JDR JDR Most Cited Articles
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eckrote, K.A.
Right arrow Articles by Goldberg, A.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eckrote, K.A.
Right arrow Articles by Goldberg, A.J.
J Dent Res 82(4): 262-266, 2003
© 2003 International and American Associations for Dental Research

RESEARCH REPORT
Biomaterials & Bioengineering

Shear in Flexure of Fiber Composites with Different End Supports

K.A. Eckrote1, C.J. Burstone2, M.A. Freilich3, G.E. Messer4, and A.J. Goldberg1,*,3

1 Center for Biomaterials, MC-1615,
2 Department of Orthodontics,
3 Department of Prosthodontics and Operative Dentistry, 1-4School of Dental Medicine,
University of Connecticut Health Center, Farmington, CT 06030;

*corresponding author, Goldberg{at}nso1.uchc.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The integrity of fiber-reinforced composite (FRC) prostheses is dependent, in part, on flexural rigidity. The object of this study was to determine if the flexure behavior of uniform FRC beams with restrained or simply supported ends and various length/depth (L/d) aspect ratios could be more accurately modeled by correcting for shear. Experimental results were compared with three analytical models. All models were accurate at high L/d ratios, but the shear-corrected model was accurate to the lowest, more clinically relevant, L/d values. In this range, more than 40% of the beam deflection was due to shear.

KEY WORDS: fiber-reinforced • composites • flexure • shear • restraints


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
As fiber-reinforced composite procedures and materials evolve, one of the fundamental issues will continue to be rigidity or stiffness of the prosthesis. Not only is stiffness important for clinical function, and the inevitable comparison with metal-ceramics, but also the success of the adhesive interfaces between the FRC lamina, between the substructure and the veneer (Rosentritt et al., 2001), and between the prosthesis and tooth structure (Loose et al., 1998) are dependent in large part on the deflection and deformation of the restoration. Most studies of FRC rigidity have used three-point flexural loading (Freilich et al., 1999; Vallittu, 1999; Behr et al., 2000; Karmaker and Prasad, 2000) of standard beams (International Standard, 1992; ASTM Standard, 1997), and it was implicit that prosthesis rigidity would be proportional to the measured flexure modulus of the material. However, in vitro tests of FRC fixed partial denture (FPD) substructures (Haser, 1999) show that prosthesis rigidity cannot be accurately predicted from these earlier reports. While several factors are probably involved, it seems reasonable that the restraint of the prosthesis retainer and abutment, in contrast to the freely supported three-point specimen, contributes to this discrepancy. Accordingly, one purpose of this study was to develop and evaluate a laboratory flexure test of FRC beams based on restrained-end loading.

In addition, the classic flexure formula (Popov, 1976)

and its derivatives, routinely used in flexure studies and standards, is probably insufficient for this problem. It is well-documented that classic small deflection theory is not accurate when the length/depth (L/d) ratio of the beam is generally below 20 (Mullin and Knoell, 1970; Zweben and Hahn, 1982; ASTM Standard, 1997). The typical fiber-reinforced composite FPD substructure would have a span across the edentulous region of less than 12 mm and a depth of at least 3 mm (Rosentritt et al., 2001; Freilich et al., 2002), resulting in an L/d of less than 4. In flexural loading of short, thick beams, plane sections do not remain plane, and thus there is a shear contribution to deflection. In addition, effects at the supports and load may have practical significance at short span lengths. Models of three-point loading corrected for shear have appeared in the general composites literature (Zweben et al., 1979). One earlier study in the dental literature did model flexural behavior of FRC orthodontic wires, attributing the departure from classic behavior to fiber damage at the support as well as to shear (Jancar et al., 1994). Other investigators (Karmaker and Prasad, 2000) have described the L/d effect for short FRC beams, but these studies examined L/d values greater than 10 and used three-point loading. Therefore, the goal of the present study was to determine if the flexure behavior of FRC beams with restrained or simply supported ends and various L/d ratios could be more accurately modeled by correcting for shear.


   MATERIALS & METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Restrained-end Test Apparatus
A restrained-end test apparatus (Fig. 1Go) was designed and fabricated. Specimens of fiber-reinforced composites with uniform rectangular cross-sections were clamped at both ends with a load applied in the center. Screws used to secure the samples in the end-restraints were tightened to 2200 gm*cm with a torque watch (Waters Manufacturing, Inc., Wayland, MA, USA). To accommodate various sample dimensions and span lengths, the clamps were adjustable on the base of the testing jig. A central load was applied until failure with an Instron testing machine (model 1113, Instron Corp., Canton, MA, USA) by means of a 10-mm-wide blade indenter with a 1.6-mm end radius, at a crosshead speed of 0.05 cm/min. For comparison purposes, flexure behavior of the FRC was evaluated with the same test parameters, but with three-point loading. Tests for both loading conditions were conducted over a range of length/depth (L/d) values.



View larger version (41K):
[in this window]
[in a new window]
  		Figure 1. Schematic of restrained-end test apparatus showing adjustable supports to accommodate various span lengths. The test sample is restrained by a bar with two screws at each support.

 
Fabrication and Testing of FRC at Various L/d
Strips of continuous, unidirectional, pre-impregnated fiber-reinforced composite (FibreKorTM, Pentron, Inc., Wallingford, CT, USA) were used to fabricate the test specimens. In this composite, 6- to 10-µm-diameter glass fibers are uniformly distributed in an unpolymerized dimethacrylate matrix (Karmaker et al., 1997; Freilich et al., 1998). Fiber volume fraction, Vf, was 42.5%, as confirmed by ashing. Each 3 x 0.3 mm strip was placed with the fibers parallel to the long axis of a Teflon mold and condensed to eliminate voids between layers. The samples were visible-light-cured for 9 min (Cure LiteTM Plus, Pentron, Inc., Wallingford, CT, USA), carefully removed from the mold, light-cured for an additional 9 min, and finally post-cured in a vacuum/heat-curing unit for 30 min at 110°C (Conquest Curing Unit, Pentron, Inc., Wallingford, CT, USA). Specimens were stored at ambient conditions for fewer than 4 days prior to being tested.

Specimen dimensions were 2 x 2 x 80 mm3 (width x thickness x length) or 2 x 1 x 80 mm3. The former were tested at span lengths of 40, 30, 20, and 10 mm, and the latter were tested at span lengths of 60, 50, 40, 30, and 20 mm (n = 3 for each L/d group). This resulted in L/d ratios of 5 to 60. Experimental rigidity values (load/deflection) were calculated within the elastic region. The load at the elastic limit, determined graphically as the limit of the initial linear region of the load-deflection curve, and the maximum load were recorded.

Analytical Models of Flexural Rigidity
The experimental values of rigidity for the FRC beams obtained over various L/d ratios were compared with three analytical models of flexure behavior. These were (a) the standard flexure formula based on classic small deflection theory, (b) a model that corrected the standard flexure formula for the contribution of shear, and (c) a model that assumed softening due to fiber damage at the sample clamp plus a shear contribution (Jancar et al., 1994). The derivations of these models are shown below, in Appendix 1 (www.dentalresearch.org), and in the cited references.

The deflection, {Delta}, of a centrally loaded, slender, restrained-end beam can be written as (Olsen, 1956):


(1)
where P is the applied load, L is the span length, E is the modulus of elasticity, and I is the moment of inertia. This relationship is based on classic small deflection slender beam theory and assumes negligible shear deformation. The theoretical rigidity of a rectangular beam of width b and depth d is therefore expressed as:

(2)

The longitudinal modulus of the composite, E, was calculated to be 34.2 GPa from the rule of mixtures (Agarwal and Broutman, 1980):

(3)
where E and V represent tensile elastic modulus and volume fraction, respectively, and the subscripts f and m refer to the fibers and the matrix of the composite, respectively. It was assumed that the contribution of the matrix to E was negligible. The tensile modulus value for glass fibers was obtained from the literature (ASM International Handbook Committee, 1987). Equations 2 and 3GoGo were used to calculate classic small deflection theoretical beam rigidity, P/{Delta}, for a restrained-end beam with various ratios of L/d. An analogous arrangement of the more widely recognized relationship,

(4)
was used as the standard flexure model for three-point loading.

The second analytical model of beam rigidity included the contribution of shear. In this model, total deflection is the sum of the deflection produced from flexure plus the deflection due to shear, {Delta}S. For rectangular beams with end restraints, the shear deflection can be represented by the following equation (Young, 1989):

(5)
where G is the shear modulus of the material. A value of 0.63 GPa was used for the shear modulus G (Hadjinikolaou, 1994). Substitutions and rearrangement result in the following relationship for beam rigidity, accounting for both flexure and shear deflection:

(6)
where C = 12/192 for restrained-end loading and 1/4 for three-point loading.

The third analytical model was based on flexure, shear, and the assumption of softening at the clamps due to fiber damage (Jancar et al., 1994). These earlier investigators calculated values for an apparent modulus, E*, that included all three effects at various ratios of L/d. In the present study, we substituted these values of E* into Eqs. 2 and 4 to calculate the theoretical rigidity of the FRC beams based on the assumption of flexure, shear, and damage at the clamp. All three models of flexure behavior were statistically compared with the experimental results by means of the chi-square goodness-of-fit test.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The experimentally determined rigidity of the FRC beams at L/d ratios of less than 30 and tested with restrained ends was compared with the three analytical models (Fig. 2AGo). Similar comparisons were made for three-point loading (Fig. 2BGo). Theoretical and measured values of stiffness up to L/d of 60 were tabulated (Appendix 2, www.dentalresearch.org). As anticipated, all three models, in both restrained-end and three-point loading, were accurate at higher values of L/d. At values below approximately 15, the standard flexure formula increasingly overestimated stiffness. The soft zone model overestimated rigidity in restrained-end loading, but is comparable with the shear-corrected formula in three-point loading. The model correcting for only shear was predictive down to the lowest values of L/d. The difference between the model and the experimental data became statistically significant (p < 0.05) below L/d values of 20 and 10, for restrained-end and three-point loading, respectively.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 2. Comparison of experimental rigidity values with the three analytical models. Error bars for the experimental values indicate standard deviations of the calculated means (n = 3). (A) Restrained-end loading. (B) Three-point loading.

 
Comparing measured rigidity (Figs. 2A and 2BGo or Appendices 2A and 2C, www.dentalresearch.org), at high values of L/d, the beams in the restrained-end device are 2-4 times more rigid than those tested with simply supported ends, consistent with the coefficients of 12/192 and 48/192 for the different loading conditions in the flexure formula. At lower values of L/d, the difference decreases, consistent with the decreasing influence of the (L/d)3 term and the common shear term in Eq. 6Go.

While the focus of this study was rigidity, the loads at the elastic limit are also quite important clinically. These values are shown for each loading condition as a function of L/d (Fig. 3Go). The maximum loads were about 50% higher and followed the same general pattern (Appendix 3, www.dentalresearch.org). The loads increased with decreasing L/d and were generally independent of the type of end support.



View larger version (14K):
[in this window]
[in a new window]
  		Figure 3. Experimental load values at the elastic limit for restrained-end loading ({blacksquare}) and three-point loading ({blacklozenge}). Error bars indicate standard deviations of the calculated means (n = 3).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
All analytical models of rigidity became more accurate with increasing values of L/d. This is consistent with generally accepted theory and previous reports of flexure behavior of fiber-reinforced composites in the materials science and dental literature. This is because even under central loads, long slender beams deform primarily due to flexure or bending, there is little contribution to deflection from shear, and end effects are small.

Even though the difference between the shear-corrected restrained-end model and experimental values was statistically significant (p < 0.05) below L/d values of 10, the model was still reasonably predictive in this clinically important region. This is consistent with the relative contributions of shear and flexure to total deflection, which were calculated with Eq. 6 as a function of L/d (Fig. 4Go). Restrained-end loading causes a greater deflection due to shear than found in the three-point test, although the difference between the test methods decreases at lower values of L/d. ISO testing standard 10477 specifies three-point loading with an L/d of 10, so even here almost 40% of the deflection is due to shear.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 4. Contribution of shear (solid lines) and flexure (dashed lines) to the total deflection. Contributions are shown for both restrained-end loading (gray lines) and three-point loading (black lines).

 
Analysis of the data in this study suggests that the rigidity of three-unit FRC prostheses is more dependent on the shear modulus, G, than previously acknowledged. Investigators frequently associate restoration stiffness with elastic modulus, E. For either restrained-end or three-point loading at L/d above about 30, this is true, but the L/d for most three-unit FRC prostheses is well below 10. Accordingly, if an increase in appliance stiffness is desired, attention needs to be directed to the shear modulus as well as to E. The in-plane shear modulus for unidirectional composites with fiber volume fractions below 50% are largely dependent on the matrix properties (Kaw, 1997). Since most dental FRC systems use similar dimethacrylate matrices, it is likely that the G values for most products are comparable. Differences in products are primarily due to fiber content, type, and orientation. Analysis of the current data indicates that independent measurement of G will be necessary for accurate modeling of flexural behavior. Literature suggests that G values for these fiber composites are in the range of 0.5 to 5.0 (O’Brien and Ryge, 1978; ASM International Handbook Committee, 1987), but with clinically relevant dimensions, this range translates to an approximately six- and three-fold difference in rigidity for restrained-end and three-point loaded beams, respectively.

The shear deformation probably also explains the dependence of FRC strength on span length. At higher span lengths, the load is supported primarily by the higher-strength fibers. As L/d is decreased, the load is supported by the fibers in tension and compression as well as the matrix in shear. Therefore, while the clinically important maximum loads (Appendix 3, www.dentalresearch.org) and loads at the elastic limit (Fig. 3Go) increase with decreasing L/d, the calculated apparent strength actually decreases (Appendices 2A, 2C, www.dentalresearch.org). For example, the apparent ultimate strengths of these composites in three-point loading at L/d of 8.5 and 23.6 are 583 ± 89 MPa and 859 ± 68 MPa, respectively.

It is interesting to compare the rigidity of the restrained-end and three-point loading models in the L/d region below 10 (Fig. 2Go). While the shear-corrected restrained-end model predicts about 20% greater rigidity than the three-point shear-corrected model, the two models are practically parallel in this region. This is understandable, since the shear term, 3L/10Gbd, is the same in both models. That is, for centrally loaded beams, the shear contribution is independent of end-restraint. This suggests that this term could be useful in clinical modeling where the support for the prosthesis is geometrically complex.

One reason for the current study was to evaluate the use of a restrained-end test arrangement instead of the more traditional three-point flexure test. Several researchers have studied the in vitro flexure behavior of FRC fixed partial dentures (Loose et al., 1998; Vallittu, 1998; Behr et al., 1999), but these reports focused on the effects of fiber reinforcement and strength or maximum load of the complete prosthesis. Haser (1999) studied non-veneered FRC substructures and reported rigidity as well as load, which allows for some comparison with the present data. The rigidity of FRC substructures bonded to steel dies was reported to be 939 N/mm. The L/d of the FRC section between the retainers was 5.0 and approximately uniform. The rigidity of beams with comparable L/d predicted with the present shear-corrected models for restrained-end and three-point loading are 767 and 607 N/mm, respectively (Fig. 2Go). Therefore, while the restrained-end test is more predictive, both test methods underestimate the rigidity of the prostheses. Perhaps of greater importance, use of the standard flexure formula predicts rigidity of 8755 and 2189 N/mm for restrained-end and three-point loading, respectively, suggesting that the shear correction is more important than the test method when the L/d is below approximately 10.

In the clinical situation, physiological tooth movement (Parfitt, 1960), deformation of the luting agent, and possibly other factors would reduce the apparent rigidity of the prosthesis. It is possible that the shear-corrected three-point test is more representative when clinical factors allow for greater mobility of the prosthesis, such as failure of the luting cement (Loose et al., 1998). The shear-corrected restrained-end model may be more accurate for cases of less prosthesis movement, possibly such as restorations over implants. Of course, neither test method is intended as a replacement for more complex approaches that attempt to reproduce effects of anatomy, periodontal ligament, and other clinical factors. The three-point and restrained-end tests are most useful for developing analytical models and for evaluating material properties, formulation changes, environmental challenges, etc.

The present data can be compared with earlier work by others who studied similar formulations and storage conditions by calculating flexure modulus with Eq. 4 for samples with nominal cross-sections of 2 mm x 2 mm tested at a span length of 20 mm. Under these comparable conditions, Lassila et al. (2002) and Karmaker and Prasad (2000) reported flexure modulus values of 24.0 and 27.5 GPa, compared with the present value of 27.1 GPa.


   ACKNOWLEDGMENTS
 
This study is supported by a grant from the Whitaker Foundation awarded through the Biomedical Engineering Alliance for Connecticut; by Connecticut Innovations, Inc.; and by Pentron Laboratory Technologies, LLC. During part of the time that this work was conducted, Drs. Goldberg and Burstone were University-approved consultants for Pentron, and patents assigned to the University of Connecticut were licensed to Pentron. Potential conflicts were managed according to University policy. This paper is based on a thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at the University of Connecticut. Portions of this study were presented as abstracts at the 1999 & 2002 IADR General Sessions (Vancouver, Canada, and San Diego, CA, USA, respectively).


   FOOTNOTES
 
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org.

Received May 13, 2002; Last revision October 7, 2002; Accepted January 10, 2003


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Agarwal BD, Broutman LJ (1980). Analysis and performance of fiber composites. New York: John Wiley & Sons, p. 22.

ASM International Handbook Committee, editors (1987). Engineered materials handbook. Vol 1: Composites. Metals Park, OH: ASM International, pp. 361, 406.

ASTM Standard (1997). Standard test methods for flexural properties of unreinforced and reinforced plastics and electrical insulating materials. ASTM D790. In: Annual book of ASTM standards, pp. 141-151.

Behr M, Rosentritt M, Leibrock A, Schneider-Feyrer S, Handel G (1999). In-vitro study of fracture strength and marginal adaptation of fibre-reinforced adhesive fixed partial inlay dentures. J Dent 27:163–168.[ISI][Medline]

Behr M, Rosentritt M, Lang R, Handel G (2000). Flexural properties of fiber reinforced composite using a vacuum/pressure or a manual adaptation manufacturing process. J Dent 28:509–514.[ISI][Medline]

Freilich MA, Karmaker AC, Burstone CJ, Goldberg AJ (1998). Development and clinical applications of a light-polymerized fiber-reinforced composite. J Prosthet Dent 80:311–318.[ISI][Medline]

Freilich MA, Meiers JC, Duncan JP, Goldberg AJ (1999). Fiber-reinforced composites in clinical dentistry. Chicago: Quintessence Publishing Co., Inc., p. 18.

Freilich MA, Meiers JC, Duncan JP, Eckrote KA, Goldberg AJ (2002). Clinical evaluation of fiber-reinforced fixed bridges. J Am Dent Assoc 133:1524–1534.[Abstract/Free Full Text]

Hadjinikolaou IN (1994). Fiber-reinforced thermoplastic composites for dental applications (dissertation). Farmington: University of Connecticut Health Center.

Haser KA (1999). Development of a restrained-end testing model for analysis of fiber-reinforced composites for dental applications (dissertation). Storrs: University of Connecticut.

International Standard (1992). Dentistry-polymer-based crown and bridge materials. ISO 10477. Geneva, Switzerland: ISO, pp. 1–8.

Jancar J, DiBenedetto AT, Hadjinikolaou IN, Goldberg AJ, Dianselmo A (1994). Measurement of the elastic modulus of fibre-reinforced composites used as orthodontic wires. J Mater Sci-Mater Med 5:214–218.

Karmaker A, Prasad A (2000). Effect of design parameters on the flexural properties of fiber-reinforced composites. J Mater Sci Lett 19:663–665.

Karmaker AC, DiBenedetto AT, Goldberg AJ (1997). Continuous fiber reinforced composite materials as alternatives for metal alloys used for dental appliances. J Biomater Appl 11:318–328.[Abstract]

Kaw AK (1997). Mechanics of composite materials. New York: CRC Press, LLC, pp. 173-175.

Lassila LV, Nohrstrom T, Vallittu PK (2002). The influence of short-term water storage on the flexural properties of unidirectional glass fiber-reinforced composites. Biomaterials 23:2221–2229.[ISI][Medline]

Loose M, Rosentritt M, Leibrock A, Behr M, Handel G (1998). In vitro study of fracture strength and marginal adaptation of fibre-reinforced-composite versus all ceramic fixed partial dentures. Eur J Prosthodont Restor Dent 6:55–62.[Medline]

Mullin JV, Knoell AC (1970). Basic concepts in composite beam testing. Mater Res Standard 10:16–34.

O’Brien WJ, Ryge G, editors (1978). An outline of dental materials and their selection. Philadelphia: Saunders Company, p. 403.

Olsen GA (1956). Strength of materials. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall, Inc., pp. 283-301.

Parfitt GJ (1960). Measurement of the physiological mobility of individual teeth in an axial direction. J Dent Res 39:608–618.[Abstract/Free Full Text]

Popov EP (1976). Mechanics of materials. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall, Inc., p. 125.

Rosentritt M, Behr M, Kolbeck C, Handel G (2001). In vitro repair of three-unit fiber-reinforced composite FPDs. Int J Prosthodont 14:344–349.[ISI][Medline]

Vallittu PK (1998). The effect of glass fiber reinforcement on the fracture resistance of a provisional fixed partial denture. J Prosthet Dent 79:125–130.[ISI][Medline]

Vallittu PK (1999). Flexural properties of acrylic resin polymers reinforced with unidirectional and woven glass fibers. J Prosthet Dent 81:318–326.[ISI][Medline]

Young WC (1989). Roark’s formulas for stress & strain. 6th ed. New York: McGraw-Hill, Inc., pp. 201-204.

Zweben C, Hahn HT (1982). Static strength and elastic properties. In: Composites design encyclopedia. Wilson DW, Wetherhold RC, Cadot HM, McCullough RL, Pipes RB, editors. Newark: University of Delaware, pp. 1-72.

Zweben C, Smith WS, Wardle MW (1979). Test methods for fiber tensile strength, composite flexural modulus, and properties of fabric-reinforced laminates. In: Composite materials: testing and design (fifth conference), ASTM STP 674. Tsai SW, editor. Philadelphia: American Society for Testing and Materials, pp. 228-262.





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eckrote, K.A.
Right arrow Articles by Goldberg, A.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eckrote, K.A.
Right arrow Articles by Goldberg, A.J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
IADR Journals Advances in Dental Research ®
Journal of Dental Research ® Critical Reviews (1990-2004)