The influence of the treatments with Na2CO3 and BTCA on the shives SHI-C and SHI-BTCA microstructure is clearly visible on the nano-CT scans.The removal of a part of the hemicelluloses in the case of sample SHI-C led to the partial deconstruction of the woody fibers cell wall,and general disorganization of the cells within the tissue,which is directly observed through the misorientation and waviness of the woody rays.The fiber walls were also strongly affected by grafting ,the woody rays degraded,the vessels distorted,and the tissues got fully disorganized as shown in Fig.3c.Indeed,the cellulose compartment seems the most affected,with a strong decrease,whereas the lignin content was increased,reflecting an enrichment effect.One can note that the pectin content is also higher in the BTCA sample,suggesting a selective chemical targeting of the treatment on cellulose polymers Fig.3.also shows EDX spectra of SHI-W,SHI-C,and SHI-BTCA samples before copper adsorption,highlighting the presence of sodium on the surface of hemp shives after Na2CO3 treatment and BTCA grafting.Fig.4 presents the X-ray nano-CT images and EDX spectra of the raw and modified hemp shives after copper adsorption.No significant effect of the exposition to copper on the shives’ microstructure can be detected but a major change in the EDX spectra can be observed.The data point out the presence of peaks that were not observed before the shives immersion in the copper salt solution,flood and drain tray which was assigned to copper.It is worthy to remind that EDX is a technique for surface characterization with limited penetration in matter thickness.

To better identify the spatial localization of the copper in the whole shives’ volume,a density-based segmentation of the 3D computed nanotomography images was done.The obtained images are proposed in Fig.5.The figures 5 b and d show the typical microstructure of hemp shives where the vessels are isolated or grouped by two or three,rarely by more,and then they deform one another.The vessels have a quite thin cell wall and a diameter of approximately 50 to 150 μm.They are surrounded by relatively thick-walled woody fibers with diameters of only a few μm and an irregularly polygonal section with a rounded cavity.It is evident that copper is adsorbed on all the free surfaces of the SHI-WCu sample,meaning shives external surface but also on the internal surface of the cells’ lumen.The situation is something different in the case of SHI-CCu shives where it is possible to observe that the copper is not only located on the internal surface of the cell wall bordering the lumen but also in the wall itself.This is attributed to the partial removal of the hemicelluloses in the cell wall induced by the treatment with sodium carbonate and by the micro- and nano-porosities created in the cell wall in which the aqueous solution can diffuse.Furthermore,the copper is massively adsorbed on the internal surface of the vessels’ wall bordering the lumen of the sample SHIBTCACu.The BTCA presence allowed more copper to be absorbed in these areas when compared to raw shives,which is in accordance with the results presented in Fig.1.The detailed analysis of EDX and nano-CT results showed that these two techniques were complementary in revealing the mechanism of copper adsorption on materials.In the EDX spectra ,we can observe an important decrease in the sodium peak and its replacement by copper cations,suggesting the presence of interaction by ion-exchange.In addition,some copper agglomerates are also observed in the internal part of the wall of some of the vessels,suggesting micro-precipitation during the adsorption onto the SHI-BTCA sample.So,the results point out that,even if the two treatments performed similarly in terms of copper removal,the spatial localization of copper and the adsorption mechanisms were significantly different.

FTIR and Raman spectroscopy were further used for the characterization of the shives before and after copper adsorption.The main IR and Raman absorption bands and their assignments according to the literature are summarized in Table 2 and Table 3,respectively.These bands were generally attributed to the three main shives components: cellulose,hemicelluloses,and lignin.The comparison of SHI-W and SHI-C samples shows an important difference in their IR spectra.Namely,the bands at 1730 cm− 1 and 1250 cm− 1 assigned to the stretching of unconjugated C=O groups present in hemicelluloses disappeared after the activation with sodium carbonate.This is in agreement with the data of the chemical composition and is also observed after the partial hemicellulose removal from jute fibers using alkaline treatment.When the FTIR spectrum of the SHI-BTCA sample was observed,the bands in the region 1500-1800 cm− 1 in particular those of the carboxyl and carboxylate groups ,are more intense than in the case of the two other samples which is ascribed to the esterification reaction.In addition,the comparison of SHI-W and SHI-BTCA samples also shows an important increase of the band at 1730 cm− 1 and a decrease of the band at 1250 cm− 1 corresponding to the elimination of hemicelluloses enhancing the lignin content in the material.The interpretation of Raman data indicated that the SHI-W and SHI-C spectra were similar with little differences in the intensity of the bands.On contrary,the Raman spectrum of the SHI-BTCA sample showed an important change in the intensity of the ester band at 1743 cm− 1.The comparison between the FTIR and Raman spectra before and after copper adsorption for the studied samples is shown in Fig.6.After copper adsorption,no changes were observed in the IR spectra of SHIWCu and SHI-CCu,while the spectrum of SHI-BTCACu showed an important change at 1590 cm− 1.Indeed,this band underwent a profound change in both intensity and width,and shifted to 1610 cm− 1.This data reflected the implication of the carboxylate groups on the copper removal through complexation and/or ion-exchange.By comparing the samples’ Raman spectra,we observed that the alkaline activation resulted in the disappearance of the ester band at 1743 cm− 1.The addition of copper contributed to obtain samples with a very high sensitivity to burning,resulting in spectra with enhanced background noise.

This is the reason why we had to work at 1% laser to avoid such noisy phenomena.However,it was difficult to explain the difference in the ratio between polysaccharides and lignin.In this case,we observed more lignin signals in samples containing copper.SHIBTCA gave very fluorescent samples,around 60,000 counts in comparison to 10,000 counts or less for SHI-C.For this sample,we also worked with a 1% laser both to avoid the detector’s ’ saturation and to burn the copper containing sample.The Raman spectrum for SHI-BTCACu also showed the involvement of the ester group in the removal of copper ions,confirming the presence of various interactions between them.Finally,the Raman spectra of samples after copper adsorption also revealed two bands at 1420 cm− 1 corresponding to carboxylate group and at 885 cm− 1 ,in agreement with IR data.Fig.7 compares the XPS wide-spectra of the SHI-W,SHI-C,and SHIBTCA samples before and after copper adsorption.The spectra obtained before copper adsorption are very similar showing the presence of oxygen and carbon C1s at their surfaces.Nevertheless,hydroponic tables canada the sample SHI-BTCA has an additional peak corresponding to the presence of sodium localized near the surface at 1072 eV.The presence of this element is due to the conversion of the carboxylic groups into a carboxylate by immersing the BTCA treated shives in a sodium hydrogen carbonate solution.The 1s level of carbon allows access to the different oxygen functions.The decomposition of level 1s of carbon is significantly different for sample SHI-BTCA with a strong decrease in C-O / C-OH bonds in favour of carboxylic functions.These results are in agreement with the data of the chemical composition.After copper adsorption,Fig.7 clearly indicates the disappearance of the sodium signal in the SHI-BTCA spectrum,due to the ion-exchange with copper cations.Fig.8 shows the main and satellite peaks of Cu 2p3/2 and Cu 2p1/2 after copper adsorption onto three studied samples,while Table 4 reports atomic concentrations of detected elements before and after adsorption.All three samples showed XPS peaks around 933 eV and 953 eV characteristic for Cu 2p3/2 and Cu 2p1/2,respectively.

The XPS analysis of the Cu 2p peaks showed that this element was not in the same oxidation state in all studied samples.Indeed,the detection of satellites on the 2p level of copper for the SHI-BTCACu sample makes it possible to deduce also the presence of copper at the oxidation number for 60%.These satellites are associated with the peak at 934.66 eV of the 2p3/2 level which corresponds to copper in the form of Cu2 [33].In the case of SHI-CCu and SHI-WCu,the detected copper was only in the I oxidation state.Indeed,there are no satellites,the 2p3/2 level showed only one component at about 933 eV.The measurement of the kinetic energy of the Auger transition of copper L3M45M45 at approximately 913.7 eV allowed us to calculate a modified Auger parameter value at 1846.7 eV close to the species Cu2O.This shows that the reactions involved in the removal of copper were different for each hemp sample.In order to know the rate of adsorbed copper on the surface,the atomic percentage of copper was calculated from the area of the Cu 2p peaks.SHI-BTCACu sample had the highest ratio Cu/C = 8.3,followed by SHI-CCu which ratio is 1.5.SHI-WCu sample had a very low Cu/C ratio of 0.4.These results agree with those obtained with other characterization techniques,notably on the fact that copper is well absorbed at the surface for SHI-BTCACu,which gives it this green colour.The less important Cu/C ratio for SHI-CCu confirms that,in this case,the mechanism of copper adsorption is different from those of SHI-BTCACu.Namely,the main interaction between copper and sodium carbonate activated shives is diffusion inside the particles.XANES spectroscopy provides quantitative insight into the oxidation states of copper present in materials.The XANES spectra of SHI-CCu and SHI-BTACCu display the same shape with a pre-edge at 8977.5 eV,the edge at 8991 eV,the maximum of the white line at 8996.5 eV,and a broad oscillation centered on 9052 eV.The absence of peak around 8983 eV,as found in Cu2O,shows that Cu was not detected and that Cu oxidation did not change during the Cu adsorption,it was kept as Cu.CuSO4,5H2O spectrum shows specific features such as a shoulder at 8990.0 eV and an oscillation at 9008.0 eV that are not present in hemp samples’ spectra.Moreover,the white line of CuSO4,5H2O spectra is slightly shifted toward lower energies and is more symmetric compared to that in hemp samples’ spectra.

These differences indicate that after adsorption onto hemp,Cu did not correspond to the crystalline form of CuSO4,5H2O.Furthermore,the hemp samples’ spectra are quite different from those of CuO and Cu2,corresponding to other Cu-bearing precipitates.The spectra of hemp samples are similar to the ones of Cu2+ measured in liquid phase; CuSO4 but also Cu2,another Cu complex,both when recorded in a liquid phase.The results obtained by XANES and those obtained by XPS or both Cu and Cu seems to differ.However,they are compatible and complementary as XANES at Cu K-edge sounded the bulk sample whereas XPS sounded the first atomic layers on the surface of the sample.XANES spectroscopy evidenced that most of the Cu atoms are present in Cu form in both SHI-CCu and SHI-BTCACu samples and XPS spectroscopy indicated that the external surfaces of SHI-WCu and SHI-CCu are coated by Cu2O and that the external surface of SHIBTCACu is coated by a mixture of Cu2O and Cu2,explaining this characteristic blue color.An increased interest in the acceptance of novel foods,that is,foods to which a consumer has not been previously exposed ,is borne of an increasing global population that should reach nine billion by the year 2050.Globalisation has resulted in increased exposure of the world’s population to foods from other cultures,and increased multicultural culinary experiences have altered normative perceptions for many.The increasing global population presents a challenge to food security as many traditional food production techniques will become environmentally unsustainable at the levels required to meet world food demands.Hemp foods might suitably address many food security issues.The hemp plant,Cannabis sativa,from which hemp foods are produced is environmentally sustainable due to a reported low water need and natural pest resistance ,can be economically lucrative with high yields and shorter growth cycles compared with many traditional crops ,and has many nutritional benefits.Nutritional benefits of hemp include being high in levels of plant protein,high in dietary fibre,a rich source of Omega 3 and 6,and contains all of the amino acids essential to human life.