,19, failures in dentin have been reported. In view of these results, the study of cohesive strength of dentin has become an important issue10,14,21,24,28.

In this study, the mean value of dentin cohesive strength for human canines was 33.95 (± 9.72) MPa, as indicated in the Table 1.


Image674.GIF (4625 bytes)


GWINNETT10 (1994) tested the cohesive strength of dentin using a shear test. The mean cohesive strength found in his study was 36.18 (±   6.81) MPa. SANO et al.22 (1994) tested mineralized and demineralized bovine and human dentin. Human coronal mineralized dentin presented a mean ultimate tensile strength of 104 (±  16.3) MPa. WATANABE et al.29 (1996) studied the shear strength of dentin related with tubule orientation and intratooth location. The values were different according to different locations. The lowest mean value was 53.5 (±  9.5) MPa in a parallel tubule orientation and from center position dentin. The highest value was 91.8 (±  12.7) MPa from cusp area dentin in samples oriented with tubules along the long axis of the specimen and with the shear stress applied perpendicularly to the tubule direction.

To compare the results from tests using different methodologies is quite hard26. Nevertheless, several differences could be enrolled in this lack of agreement. The differences between these results and our results can be explained by the design of the testing methods employed.

The samples tested in studies by SANO et al.21 (1994) and WATANABE et al.28 (1996) were smaller than our samples, which had a larger area of dentin. In a smaller cross-sectional area, if the natural defects or stress-concentrating voids are uniformly but relatively sparsely distributed in dentin, then these samples may have more uniform stress distribution and thus produce higher tensile strengths21.

On the other hand, GWINNETT10 (1994) used larger samples than ours, thus a higher cohesive strength could be expected in our results. However, both studies had similar results. In his study, GWINNETT10 (1994) worked with samples from recently extracted human third molars, whereas we tested the dentin cohesive strength of samples from human canines that had been extracted more than one year. Dentin following extraction becomes more brittle than that of vital teeth, meaning that a physical, and probably also a chemical, change takes place in the tissue after loss of the pulp4. The elapsed time after extraction and the environment in which the teeth are placed may influence results from laboratory tests12. Additionally, the teeth used in our study were from older patients. As teeth age, they continue to calcify, resulting in the dentinal tubules becoming narrower2. PASHLEY et al.18 (1993) demonstrated that different tooth groups, for example molars and cuspids, can produce different results in bond strength. As we selected canines and GWINNETT10 (1994) selected third molars, the data could be influenced because of the different characteristics of each dentin18. Finally, our samples were square and his samples were cylindrical, being tested with different methods of load application in shear test. Geometric design of the experiment and the nature of the application of forces, according to VAN NOORT et al.27 (1991), may interfere in the results from laboratory tests in adhesive dentistry.

GWINNETT11 (1994) has used the test of dentin cohesive strength as a control in relation to bond strength test of dentin bonding agents. The author stated that the rising of bond strength near to values of dentin cohesive strength increases the possibilities of cohesive failures in dentin.

Cohesive failures of dentin have also been found in laboratory tests, in spite of low values of bond strength19. In other studies, where the authors found values of bond strength higher than 50 MPa3,22, only adhesive failures occurred. Up to date there are some doubts regarding the occurrence of cohesive failures in in vitro adhesion tests.

KATO; NAKABAYASHI13 (1995) found that after acid etching of dentin the exposed collagen, that was not completely impregnated by dentin adhesives, was easily weakened by hydrolysis and, in the long run, the durability of adhesion decreased. The authors have also found that cohesive failures in tensile test happened in the exposed collagen area, that was not completely reinforced by diffusion of the adhesive, under SEM examination.

On the other hand, SANO et al.23 (1995) demonstrated that demineralized dentin, after being infiltrated by adhesive resins, can restore and even exceed the ultimate tensile strength of mineralized dentin. In another study, SANO et al.22 (1994) offer an alternative explanation on the cohesive failures found in adhesive tests. They said that cohesive failures were due to the non uniform stress distribution during the performing of the test. This is reinforced by the results from their study, when they compared the relationship between surface area of adhesion and tensile bond strength using a micro-tensile bond test. They disclosed that smaller surface areas were associated with higher tensile bond strengths and produced adhesive failures, whereas larger surface areas showed lower tensile bond strengths and the presence of cohesive failures in dentin22.

In our study, when the fracture surface was observed at X40 magnification it was possible to see the removal of dentin from the loaded size of the sample base. The fracture pattern of the samples in shear is in accordance with the stress distribution that could be observed in the FEA (Figure 3). Like the results found by DELLA BONNA; VAN NOORT6 (1995) in their composite samples, the stress distribution from the shear test in our dentin samples showed a higher stress in the point of load application and also in the dentin base of the sample, which was the beginning point of fracture. As stated by those authors6 the cohesive failure pattern found in shear bond strength tests can be dependent on the test geometry (shear) instead of on dentin properties. Based on these statements, additional studies could be necessary to investigate the cohesive strength of dentin using a tensile test model. More homogeneous stress distribution was found than in the shear test, when a tensile test with the same sample size was performed in FEA, as can be seen on Figure 4.


Image675.GIF (32892 bytes)

FIGURE 3 - Cohesive strength of dentin. Stress distribuition from the shear test in Finite Element Analysis.



Image676.GIF (39287 bytes)

FIGURE 4 - Cohesive strength of dentin. Stress distribuition from the test in Finite Element Analysis.


The methodology developed in our study could be used, in a similar manner as used by GWINNETT11 (1994), as a control when shear bond tests are being performed, using similar teeth group, sample size and load application.



Based on the utilized methodology, the conclusions are the following:

1. The cohesive strength of dentin found in this study, with human canines from old patients, was 33.95 (± 9.72) MPa.

2. The Finite Element Analysis showed a stress distribution compatible with the sample fracture pattern.



The authors would like to thank Dr. Richard Van Noort, University of Sheffield, for helping the preparation of this paper. Also, we would like to thank Aldo Francisco Gomes, School of Dentistry, University of São Paulo, for the graphs presented in this study.



DEMARCO, F. F. et al. Resistência coesiva da dentina. Rev Odontol Univ São Paulo, v. 11, n. 3, p. 189-193, jul./set. 1997.

A resistência de união dos adesivos dentinários tem sido aumentada com o desenvolvimento de cada nova geração. Pouca importância tem sido dada à resistência coesiva da dentina. A proposta deste estudo foi avaliar a resistência coesiva da dentina. Dezessete caninos humanos hígidos, os quais tinham sido extraídos há mais de um ano, foram usados. Os dentes foram desgastados até a obtenção de corpos-de-prova em dentina, de formato quadrangular, com tamanho aproximado de 2 X 2 mm. Os dentes foram incluídos em resina acrílica e, então, submetidos ao teste de cisalhamento em uma máquina de ensaios universais Wolpert, com uma velocidade de 0,5 mm/min. A resistência coesiva média da dentina no teste de cisalhamento foi de 33,95 (±   9,72) MPa. O tipo de fratura foi analisado com um aumento de 40X. Foi realizada uma análise com elemento finito, para observar a distribuição do estresse relacionada com o teste de cisalhamento. O padrão de fratura encontrado foi compatível com o tipo de teste executado e com a distribuição do estresse obtida a partir da análise de elemento finito.

UNITERMOS: Dentina; Adesivos dentários; Resistência de união.




1. BURKE, F. J. T.; McCAUGHEY, A. D. The four generations of dentin bonding.  Am J Dent, v. 8, n. 2, p. 88-92, Apr. 1995.        [ Links ]

2. BURROW, M. F. et al. The influence of age and depth of dentin on bonding. Dent Mater, v. 10, n. 3, p. 241-246, July 1994.        [ Links ]

3. CARVALHO, R. M. et al. Determinação da resistência adesiva à dentina através de um dispositivo de micro-tração. Rev Fac Odontol Bauru, v. 2, n. 3, p. 77-82, jul./set. 1994.        [ Links ]

4. CAUSTON, B. E.; JOHNSON, N. W. Changes in the dentine of human teeth following extraction and their implication for in vitro studies of adhesion to tooth substance. Arch Oral Biol, v. 24, n. 3, p. 229-232, May 1979.        [ Links ]

5. COOLEY, R. L.; DODGE, W. W. Bond strength of three dentinal adhesives on recently extracted versus aged teeth.  Quintessence Int, v. 20, n. 7, p. 513-516, July 1989.        [ Links ]

6. DELLA BONA, A.; VAN NOORT, R. Shear versus tensile bond strength of resin composite bonded to ceramic. J Dent Res, v. 74, n. 9, p. 1591-1596, Sept. 1995.        [ Links ]

7. DEMARCO, F. F. et al. Shear bond strength of glass ionomer cement. J Dent Res, v. 74, spec. issue, p. 800, Apr. 1995. (Abstr. n. 212).        [ Links ]

8. DEMARCO, F. F. et al. Tensile bond strength of two dentin adhesives systems to dentin. Braz Dent J, v. 8, n. 2, 1997.        [ Links ]

9. FOWLER, C. S. et al. Influence of selected variables on adhesion testing. Dent Mater, v. 8, n. 4, p. 265-269, July 1992.        [ Links ]

10. GWINNETT, A. J. A new method to test the cohesive strength of dentin. Quintessence Int, v. 25, n. 3, p. 215-218, Mar. 1994.        [ Links ]

11. GWINNETT, A. J. Shear bond strength, microleakage and gap formation with fourth generation dentin bonding agents. Am J Dent, v. 7, n. 6, p. 312-314, Dec. 1994.        [ Links ]

12. JAMESON, M. W. et al. Effect of storage media on subsequent water loss and regain by human and bovine dentine and on mechanical properties of human dentine in vitro. Arch Oral Biol, v. 39, n. 9, p. 759-767, Sept. 1994.        [ Links ]

13. KATO, G.; NAKABAYASHI, N. Influence of exposed collagen etched with phosphoric acid to bonding. J Dent Res, v. 74, spec. issue, p. 403, Mar. 1995. (Abstr. n. 19).        [ Links ]

14. LEHMAN, M. L. Tensile strength of human dentin. J Dent Res, v. 46, n. 1, p. 197-201, Jan./Feb. 1967.        [ Links ]

15. LOS, S. A.; BARKMEIER, W. W. Shear bond strength of composite to dentin using newer generation dental adhesive systems. Esthet Dent Update, v. 5, n. 2, p. 46-49, Apr. 1994.        [ Links ]

16. MARSHALL Jr., G. Dentin: microstructure and characterization. Quintessence Int, v. 24, n. 12, p. 606-617, Dec. 1993.        [ Links ]

17. MORI, M. Estudo da distribuição das tensões internas em um dente natural e em dente restaurado com coroa metalo-cerâmica e retentor intra-radicular fundido, sob carga axial - método do elemento finito. São Paulo, 1994. 61 p. Thesis (Ph. D) - Faculty of Dentistry, University of São Paulo.        [ Links ]

18. PASHLEY, E. L. et al. Bond strength to superficial, intermediate and deep dentin in vivo with four dentin bonding systems. Dent Mater, v. 9, n. 1, p. 19-22, Jan. 1993.        [ Links ]

19. PERINKA, L. et al. Dentin thickness, hardness, and Ca-concentration versus bond strength of dentin adhesives. Dent Mater, v. 8, n. 4, p. 229-233, July 1992.        [ Links ]

20. REES, J. S.; JACOBSEN, P. H. The elastic moduli of enamel and dentin. Clin Mater, v. 14, n. 1, p. 35-39, Jan. 1993.        [ Links ]

21. SANO, H. et al. Tensile properties of mineralized and demineralized human and bovine dentin. J Dent Res, v. 73, n. 6, p. 1205-1211, June 1994.        [ Links ]

22. SANO, H. et al. Relationship between surface area for adhesion and tensile bond strength: evaluation of a micro-tensile bond test. Dent Mater, v. 10, n. 4, p. 236-240, July 1994.        [ Links ]

23. SANO, H. et al. Tensile properties of resin-infiltrated demineralized human dentin. J Dent Res, v. 74, n. 4, p. 1093-1102, Apr. 1995.        [ Links ]

24. SMITH, D. C.; COOPER, W. E. G. The determination of shear strength. Br Dent J, v. 130, p. 333-337, Apr. 1971.        [ Links ]

25. SWIFT Jr., E. J. et al. Bonding to enamel and dentin: a brief history and state of the art, 1995. Quintessence Int, v. 26, n. 2, p. 95-110, Feb. 1995.        [ Links ]

26. STANLEY, H. R. An urgent plea for a standardized bonding (adhesion) test. J Dent Res, v. 72, n. 10, p. 1362-1363, Oct. 1993.        [ Links ]

27. VAN NOORT, R. et al. The effect of local interfacial geometry on the measurement of the tensile bond strength to dentin. J Dent Res, v. 70, n. 5, p. 889-893, May 1991.        [ Links ]

28. WATANABE, L. G. et al. Dentin shear strength: effects of tubule orientation and intratooth location. Dent Mater, v. 12, n. 2, p. 109-115, Mar. 1996.        [ Links ]

29. WILLIAMS, K. R. et al. Finite element stress analysis of restored teeth. Dent Mater, v. 3, n. 4, p. 200-206, Aug. 1987.        [ Links ]


Recebido para publicação em 17/09/96
Aceito para publicação em 07/04/97



* Ph. D. student, Operative Dentistry, School of Dentistry, University of São Paulo. Assistant Professor, School of Dentistry, Federal University of Pelotas, Rio Grande do Sul.
**Ph. D. student, Operative Dentistry, School of Dentistry, University of São Paulo. Assistant Professor, School of Dentistry at Ribeirão Preto, University of São Paulo.
***Dean, School of Dentistry, University of São Paulo.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License