Rtespa 525

A JPK AFM Nanowizard Cellhesion 200 (JPK Instruments AG, Berlin, Germany) interfaced with a motorized XY stage was used to conduct nano-indentation measurements and extract stiffness maps. A stiff cantilever with a sharp tip (RTESPA-525, Bruker, nominal spring constant of 200 N/m) was used. Exact cantilever spring constant was determined using a direct method of a reference cantilever. The AFM setup was mounted on an Olympus inverted microscope and a CCD camera allowed monitoring of the position of the cantilever over defined regions of the sample. Forcedistance curves were taken with an approach speed of 2 µm/s and a set force of 200 µN by an automated raster scan using a motorized stage with 10 µm step size. Prior to indentation tests, the sensitivity of the cantilever was set by measuring the slope of a force-distance curve acquired on Sapphire substrate (PFQNM-SMPKIT-12M kit, Bruker). Using JPK Data Processing software, the contact point was found. The stiffness values Stiffness=E/(1-ʋ 2 ) were extracted from the forcedistance curves by fitting the contact portion of curves to a Hertz contact model between a sharp triangular tip and an infinite half space, where E is the elastic modulus and ʋ the Poisson ratio as previously demonstrated in literature [36] .

Similar to the FIB-SEM specimen-preparation procedure, eight embedded specimens were mechanically polished using diamond lapping films (3M) and argon-ion polished (SM-090101 Cross-Section Polisher, JEOL). The SPM system employed was a Bruker MultiMode 8 SPM equipped with a Nanoscope V controller and Nanoscope analysis software (Bruker Nano Surface Business, Santa Barbara, CA, USA). Cross-sectioned surfaces were quantitatively characterized using the PeakForce QNM (Quantitative NanoMechanics) SPM module under ambient conditions. Elastic modulus (EM) was calculated by Nanoscope analysis software using the Derjaguin-Muller-Toropov (DMT) model (Derjaguin et al., 1975; Wang et al., 2013) . Samples were scanned using a probe (RTESPA-525, Bruker) with a nominal radius of 8 nm and a nominal spring constant of 200 Nm -1 with 0.501 Hz scan rates, a 250-mV amplitude set point and a 150-nm peak force amplitude. After SPM, the samples were observed by SEM (JSM-6701F, JEOL).

Mechanical characterization was performed with a Multimode AFM instrument (Bruker Corporation, USA) using PF-QNM imaging mode and a RTESPA-525 (Bruker) probe with a spring constant of 139 N/m. The probe was calibrated using the so-called Sader method (http://www.ampc.ms.unimelb.edu.au/afm) with a Scanning Electron Microscope (Jeol JSM 6460LV) for measurements of the cantilever length and width. The tip radius, 32 nm, was tuned during measurement on a Highly Oriented Pyrolytic Graphite (HOPG) standard from Bruker, providing an indentation modulus of around 18 GPa. The applied maximum load was set at 200 nN for all the measurements. As the measurements are longer and more subjected to artefact and drift with this technique, a reduced number of fibres have been measured compared to nanoindentation (i.e., 5 developing flax bast fibres, 5 mature flax fibres and 5 aramid fibres).