Microstructural Characterization of Looseness in Bovine Leather using Ultrasound

Looseness is defect that affects the quality of a significant proportion of leather, particularly bovine leather. This defect causes a wrinkled appearance on the leather surface and results in a significant reduction in the leathers value. The structural mechanism and causes of looseness in leather are not well understood; therefore this study aims to gain a better understanding of this. We have used small angle X-ray scattering, ultrasonic imaging, electron microscopy and tensile testing to characterize and compare the structures of loose and tight bovine leather. Loose leather appears to have more highly aligned collagen fibrils, resulting in less overlapping of fibrils and weaker connections between the layers in leather. Loose leather has a looser fibre packing arrangement, with larger gaps in between fibre bundles, particularly in grain-corium boundary region. We are able to see this characteristic loose region using in-situ ultrasonic imaging and electron microscopy. We have used a range of techniques to gain a better understanding of the physical basis of looseness. This knowledge will be useful for further studies into the cause and prevention of looseness from developing during leather production. 1 – Introduction Around 5-10% of bovine leather produced displays a defect known as looseness. The defect looseness appears as undesirable wrinkles on the leather surface and results in a significant reduction of value of the affected hide. An entire finished leather is deemed rejected when looseness appears more than half-way up one side of the leather, and the leather becomes waste. The test for looseness involves a simple break test that occurs in the latter stages of the processing line, where an operator folds a piece of leather with the grain side in to observe and rank the surface wrinkles. The significant proportion of reject leather in the late stages of processing results in a significant waste of processing time and resources for leather tanneries. Despite the substantial loss in profit due to looseness, the industry has little understanding of the structure of loose leather and the causes. This work aims to gain more of an understanding of this. 2 – Materials and Methods 2.1 – Leather preparation Hides, weighing around 35kg, were obtained part-processed from either a local tannery or a pilot tannery. They had been green fleshed and processed using a conventional recipe based on sodium sulphide, sodium hydrosulfide and lime, where they were added to a 4 m diameter by 3.8 m wide drum along with the chemicals to depilate and open up the structure in order to removal of non-collagenous proteins. Around 8500 kg of hides were processed at a time and hides were removed from the rotating XXXIII IULTCS Congress November, 24 – 27, 2015 Novo Hamburgo/Brazil 2 drums after 8 h of liming. They were then lime split to a thickness of 2.5-4.0 mm. It was through the lime splitter that the hides with more “draw” than others were selected for further processing at the LASRA pilot tannery with the hope of selecting some hides that produced loose leather for analysis. A conventional recipe for shoe leather production was used for onward processing at the LASRA pilot tannery. Selected hides were sammed and sett, toggle dried at 40°C and conditioned prior to Molissa staking. 2.2 – Looseness Evaluation The standard break test method was used to evaluate looseness in the crust leather. This test involves folding the leather inwards grain side up and scoring the size of the surface wrinkles against a standard break scale. From the prepared leather, three crust leathers were selected and divided into five regions. Samples were taken from each region. Two operators measured and scored each leather sample and the results were averaged. From these scores the samples were deemed either loose or tight. Samples were considered tight if a break score of three or less was given, and loose if they had a break score of five or greater. Samples between 3 and 5 were not evaluated in this study. 2.3 – Ultrasonic Imaging A Dermascan C USB instrument (Cortex Technology, Denmark) was used for the ultrasonic measurements. A 20 MHz 2D-scanning probe with an internal water chamber was used to carry out the scanning at 6-8 frames per second. To achieve good image quality and minimize signal attenuation the scans were carried out under water and after the leather had been soaked for at least 24 h. Calibration of the sound velocity in leather was carried from the time of signal reflection and found to be 2561 m/s. The set bandwidth is 0.75 which results in an axial resolution of 97μm. The lateral resolution is set according to the mechanical scanning of the transducer at 150 μm. The focal point sits at a depth of 13 mm which falls approximately within the leather sample. A custom gain profile was created and the gain level was used for smaller adjustments of the signal amplification to improve signal penetration and image quality. The custom gain profile had a minimum sound intensity of 21 dB near the grain surface, increasing steadily to 42 dB within the corium. 2.4 – Scanning electron microscopy Samples were fixed for 8 h in a modified Karnovsky’s fixative at room temperature. This was followed by three washes in phosphate buffer and dehydration in a graded ethanol series. Samples were critical point dried using the Polaron E3000 series II critical point drying apparatus with liquid CO2 and ethanol. Samples were then mounted on to aluminium stubs and sputter-coated with gold (Baltec SCD 050 sputter coater) before being viewed in the FEI Quanta 200 Environmental Scanning Electron Microscope at an accelerating voltage of 20kV. 2.5 – Small angle X-ray scattering The Australian Synchrotron SAXS/WAXS beamline was used to record diffraction patterns. The beamline has an energy resolution of 10 from an undulator source and a cyro-cooled Si (III) doublecrystal monochromator. The beam size is 250 x 150 μm full width half maximum (FWHM) at the sample, and has a photon flux of around 2 x 10 ph/s. Diffraction patterns were recorded with an exposure time of 1-2 s and an X-ray energy of 11 keV. A Pilatus 1M detector was used with an area of 981 x 1043 pixels, a sample-to-detector distance of 3371 mm and an active area of 170 x 170 mm. To calibrate the energy at 9.659keV, the zinc-K edge from zinc foil was used and maintained within 2 eV of the nominal energy. For data collection, the samples were mounted individually on a remotely controlled sample plate and placed in the path of the X-ray beam. Spectra were recorded from the grain to the corium at 0.2 mm XXXIII IULTCS Congress November, 24 – 27, 2015 Novo Hamburgo/Brazil 3 increments under constant humidity conditions. The SAXS diffraction data was processed using SAXS15ID software. To measure the spread of orientation of the collagen fibrils within samples the Orientation Index (OI) was calculated. OI was calculated from the range in azimuthal angle of the most intense Bragg’s peak at around 0.058-0.060 Å, and is defined as (90° OA)/90° where OA is the minimum azimuthal angle range that contains 50% of the fibril scattering intensity centred around the maximum peak intensity. An OI value of 0 indicates a completely isotropic orientation of fibrils within a material, where as an OI of 1 indicates completely parallel fibrils to each other. 2.6 – Tear and tensile test A standard method for double-edge tear testing was used to determine tear strengths . After being cut to size, samples were stored at 20°C and 650 g kg relative humidity for 24 h before tear tests were carried out using an Instron 4467. Thickness was recorded using method BS EN ISO 2589:2002. Six measurements were taken on both loose and tight leather, 3 being parallel and 3 being perpendicular to the line of the backbone of the animal. 3 – Results and Discussion Looseness in leather results from a structural deficiency which weakens the connections between layers within the leather structure. Until now, the details on the structure, and the region that the structural weaknesses occur, have not been known. Therefore we have investigated cross sections of leather with multiple structural characterisation techniques in order to gain a better understanding of looseness and how it occurs. Below are the results and structural characteristics observed from each technique: 3.1 – Break and Looseness Loose leather displays larger wrinkles than tight when folded inwards towards the grain. An example of break for a loose (break of 7) and tight (break of 1) is shown in Figure 1. below. Figure 1: Comparison of the break in a) tight leather (break = 1) and b) loose leather (break = 7). Photos were taken of the concave surface of the grain while simulating the break test. Scale bar 10 mm. © J Sci Food Agric DOI: 10.1002/jfsa.7392. XXXIII IULTCS Congress November, 24 – 27, 2015 Novo Hamburgo/Brazil 4 3.2 – Small Angle X-ray Scattering SAXS analyses of cross sections of loose and tight leather showed a difference in fibril OI. Loose leather had a relatively uniform OI throughout the thickness of leather in comparison to tight leather, which showed a larger change in OI between the grain and corium regions. In tight leather, the corium had a much lower OI, suggesting a more isotropic fibril arrangement. Overall, loose leather had a higher average OI throughout the entire thickness of 0.61 (σ = 0.01) compared with 0.43 (σ = 0.03) for tight leather, P < 0.001 for α = 0.05). 3.3 – Ultrasonic Imaging Clear differences can be seen between the ultrasonic images of tight and loose leather, as demonstrated in Figure 2. Figure 2: Ultrasonics image of a,c,e,g) tight leather and b,d,f,h) loose leather (scale bar 1 mm). The grain is at the top of each image and the image was taken from the grain side. Some of the corium (bottom of the leather) is not visible as the ultrasonic signal is greatly attenuated as it passes thought the sample positions of these samples with their looseness scores are tight: a(2), c(2), d(2) all OSP, g(2.5) upper axilla; loose b(7) belly, d(6) lower axilla, f(5) belly, h(6.5) lower axilla. © J Sci Food Agric DOI: 10.1002/jfsa.7392. XXXIII IULTCS Congress November, 24 – 27, 2015 Novo Hamburgo/Brazil 5 We observe a band region of low intensity in the scans taken of loose leather, wet blue and pickle that are not apparent in the tight materials. This band region could represent a less dense region of collagen fibres, with space or gaps existing between collagen fibres. The gap regions shown in the images would be filled with water due to the images being taken while the samples were submerged under water, however usually these gaps would be air gaps. Tight leather appears to have a very uniform and dense packing of fibres in the grain and grain-corium regions.