Characterization Of Multi-layered Fish Scales (Atractosteus spatula) Using Nanoindentation, X-ray CT, FTIR, and SEM

1. Fish Scale Sample Preparation

For this study, scales were obtained from the US Army Engineer Research and Development Center (ERDC) Environmental Laboratory at mid-length (29^th caudal column) from an approximately 600 mm long gar (A. spatula). The fish scales were obtained according to the ERDC and National Institute of Health (NIH) animal care guidelines.

1. Materials
   Record the spatial location on the fish of the scales obtained for the study. Make sure to adhere to the appropriate organization or governmental guidelines for obtaining biological specimens such as the NIH animal care guidelines. Store the scales in an adequate medium such as phosphate buffered saline solution to preserve hydration and maintain mineral content once they are removed from the fish. Avoid prolonged storage that can result in mineral loss, which may influence the nanoindentation data. Use a medium bristle brush and tweezers to remove any soft tissue from the hard scales.
2. Specimen Mounting and Sectioning
   Examining a transverse section of the fish scale’s short-axis (Figure 1) using FTIR and nanoindentation requires first mounting the scale in a stiff medium such as a two-part epoxy consisting of a resin and hardener. Use general-purpose RT curing epoxies with low peak temperatures such as the commercial general-purpose epoxy used in this study that had a peak temperature of less than 55 °C.

Figure 1. X-ray CT images of A. spatula scale depicting the short-axis transverse section examined in this study of A. spatula using nanoindentation and FTIR [A (anterior), P (posterior), D (dorsal), V (ventral)].

2. 1. Hold the fish scale in a 32 mm diameter sample mold using a commercially available plastic sample holder. This keeps the sample oriented correctly while mounting in the epoxy.
   2. Once the sample is held in the mold, pour the uncured epoxy on the sample and then allow the epoxy to cure according to the manufacturer's directions.
   3. After the epoxy is cured, section the mounted sample using a diamond blade high-precision cut-off saw at the midline of the sample.
   4. Sonicate in distilled water for 15 min to remove any debris from the specimen.
3. Polishing for Nanoindentation and FTIR
   To obtain a smooth flat surface for nanoindentation as depicted in Figure 2, the following polishing procedure and parameters are suggested based on discussions with the polisher manufacture and test samples. However the parameters may need to be adjusted for different biomaterials based on responses such as material removal rates. Ultrasonication of samples in a bath of distilled water between polishing steps is vital to ensure particles from a coarser polishing step are not introduced into a subsequent finer polishing step.

Figure 2. Image of a polished short-axis transverse section A. spatula scale mounted in epoxy.

4. 1. Coarse polish with a 15 µm SiC pad and water as a lubricant until the sample is plane using the automatic polishing head force of 7 lbf and speed of 200 revolutions per min (rpm).
   2. Sonicate the sample in a bath of distilled water for 15 min.
   3. Intermediate polish with a 6 µm SiC pad using water lubricant at a platen speed of 130 rpm and a force of 7 lbf for 5 min.
   4. Sonicate sample in a bath of distilled water for 15 min.
   5. Polish with a 1 µm SiC pad using water lubricant at a platen speed of 130 rpm and a force of 7 lbf for 5 min.
   6. Sonicate the sample in a bath of distilled water for 15 min.
   7. Final polish with a 50 nm colloidal silica suspension using an appropriate polishing pad such as a high-density, non-woven, low-nap porous polyurethane that a manufacturer suggests for 50 nm suspensions. Polish at a speed of 130 rpm with a force of 7 lbf for 5 min.
   8. Sonicate the sample in a bath of distilled water for 15 min.

2. Nanoindentation Testing

1. Calibrate the nanoindentation system before each batch of testing per the manufactures guidelines. The calibration should include determining the system’s area function for the Berkovich tip and frame stiffness. Additionally, perform a microscope-to-indenter calibration at this step to ensure the indents correlate to the chosen microscope locations.
2. Load the sample into the nanoindenter and use the optical microscope controls on the nanoindenter to bring the sample into focus.
3. Use the software controls to move the specimen to the location for the first indent. Ideally, this is approximately 10 µm in the epoxy from the edge of the ganoine layer along the centerline of the scale’s cross-section.
4. Perform 4 parallel rows of indents spaced 15 µm apart to obtain a statistically significant data set starting at this location. Set the nanoindenter to a maximum load of 5 mN, loading and unloading rates of 0.1 mN/sec, a hold time of 30 sec, and a minimum indent spacing of 5 µm for each row. The row of indents should be setup to run orthogonal to the ganoine surface, and a sufficient number of indents should be specified to travel across the scale’s cross-section while traversing approximately 10 µm into the epoxy past the bony layer.
5. When the batch is completed, have the nanoindenter create fiducial indents with a maximum load of 100 mN at the first and last indent, which should be in the epoxy before the ganoine layer and after the bone layer, respectively. These correlate to the start and end points for each row of indents.
6. After nanoindentation, place the sample back in the PBS solution to avoid further dehydration.
7. Use the nanoindentation software to determine the modulus and hardness based on the Oliver-Pharr method¹³ if a time-independent material response is observed. Otherwise the hold-time may need to be extended to overcome the creep observed from unloading too quickly.

3. Spatially Resolved ATR-FTIR Spectroscopy

The use of a slide-on ATR accessory attached to an FTIR microscope is one suggested method to collect spatially resolved Fourier transform infrared (FTIR) spectra of the layers in a fish scale sample. The ATR accessory allows for collection of high quality spectra with very small (~10 μm²) spatial resolution, which is not attainable with any other FTIR technique. The same polished sample (Figure 2) prepared for nanoindentation experiments was used in these experiments.

1. Choose a sample with a surface and dimensions appropriate for the FTIR microscope being used for analysis to ensure high quality spectra is obtained from ATR-FTIR spectromicroscopy.
2. Prepare the FTIR microscope to collect data. FTIR microspectroscopy requires calibration of the FTIR signal under the same sampling conditions as will be used for measuring the sample. Typically, this includes cooling the detector and allowing time for it to stabilize as well as collecting all background spectra and sample spectra under the same environmental conditions. This can be especially important, because CO₂ and water vapor in air can dramatically affect FTIR spectra. It is also important to ensure that the instrument’s optics are aligned.
3. Collect an appropriate background spectrum to subtract the sample against. For these experiments, a polished, gold-coated microscope slide was used as a background for FTIR spectromicroscopy.
4. Using an appropriate objective, focus on the sample and select an area of interest for analysis.
5. Once an area of interest is found, attach the ATR accessory to the FTIR microscope objective, raise the sample until it makes intimate contact with the ATR internal reflection element, and collect a sample spectrum.
6. After collecting the FTIR spectra, perform the necessary standard data processing required.

4. X-ray Computed Tomography (CT)

1. Obtain and prepare scale as discussed in Section 1.1
2. Scanner Setup
   1. Warm-up the X-ray source according to manufacturer’s specifications.
   2. Set X-ray voltage and current to 50 kV and 160 µA, respectively.
   3. Set exposure time to 1,450 msec.
   4. Select a 1.0 mm aluminum filter.
   5. Prior to loading sample, make a flat-field correction when the X-ray source is off (dark-field) and on (bright-field).
3. Mount and Load Specimen
   Scales must be mounted in a manner so that they do not shift or move throughout the length of the scan. These samples must also be mounted using materials that are nearly transparent to X-rays. A combination of Styrofoam and Parafilm can be used to secure the scale to the CT stage.
   1. Rigidly mount the sample so that the longest dimension is parallel to the detector.
   2. Secure the mounted sample to the scanner stage.
   3. Position the sample so that it will be in the center of rotation throughout the scan.
   4. Select the highest resolution that allows the entire scale to be in the field of view (FOV), in this case 7.5 µm.
4. Acquisition Settings
   Perform scans in this study with a rotational step of 0.25° and a frame averaging value of 15. If lower resolutions are acceptable, increase the step size and/or decrease the frame averaging to reduce the total scan time.
5. Reconstruction Parameters
   Once a data set is obtained, reconstruct the X-ray projection images to create a data set containing cross-sectional images. Select Skyscan’s NRecon software default settings except for the following.
   1. Change the Ring Artifact Correction to 20.
   2. Change the Beam Hardening Correction to 25%.
   3. Adjust the CS Static Rotation to make cross-sectional image level.
6. Image Processing
   Use Skyscan’s CTAn software to obtain the final 3D gray-scale image. Adjust the gray-scale range to an appropriate level to remove the artifacts from the Styrofoam and Parafilm.

5. SEM Imaging and EDX Analysis

Specimens prepared by polishing for nanoindentation and micro-/nano-structure characterization were examined using scanning electron microscopy (SEM). Low-vacuum mode was utilized to minimize dehydration of specimens and the need for application of conductive coatings. Local chemical analysis was performed on polished specimens in conjunction with SEM imaging using energy-dispersive X-ray spectroscopy (EDX). EDX analyses were performed on the same line/grid that was analyzed by nanoindentation in order to provide correlations between chemical and mechanical properties. Freshly fractured surfaces were also examined by SEM to provide better information on the morphology and orientation of biomineralized structures present in the fish scales. To improve resolution for observation of nano-scale structure on fractured surfaces, specimens were sputter coated with gold (Au) and imaged in high-vacuum mode. The following provides additional details on the procedures used.

1. SEM Imaging of Polished Surfaces
   1. Place polished specimen into SEM chamber and pump chamber into low-vacuum mode with chamber pressure of 0.1 to 0.5 mbar.
   2. Adjust working distance to approximately 5.0 mm.
   3. Activate high voltage (HV) and navigate to region of interest on specimen that includes the transition zone between ganoine and bony sublayers or other areas of interest.
   4. Obtain images at 15 kV HV and beam current of approximately 3.9 nA.
   5. Focus image and perform all necessary alignments and stigmation adjustments.
   6. Capture images from at least three regions of interest at relevant magnifications (typically 250X to 10,000X) using the low-vacuum backscattered electron (BSE) detector to aid in identification of changes in biomineral content and density (i.e., dense bone vs. porous bone).
2. SEM Imaging of Fractured Surfaces
   1. Affix freshly fractured specimen to a 90° SEM stub using double-sided carbon tape with the fracture surface facing up.
   2. Sputter coat with Au to provide a sub-nm thick conductive layer on the fracture surface.
   3. Place specimen into SEM chamber and pump chamber into high-vacuum mode.
   4. Adjust working distance to between 3.0 and 5.0 mm.
   5. Activate HV and navigate to regions of interest on the specimen. Primary areas of interest in this case were the structure present in the ganoine and bony layers.
   6. Obtain images at between 5 kV and 15 kV HV and a lower beam current of 0.24 nA to improve resolution.
   7. Initially focus specimen and perform preliminary alignments.
   8. Increase magnification to greater than 5,000X and switch from normal field emission lens into immersion/ultra-high resolution (UHR) lens.
   9. Perform UHR alignments and stigmation adjustments.
   10. Capture images from at least three regions of interest at relevant magnifications (typically 5,000X to 250,000X) using the through lens detector (TLD) operated in secondary electron (SE) mode.
3. EDX analysis of polished surfaces (performed in conjunction with SEM imaging). These parameters are material dependent and will need to be adjusted so the EDX interaction volume is similar in size to the nanoindentation interaction volume as discussed by Moser¹⁴.
   1. Navigate to region of interest on polished specimen that includes nanoindentation grid indicated by fiducial marks at the end of each line of indents.
   2. Ensure HV is at least 15 kV, beam current is at least 3.9 nA, and working distance is greater than 5.0 mm.
   3. Capture BSE image of region to be analyzed using EDX.
   4. Using EDX analysis software, capture the same image to aid in locating areas to perform chemical analysis along the line of indents.
   5. Using the “Line Analysis” technique, position a line to perform chemical analysis along the line of interest of indents starting at the first indent and ending at the last indent.
   6. Specify the number of analysis points to be placed along the line. It is best to use the same number of analysis points and indents that are present to provide a direct spatial correlation between chemical composition and mechanical properties.
   7. When the line is positioned and points specified correctly, initiate the line analysis using EDX software.
   8. When the line analysis is completed, identify elements of interest to be quantified from the point spectra obtained along the specified line on the polished surface of the specimen.
   9. Once elements of interest are identified, perform a background calibration to account for Bremsstrahlung radiation and other effects.
   10. Choose the software’s deconvolution analysis option to obtain quantitative analysis on each point along the specified line to quantify the chemical composition at each point.
   11. Save quantitative chemical analysis results along with the image of specified line that was analyzed to aid in spatial correlation with mechanical properties measured using nanoindentation.