New York University

David B. Kriser Dental Center

College of Dentistry

Graduate School of Arts and Science                                                                                                         Nov. 7, 1998

Department of Dental Materials Science

345 East 24th Street

New York, NY  10010

Tel.:  212-998-9637

Fax: 212-995-4085

Dr. Samuel Waknine

DRM Research Laboratories Inc.

29 Business Park Drive

Branford, CT  06045

Dear Dr. Waknine,

Please find enclosed a final report entitled “Performance probability - Material/Operator contribution -Direct restorative composite materials”.  The study was undertaken by Dr. Seung-Il Eom as part of MS thesis requirements for the Masters degree in Dental Materials Science under my supervision.  Additional results and analysis which do not form part of the MS thesis are also included in this report.

The Biaxial flexure strength (BFS) and Diametral tensile strength (DTS) of “DiamondLite” microhybrid composite restorative material (DRM Res. Labs. Inc., product) were compared to two other popular micrhybrid direct composite formulations.  The mechanical property comparisons indicate that “DiamondLite” ranks first among the materials evaluated, in BFS and DTS.  Further, DLTE satisfies the DTS requirement as per ADA Spec. #27, at the 0.1% failure probability compared to the other materials.

The report submits the results of the Weibull regression analysis of test data.  The characteristic strength values are presented for various levels of failure probability as well.

Please feel free to contact for any clarifications or additional information.

We thank DRM Labs for the Material and Equipment grants required for this study.

Sincerely,

T.V.V. Raghavan

T. V. Vijayaraghavan,  PhD

Associate Professor, Dental Materials Science

Division of Restorative and Prosthodontic

NYU Kriser Dental Center

Performance probability - Material/Operator contribution -Direct restorative composite materials

T. V. Vijayaraghavan,  PhD. and Seung-Il Eom, DDS, MS

Department of Dental Materials Science, New York University Kriser Dental center, NY 10010

Objective

The objective of this study is to evaluate the Biaxial flexure strength (BFS) and  Diametral tensile strength (DTS) of three direct composite materials (DiamondLite, Z100 and Prisma TPH) and evaluate the performance probability from mechanical strength data.

Introduction and background

Composite resin dental restorative materials are blends of hard glass or glass ceramic particulate fillers and a softer polymeric resin matrix. Due to the technique sensitivity of these materials, operator manipulation and handling differences can alter the homogeneity of the material and lead to property differences in the set form.  Flaws such as, voids, micro-porosities on surface or bulk, may be material inherent and/or infiltrated or incorporated during manipulation and handling; In the presence of an external load or force they act as local stress concentration points and failure is initiated at lower forces[i]^,[ii].

The spread in test data at constant level of experimental errors, reflects variations in material homogeneity.  The distribution of sample strength in the test data is determined by the statistical nature (size and distribution) of the weak links leading to failure, i.e., differences in ‘flaw population characteristics’ from sample to sample.  In the case of technique sensitive materials such as composites, the wide range of operators who would handle this material, can compromise the material performance by inappropriate handling or manipulation technique.  In the presence of an external load, flaws localize stresses and failure is initiated at lower tensile forces and leads to lower strength values or brittle behavior.

The behavior of brittle (flaw dependent strength) materials is best represented by the “Weibull distribution function”[iii]^,[iv].  ‘Weibull statistical analysis’[1] of test data is one method to characterize the variability of test data of brittle materials[2].  The data from ‘N’ number of trials or replicates of a sub-group is analyzed using ‘Weibull statistics’ to estimate two parameters; 1) s₀, termed the characteristic or median (probable) strength and ‘m’ the ‘Weibull modulus’ or shape parameter, which can be used to estimate the failure probability.  A high value of ‘m’ indicates a greater level of homogeneity in material character, while a lower value of ‘m’ would be an indicator of heterogeneity or a wider spread in test data about s₀.  From the value of ‘s₀’ and ‘m’ the failure stress for a failure probability of 0.001, 0.1, 0.9 and 0.999(1, 10, 90 or 999 in 1000).labeled as s_0.001, s_0.1, s_0.9, and s_0.999 can be estimated.  This allows ways to optimize the performance of the material.

The values of s_0.001, s_0.1, indicate the possible performance of the material, for an operator, who may be a junior operator or a first time user; while the values s_0.9, and s_0.999 reflect the performance obtainable from a master technician or an experienced user.  The value of s ₀, may be considered as a performance of the material achievable  by a ‘median’ operator.  The inherent limitation or attainable performance of the material is also represented by the value of s_0.999; i.e., as good as it gets!. The conformity of the test data to the ‘Weibull distribution function’ for a given sub group of N trials is indicated by the value of R(correlation coefficient); the higher the value, the greater the confidence in the test data.  When R is >0.9 the test data has a greater validity.

The estimate of the uniaxial tensile strength for brittle materials is obtained from Diametral Tensile, Three or Four point bend flexure testing.  Flexure testing or bend tests have the advantage that a state of pure tension can be established on one side of the specimen.  However, the stress state under In-Vivo service conditions in the oral environs are not purely uniaxial, but biaxial or triaxial in nature[v].  A biaxial flexure test has been used frequently for the determination of fracture characteristics of brittle materials (Ban and Anusavice) and is increasingly being adopted for the evaluation of brittle dental materials^,[vi].  The ease of sample preparation, the elimination of edge effects, the similarity to clinical size scale and intra oral loading conditions are the main advantages of biaxial compared to uniaxial flexure testing.  Further, the evaluation of slightly warped specimens and the possibility of estimation of uniaxial flexure data from biaxial test data are additional advantages of Biaxial flexure testing [vii].

This study is a companion study to previous reported data[3] on crown and bridge materials.

Materials and Methods

Table 1. lists the materials used.

Sample preparation                                 Biaxial flexure disk samples 14 mm in diameter and 1.4 mm thick were prepared using a bulk packing procedure adopted in the previous study².  A length of the cylindrical segment of the composite dispensed from the syringe was placed on a flat glass slab, and patted down using gentle pressure along the long axis with a stainless steel placement instrument(#324).  A transparent plastic film followed by a glass slide or slab was placed below the mold.  The cylindrical paste form was placed inside a stainless steel split ring mold(14 mm diameter and 1.4 mm thick).  A second plastic film followed by a glass plate was overlaid on top and gently pressed to distribute the resin evenly within the circumference of the ring .  Two butterfly clamps were used to squeeze the excess composite through the split ring opening.  Approximate time under clamps was 4-5 minutes for both methods of sample preparation.

                      A Halogen visible light curing booth (DiamondLite, VL Halogen (DRM Res. Labs. Inc., Branford, CT), generating 500 mW/cm² at 450-550 nm wavelength band, was used to cure the composite.  The cure time  per surface was 40 seconds for all materials (DLTE, Z100 and PTPH).  After curing, specimens were separated from the ring mold and stored in distilled water maintained at 37 °C for 24 hours prior to testing.  A total of 20 samples were prepared per group.  All samples were evaluated visually with a low powered magnifier for surface defects and markings etc.  These samples were excluded from the test group.

Diametral tensile test was performed as per ADA specification #27.  Stainless steel molds 6 mm in diameter and 3 mm in height were used to prepare DTS test specimens.  The curing modality, time and storage (post curing ) conditions were identical to those used for BFS test samples.

Testing         BF tests were performed using an "Instron" machine (Instron Corp, Canton, MA), with the cross head displacement rate set at 0.50 mm/minute(0.2 inches/minute).  Disk specimens were placed on eight steel balls[4] (3.2 mm in diameter) spaced along the circumference of a circle of diameter 10 mm.  A flat ended piston of diameter 1.2 mm  was used to load the sample at the center of the disk.  Tests were carried out at room temperature.  The BFS was calculated from the maximum load at failure recorded on a chart recorder attached to the test machine.

Biaxial Flexure Strength                         The maximum tensile stress under biaxial loading condition, prior to failure,  was calculated according to the following equations developed by Marshall (1980)[viii].  The failure stress, s _f , can be expressed as:

s _f = A. P_f /t²                                              (1),  where, ‘A’ is given by the expression,

A = (3/4p) [2.(1+n).ln(a/r₀^*) + (1-n).(2a²-r₀^*2)/2b² + (1+n)]                                        (2)

P_f, is the applied load at fracture in Newtons, n, the  Poisson`s ratio (for composite materials a value of 0.24 was used), ‘a’, the radius of the support circle, ‘b’ is the radius of disc specimen, ‘r₀^*’ the equivalent radius given by                                              r₀^* =(1.6 r₀² + t²)^1/2 - 0.675t,

where, ‘t’ is the thickness of the disc specimen, and ‘r₀‘is the radius of the piston².

                      The Diametral tensile testing was performed as per ADA specification #27.

Statistical Analysis                                  Mean and standard deviation were obtained from raw data.  Student t-Test was used for evaluation of significant differences between mean value pairs.  Weibull regression analysis¹ was performed using a computer, to evaluate the characteristic strength, s₀ and m, the Weibull modulus from the raw BFS and DTS data.

Results

The  results of the Weibull regression analysis of BFS and DTS raw data are shown in tables 2 and table 3, respectively.  Of the materials studied DLTE showed the highest values of characteristic BFS and DTS.  The high values of correlation coefficient, R, for the linear regression analysis indicate that the distribution of raw data shows a good fit to the ‘Weibull distribution function’ for all the materials.  For DLTE the BFS values obtained by Operators ‘a’ and ‘b’ were not significantly different (p<0.05) for DLTE..

Discussion

The sample preparation procedure adopted in this study may be termed ‘bulk packing’ since a volume of the composite from the syringe was used without any manipulation.  Surface defects or flaws incorporated into the sample in addition to the preexisting volume defects from fabrication are the likely sources for fracture initiation.  In an incremental filling mode commonly adopted for direct restoration in the clinic, the layering of composite may incorporate additional voids or flaws in the volume[5]

BFS is sensitive to both surface and volume defects.  While the mode of packing the material, i.e., bulk packing, is expected to minimize the inclusion of volume flaws; surface defects are expected to be the prime reason for the low values of strength obtained in the sub groups, indicating technique sensitivity of the material to manipulation. The localization of tensile forces at surface flaws or sub-surface flaws is the primary cause of failure in BF test mode for this mode of sample preparation..

Factors that may affect this include sharp instruments, surface air or dust entrapment during cavity filling, the surface finish, wetting characteristics, etc.  Visual or ocular assisted examination, for subsurface and surface defects and their elimination prior to curing may aid in achieving values closer to the attainable strength value for the material(s_0.999).

The spread in BFS and DTS data is narrower for the DLTE material (marginally higher ‘m’ value, non-significant) in comparison to PTPH or Z100 materials.  For all materials the drop in the s value at the low failure probaility end (s _0.001 and s_0.1) is greater compared to the high probability end (s_0.9, and s_0.999) of the distribution.  This indicates that manipulation and handling procedures are responsible for the low strength values obtained in the test data.

In the case of direct materials, incremental filling or multiple layering techniques may lead to an increased probability for incorporation of flaws in the bulk of the restoration.  Finishing methods adopted by the clinician, relating to surface grinding for occlusion matching and polishing methods are expected to have an effect on the production of surface flaws.  A study based on mathematical modeling by Versilus et al[ix] suggest that incremental filling does not reduce polymerization shrinkage and/or reduce residual stresses.  The effect of surface finishing procedures on the BFS is currently being studied.

For direct composite materials, which are usually handled by clinicians, the characteristic strength values should represent a greater correlation due to limited manipulation and handling (cavity filling, repair), unlike for indirect composite materials (crowns and bridge, veneers etc.) where multiple shades and contouring, create a need for greater care in handling to maintain the materials integrity.

Clinical implications

The mean DTS requirement as per ADA Spec. #27[x] for Type II Direct filling resin composite materials is about 34 MPa.  This property requirement does not take in to account clinical parameters such as manipulation, handling and operator expertise into account on restorative function.  For all materials the mean DTS and the characteristic DTS value exceed 34 MPa.  Of the three only DLTE exceeds a value of 34 MPa at probability of failure of 0.1%, Table 3.  Based on DTS, a higher level of clinical performance reliability for DLTE material (s_0.1>s_0.001>34 MPa), is expected in comparison to materials where the stress for 0.1% failure probability is less than 34 MPa (s_0.1>34 MPa>s_0.001).

Conclusions

1.     The reliability of strength data was evaluated using Weibull statistical analysis.  The high value of correlation coefficient obtained for the regression fit indicate that the BFS and DTS test data conforms to the two parameter’ Weibull distribution function’.

2.     There was no significant difference in mean BFS values as a function of operator for DLTE material (p>0.05) for the sample preparation procedure adopted in this study.

3.     The characteristic strength, s₀, s_0.01 and s_0.9 s_0.999(BFS and DTS) were estimated to be higher for DLTE material compared to PTPH and Z100 materials; indicating an increased confidence in clinical performance probability.

Table 1. List of VLC materials

Table 2. Results of Weibull Regression Analysis

Table 3. Mean, SD and Weibull analysis results for DTS

References