To investigate the single-edge notched (SEN) bend fracture toughness ( K IC ) testing methodology as a reproducible and discriminatory mechanical testing protocol for encapsulated and hand-mixed glass-ionomers (GI).
SEN bend test-pieces (35.0 ± 0.1 mm length, 6.0 ± 0.1 mm width, 3.0 ± 0.1 mm thickness with a sharp notch formed at mid-length by embedding a scalpel blade) were prepared for K IC testing using three encapsulated GI products (Chemfil Rock, Fuji IXGP Fast Capsule and Ionofil Molar AC). In addition, test-pieces were prepared from a hand-mixed GI product (Ionofil Molar) which contained between 100% and 20% of the manufacturer’s recommended powder content (in 10% decrements) for a constant weight of liquid. Groups of 20 test-pieces were prepared for each encapsulated GI product ( n = 3) and hand-mixed GI powder:liquid mixing ratio ( n = 9). Data were statistically analyzed and the coefficients of variation (CoV) determined for each encapsulated GI product and hand-mixed GI powder:liquid mixing ratio.
The K IC testing methodology failed to discriminate between the encapsulated GI products that were investigated ( p = 0.225). For the hand-mixed GI, the K IC testing methodology also failed to discriminate between the powder:liquid mixing ratios investigated ( R 2 = 0.576). The pooled CoV (10%) for the encapsulated GI products and for the powder:liquid mixing ratio groups (12%) identified the reproducibility of the test for this experiment. For the hand-mixed GI mixing ratio groups with between 100% to 50% of the recommended powder content, no trend could be discerned.
The K IC testing methodology failed to discriminate between different encapsulated GI products and hand-mixed GI powder:liquid mixing ratio groups investigated, despite K IC being an intrinsic material property and the coefficient of variation being acceptable.
The compressive fracture strength (CFS) test is the strength test specified by the International Organisation for Standardisation (ISO) standard for water-based cements. Concerns have been raised regarding the reliability of the CFS testing protocol for glass-ionomers (GIs) in a ‘test-house variability’ study in 1990 in which inter- and intra-operator variability was identified. Fleming et al. revisited the CFS testing methodology given in ISO 9917-1:2007 for GI products to identify the reproducibility of test results achieved in terms of inter- and intra-operator variabilities. These authors highlighted the batch censoring specified in the ISO 9917-1:2007 to be unsafe as it misidentified operator variability and therefore a statistical approach employing a minimum of 20 test-pieces is required for a meaningful interpretation of the CFS data. They concluded that the CFS test could be performed reliably if test-piece preparation techniques and laboratory conditions are standardized, although such level of control is not routinely applied, as is evident from the large variation in CFS data reported in the dental literature .
To identify a potential replacement testing protocol for the CFS test, Dowling et al. investigated the three-point flexure strength (TFS) and biaxial flexure strength (BFS) testing methodologies using three encapsulated GIs (Chemfil Rock, Fuji IX GP Fast Capsule and Ionofil Molar AC). With their results, these authors could not differentiate between the reproducibility of the testing protocols investigated ( p = 0.632), with the pooled coefficients of variation (CoV) of the CFS, TFS and BFS data reported being 10% (0.104), 12% (0.116) and 11% (0.109), respectively. Interestingly, both the TFS ( p = 0.271) and BFS ( p = 0.134) testing methodologies failed to discriminate between the three encapsulated GI products, despite being supplied by different manufacturers with different glass compositions, liquid compositions and powder:liquid mixing ratios. In contrast differences in CFS ( p = 0.001) were evident. Baig et al. used the same encapsulated GI products to examine the reliability of the Hertzian indentation (HI) testing protocol for disc-shaped test-pieces of 10 mm diameter and 3 mm thickness . They identified the HI test to be a reproducible testing methodology with a lower pooled CoV, 7% (0.071). In addition, the HI test was able to discriminate between these three encapsulated GI products ( p < 0.001). However, the ‘Hertzian load-bearing capacity’ reported by Baig et al. is the maximum load-to-failure, which was not a material property and therefore could be interpreted as the ‘quality of the cement’ rather than a ‘predictive value’ .
The validity of the CFS test has also been questioned because the stress at failure calculation does not take into account the failure mechanism operating during CFS testing, whereby cylindrical specimens fail by ‘some unresolved combination of tension and shear’ stresses . The TFS , BFS and HI tests have been identified as valid testing methodologies for encapsulated GI products because failure occurs from the surface under tension. Bar-shaped test-pieces used for TFS testing are not geometrically representative of loading clinical GI restorations. Also, the force at failure will depend upon the perfection of the test-piece surface and well as any inherent material property. The loading geometry of the disc shaped BFS test-piece eliminates the test-piece edge effect. This test-piece more closely resembles the surface area to volume ratio of a dental restoration, although loading such a test-piece through a 4 mm diameter ball when supported on an annulus (i.e. in the absence of a supporting substructure) fails to replicate the situation met in vivo. In terms of simulating the clinical scenario, loading HI test-pieces using a 20 mm hard steel ball to simulate cusp contact when resting freely on a dentin analogue seems most relevant, although the ‘Hertzian load-bearing capacity’ recorded as the maximum load-to-failure is not a material property.
An ideal mechanical testing protocol for GIs should be reliable, valid and discriminatory between encapsulated GI products from different manufacturers and between hand-mixed GIs prepared at varying powder:liquid mixing ratios routinely employed in the clinic . Fracture toughness ( K IC ) is an intrinsic material property, independent of specimen dimensions, that represents a material’s resistance to crack propagation and has been employed to test resin-bonded composites (RBCs) , GIs , ceramics and dental amalgams . The aim of the current study was to investigate whether the single-edge notched (SEN) bend test-piece K IC testing protocol is a reproducible and discriminatory testing methodology to act as a performance indicator for encapsulated and hand-mixed GI restorative materials. The hypotheses proposed were: (1) that the test would be a reliable and discriminatory mechanical testing protocol to act as a performance indicator by discriminating between different encapsulated GI products, and (2) that the presence of a linear deterioration in K IC on reducing the powder content from the optimum that specified by the manufacturer for a hand-mixed GI restorative material is indicative of a reliable and discriminatory mechanical testing protocol.
Materials and methods
A polytetrafluoroethylene (PTFE) lined brass split-mould resting on top of a PTFE lined brass base plate was used to prepare SEN bend test-pieces (35.0 ± 0.1 mm length, 6.0 ± 0.1 mm width, 3.0 ± 0.1 mm thickness). Prior to test-piece manufacture, the PTFE lined surfaces (forming the two side walls) of the mould ( Fig. 1 ) were covered with two layers of transparent adhesive tape (Sellotape, Henkel Ltd., Winsford, Cheshire, UK) to facilitate test-piece removal. The mould was designed and constructed with PTFE on the surfaces in contact with the material to avoid adhesion. The PTFE was backed with brass to give the mould rigidity. However, applying adhesive tape to the walls and base produced an improvement. In addition the tape aided removal of the set cement. Peeling the tape from the top surface of the mould created a tab which could be lifted to apply a uniform force to the test-piece. This avoided direct and localized force on the test-piece which risked fracture during demoulding . In addition, a single layer of the transparent adhesive tape was applied to the PTFE surface of the brass base plate ( Fig. 1 ). Following application of the adhesive tape, the structural components of the split-mould were re-assembled and secured with multiple screws.