Scanning acoustic microscopy (SAM), when applied to biological samples has the potential to resolve the longitudinal acoustic wave speed and hence stiffness of discrete tissue components. The heterogeneity of biological materials combined with the action of cryosectioning and rehydrating can, however, create variations in section topography. Here, we set out to determine how variations in specimen thickness influence apparent acoustic wave speed measurements

Cryosections (5μm nominal thickness) of human skin biopsies were adhered to glass slides before washing and rehydrating in water. Multiple regions (200x200 μm; n = 3) were imaged by SAM to generate acoustic wave speed maps. Subsequently co-localised 30x30 μm sub-regions were imaged by atomic force microscopy (AFM) in fluid. The images were then registered using Image J. Each pixel was allocated both a height and wave speed value before their relationship was then plotted on a scattergram. The mean section thickness measured by AFM was 3.48 ± 1.12 (SD) μm. Regional height variations influenced apparent wave speed measurements. A 3.5 μm height difference was associated with a 400 ms-1 increase in wave speed. In the present study we show that local variations in specimen thickness influence apparent wave speed. We also show that a true measure of wave speed can be calculated if the thickness of the specimen is known at each sampling point.

In this study we have established a new approach to more accurately map acoustic wave speed (which is a measure of stiffness) within soft biological tissues at micrometer length scales using scanning acoustic microscopy. By using thin (5 μm thick) histological sections of human skin and porcine cartilage, this method exploits the phase information preserved in the interference between acoustic waves reflected from the substrate surface as well as internal reflections from the acoustic lens. A stack of images were taken with the focus point of acoustic lens positioned at or above the substrate surface, and processed pixel by pixel using custom software developed with LABVIEW and IMAQ (National Instruments) to extract phase information. Scanning parameters, such as acoustic wave frequency and gate position were optimized to get reasonable phase and lateral resolution. The contribution from substrate inclination or uneven scanning surface was removed prior to further processing. The wave attenuation was also obtained from these images.

Dentin matrix protein 1 (DMP1) is an acidic noncollagenous protein which plays an important role in mineralized tissue formation. Dmp1 null adult mice are ricketic and osteomalacic and are a model for hypophosphatemic rickets [1]. Mutation in humans results is Autosomal Recessive Hypophosphatic Rickets [1]. The degree of bone mineralization significantly contributes to bone tissue mechanical properties, but precise relationships and interactions between chemical and mechanical variables are unknown. The objective of this study was to relate the differences in chemical properties in the Dmp1 wildtype (WT) and null (KO) mouse femoral cortical bone to their mechanical properties by using FTIR imaging and Scanning Acoustic Microscopy (SAM). Interactive mechanical (elastic modulus) and chemical images (i.e., mineral/matrix ratios) were generated from the same region of bone at a lateral resolution of ˜10 um. Mechanical analysis showed that elastic modulus, 75 percentile around 7.1 GPa, was ˜60% less in Dmp1 KO than that in WT, in which elastic modulus was 75 percentile around 15.2 GPa. The mineral-to-matrix ratios in Dmp1KO (4.96±1.63) were ˜2 times lower than that in Dmp1 WT (8.65±1.14). The mineral crystallinity and collagen crosslink ratios were not significantly different between KO and WT. Conclusions: The results relate the bone elastic modulus changes in Dmp1WT and KO mice with chemical changes within a specific bone site. These measurements provide a new tool for describing the variability of bone chemical and mechanical properties.

Non-destructive investigation of buried interfaces at high-resolution is critical for integrated circuit and advanced packaging research and development. In this letter, we present a novel non-contact microscopy technique using ultrahigh frequency (GHz range) longitudinal acoustic pulses to form images of interfaces and layers buried deep inside a silicon device. This method overcomes fundamental limitations of conventional scanning acoustic microscopy by directly generating and detecting the acoustic waves on the surface of the sample using an ultrafast pump-probe optical technique. We demonstrate our method by imaging copper lines buried beneath a 6-μm silicon wafer; the lateral spatial resolution of 3 μm is limited by the laser spot size. In addition to the high lateral spatial resolution, the technique has picosecond (ps) time resolution and therefore will enable imaging individual interconnect layers in multi-layer stacked devices.