1：Experimental Design and Method Selection:

Deuterium exchange combined with near infrared spectroscopy was used to study the roles of accessible and inaccessible cellulose in the load transfer of eucalyptus wood. Monitoring the drying process helped to assign bands of deuterated wood samples. Polarized NIR spectra of protonated and deuterated samples confirmed that inaccessible hydroxyl groups in eucalyptus wood were preferably oriented in the longitudinal direction. The spectral changes on NIR spectra caused by mechanical strain could be highlighted by averaging loading and unloading cycles to compensate for effects of desorption and isotope re-exchange due to environmental fluctuations.

2：Sample Selection and Data Sources:

Five thin Eucalyptus regnans samples with a thickness of 0.5 to 1 mm in preferably radial direction were cut from sapwood of a quarter-sawn board with an air-dry density of 0.54 g/cm3 sourced from Canterbury, New Zealand.

3：5 to 1 mm in preferably radial direction were cut from sapwood of a quarter-sawn board with an air-dry density of 54 g/cm3 sourced from Canterbury, New Zealand. List of Experimental Equipment and Materials:

3. List of Experimental Equipment and Materials: Transmission NIR spectra were acquired using a Tensor 37 infrared spectrometer (Bruker, Germany) at a resolution of 8 cm-1 averaging 128 scans before and after deuteration. Deuterium oxide (99.9%, Sigma Aldrich, Australia) was used for deuteration.

4：9%, Sigma Aldrich, Australia) was used for deuteration. Experimental Procedures and Operational Workflow:

4. Experimental Procedures and Operational Workflow: Samples were conditioned to constant weight in a sealed container over a saturated K2CO3 solution, equating to a relative humidity of 42.3%. Subsequently, four samples were oven dried at 103 °C before being immersed in a sealed container containing 25 mL deuterium oxide at room temperature for 7 days. One of the samples was dried in a container over phosphorus pentoxide (P2O5) to 0.8% while spectra were taken at different drying stages to monitor the drying process. The other three samples used to study the effect of polarization and mechanical stretching were dried over P2O5 close to FSP (monitored by sample mass) and then equilibrated over saturated K2CO3 solution in D2O.

5：3%. Subsequently, four samples were oven dried at 103 °C before being immersed in a sealed container containing 25 mL deuterium oxide at room temperature for 7 days. One of the samples was dried in a container over phosphorus pentoxide (P2O5) to 8% while spectra were taken at different drying stages to monitor the drying process. The other three samples used to study the effect of polarization and mechanical stretching were dried over P2O5 close to FSP (monitored by sample mass) and then equilibrated over saturated K2CO3 solution in D2O. Data Analysis Methods:

5. Data Analysis Methods: NIR spectra were normalized using the “scale” function in R. “Slope” spectra were defined as the rate of change of the intensity with increasing tensile strain and calculated as the slope coefficient of the linear regression between normalized spectral intensity at each wavenumber and applied strain.