Multi-modal characterization of polymeric gels to determine the influence of testing method on observed elastic modulus
David M. Kingsley, Caitlin H.McCleery, Christopher D.L.Johnson, Michael T.K.Bramson, Deniz Rende, Ryan J.Gilbert, David T.Corr
Journal of the Mechanical Behavior of Biomedical Materials, available online 10 January 2019, https://doi.org/10.1016/j.jmbbm.2019.01.003
Demand for materials that mechanically replicate native tissue has driven development and characterization of various new biomaterials. However, a consequence of materials and characterization technique diversity is a lack of consensus within the field, with no clear way to compare values measured via different modalities. This likely contributes to the difficulty in replicating findings across the research community; recent evidence suggests that different modalities do not yield the same mechanical measurements within a material, and direct comparisons cannot be made across different testing platforms. Herein, we examine whether “material properties” are characterization modality-specific by analyzing the elastic moduli determined by five typical biomaterial mechanical characterization techniques: unconfined-compression, tensiometry, rheometry, and micro-indentation at the macroscopic level, and microscopically using nanoindentation. These analyses were performed in two different polymeric gels frequently used for biological applications, polydimethylsiloxane (PDMS) and agarose. Each was fabricated to span a range of moduli, from physiologic to supraphysiologic values. All five techniques identified the same overall trend within each material group, supporting their ability to appreciate relative moduli differences. However, significant differences were found across modalities, illustrating a difference in absolute moduli values, and thereby precluding direct comparison of measurements from different characterization modalities. These observed differences may depend on material compliance, viscoelasticity, and microstructure. While determining the underlying mechanism(s) of these differences was beyond the scope of this work, these results demonstrate how each modality affects the measured moduli of the same material, and the sensitivity of each modality to changes in sample material composition.
Compression testing
Cylindrical test specimens were punched from the prepared gels, using a 4-mm biopsy punch, to approximate the ratio of sample geometries (4-mm diameter, 2-mm thickness) specified by ASTM standard D575-91. Samples were characterized on a Mach-1 mechanical tester (BMM Inc., Laval, QC, Canada) equipped with either a 70-N or 1.5-N load cell, depending on sample rigidity. Each test specimen was compressed to approximately 0.50 strain (absolute compression distance of 1.125 mm) at a constant rate of 0.03 mm/s. A stress-strain curve was constructed based on the measured force and displacement data (normalized to initial sample cross-sectional area and length, respectively). The elastic modulus was estimated from the linear region of this curve, which occurred between 0.05–0.25 strain for PDMS, and 0–0.10 strain for agarose. Six material test specimens were punched from three independently made gels, for all PDMS and agarose groups (n = 18 per group).
Indentation testing
Indentation characterization was performed on samples within 90-mm Petri dishes, using a Mach-1 mechanical tester (BMM Inc., Laval, QC, Canada) with a spherical tip probe (6.35 mm diameter). A 1.5-N or 70-N load cell was used for all samples, depending on sample rigidity. To characterize, the probe height was first calibrated to a Petri dish containing no gel, to standardize the height of the probe. Gel tests began using the system's “Find Contact” mode, wherein the probe is lowered at a rate of 0.03 mm/s until a load was detected. Based on the determined initial sample height, the test was set to a total probe displacement of 30% strain, at a constant rate of 0.03 mm/s for all samples. The elastic moduli of indentation data were estimated using a Hertzian model, which accounts for the contact mechanics between a sphere and an elastic solid [12, 13]. To utilize the Hertz model, measured indenter force was plotted against the indentation depth raised to the 3/2 power, and the slope of the initial linear region, (between 0–0.1 strain for PDMS, and 0.03–0.09 strain for agarose), was used to solve for elastic modulus, E, in Eq. (2),
where F is detected force, R is the radius of the indenter, d is the displacement, and ? is the Poisson's ratio of the sample. Three independent gels were fabricated for testing per sample condition with six indentations per gel (n = 18).