Among hemp samples, Hemp 5 displayed the highest WHC and HWHC, which could be attributed to its higher protein content and denaturation/aggregation occurring during its double AE-IP. In fact, Hadnađev et al. showed that AE-IP hemp protein showed a two-fold higher WHC than their counterparts obtained through micellisation. On the contrary, it seems that pea protein lost its ability to hold water upon heating likely due to the exposure of more hydrophobic groups on the protein surface.This process occurs upon heating and subsequent cooling of a concentrated protein solution. In fact, to create a heat-set gel, proteins should be unfolded upon heating so that both polar and non-polar residues are exposed and able to form hydrogen bonds and hydrophobic interactions during the cooling process. In this process, the rate of protein aggregation and network formation is higher than that of unfolding . Therefore, proteins with a high number of non-polar residues can form the network faster . In this respect, Hemp 2, with the highest H0, exhibited the lowest LGC . Meanwhile, Hemp 5, despite having the highest protein content among hemp samples, showed one of the highest LGC, which could be explained by its significant denaturation/aggregation . The low LGC of Hemp 1 could be explained by its highest content of carbohydrates , sub irrigation cannabis which could have hindered its ability to form a stable gel at lower concentrations. HMMAs were made considering Hemp 3 and Hemp 5 samples as representatives for dry- and wet-fractionated hemp protein samples, respectively.
Low-field nuclear magnetic resonance has been recently utilized as a fast and non-destructive method to study the distribution and mobility of water in food matrices such as meat and meat analogues . In this technique, a sample is subjected to radio frequency pulses inside a magnetic field and the water or lipid state in the sample can be assessed based on the relaxation time of proton nuclei of water molecules . The relaxation time of the protons indicates how tightly or loosely water molecules are bounded to the solid phase of the matrix. In general, an NMR decay curve can be fitted to three or more exponentially decaying compartments , showing the water distribution in the sample. Accordingly, a relaxation time curve provides two different parameters for the water state in a food matrix, namely water mobility and the amount of water distributed within each mobility component . Fig. 5 shows the T2 relaxation time curve for pea, Hemp 3, and Hemp 5 HMMAs, and four distinct peaks roughly centred at 1–2 ms , 10–14 ms , 45–50 ms , and 138–220 ms were observed. The first water population with the lowest relaxation time reflects those water molecules closely bound to proteins or polysaccharide molecules . In this respect, Hemp 3 extrudate displayed the lowest absolute area in this region probably due to the higher number of hydrophobic patches, i.e., higher surface hydrophobicity , which could expel water upon protein structuring on fibrous structures. The second two compartments, T21 and T22, represent the immobilized water entrapped in protein or polysaccharide gel structures. As can be seen in Fig. 5 and Table 4, pea extrudate had the highest absolute area in T21 and a small peak in T22. This occurrence indicates that most of the water in this sample is entrapped in a protein network, as opposed to being immobilized in a polysaccharide network, which is consistent with the fact that pea displayed the highest amount of protein and the lowest amount of TDF . Hemp HMMAs, with lower protein and higher carbohydrate contents, showed shorter relaxation time and higher absolutes areas of T22 peaks than pea, respectively , indicating more water being entrapped by the polysaccharides.
It is noteworthy that the main peak T21 is sensitive to the homogeneity of the water in the samples , increasing its susceptibility to become broader in compositionally heterogeneous samples . As an example, Aursand, Gallart-Jornet, Erikson, Axelson, and Rustad observed that the addition of salt to cod and salmon meat leads to peak broadening due to the salt hydration effect. This occurrence would explain the T21 peak broadening of Hemp 3 and Hemp 5, which possessed more ash, lipids, and TDF than pea . Lastly, the third region of relaxation time curve reflects the free water in the system, which is the fraction that can be lost upon the cooking process of meat. Typically, this water population appears above 100 ms in an animal meat system . Thus, the longer relaxation time of T23 region shown for the HMMAs indicates that water is more prone to leave the plant protein structure, which more closely resembles the process that occurs in animal meat. In this regard, Hemp 3 HMMA would be the best choice from a free water standpoint, since it had the highest absolute area in this region, followed by Hemp 5 and pea . Tensile and small amplitude oscillatory rheological experiments were performed to evaluate the mechanical properties of the plant-based HMMAs and their relationship to their visual appearance and water mobility. HMMA made with hemp exhibited a remarkably greater fibrousness, following the order of EHemp5> EHemp3 >EPea . It was also observed that the extrudates made of EHemp5 presented longer and more defined fibres that the EHemp 3 counterpart. Meanwhile, EPea exhibited a more isotropic structure, where fibres were hardly visible. The visual appearance of the samples was in agreement with the tensile tests performed parallel and perpendicular to the fibre direction, which highlighted a significantly higher anisotropic index for EHemp5 . It is noted, however, that higher anisotropy also resulted in higher sample hardness .
We note that the significantly higher proportion of less interacting water protons with macromolecules of EHemp5 compared to EPea could suggest that the marked phase separation leading to fibrous-like structures could also result in higher proportions of free water . EHemp3 did not follow this rule, which could be likely attributed to peak broadening during LFNMR analysis. Thus, fibrousness, hardness and water mobility should be jointly considered when designing fibrous HMMA. The comparison between SDS-PAGE at reducing and at non-reducing conditions of hemp HMMAs revealed some bands representing edestin that were only present using DTT, in contrast to EPea, whose bands were less intense both with and without DTT. This provides evidence of the higher contribution of S–S bonds in hemp protein aggregation during extrusion than in pea protein. This occurrence aligns with previous reports showing the relative importance of physical and covalent interactions between plant proteins, where S–S bonds were key for anisotropy formation using gluten , and non-hydrophobic and ionic interactions were the main contributors of gel strength using pea . Due to the protein unfolding occurring during extrusion, the hidden non-polar and sulfhydryl groups are exposed to the surrounding aqueous phase. This phenomenon might cause the formation of new assemblies via non-covalent interactions and disulphide bonds . Hemp protein is richer in sulphur containing amino acids than pea , which suggests the higher contribution of S–S bonds to form anisotropic structures under laminar flow during cooling of Hemp 5, and could also explain the higher anisotropy of EHemp5, followed by EHemp 3 and EPea. Another explanation could be given by the higher surface hydrophobicity of hemp protein concentrates than the pea counterpart . The combination of highly hydrophobic hemp proteins with the hydrophilic maize starch could have promoted the formation of an optimal multiphasic system during extrusion, where the dispersed phase could have been elongated under tensile and shear stresses into visible fibres in the cooling die. This colloidal approach has been suggested before , although the relative importance of the molecular model and the colloidal approach are still the subject of debate. The differences in the fibre-forming ability between EHemp3 and EHemp5 should be ascribed to their different protein purity .
The high level of lipids in Hemp 3 protein concentrate could have also played a role in the flowability and shear and tensile stresses of the melt inside the extruder, lowering the SME response during processing from 325 kJ/kg for EHemp5 to 285 kJ/kg for EHemp3. The content of phytic acid in dry-fractionated hemp was found in the upper part of the typical high valuesreported for oilseeds, vertical grow which was in turn higher than the content typically found in cereals and legumes . For this reason, the content of phytic acid was also measured in HHMAs. The content of phytic acid naturally decreased in HMMA in proportions generally explained by the dilution effect of the protein concentrates with starch and water. Therefore, and as expected, extrusion did not reduce the content of phytic acid due to its known high thermal stability . Likewise, Pontoppidan, Pettersson, and Sandberg already reported that the degradation of myo-inositol hexakisphosphate in feedstuffs during extrusion is too limited to have any nutritional effect on the availability of phosphorous and minerals. From the phytic acid perspective, if our developed HMMA prototypes were to be consumed as they come out from the extruder, that is, without further water removal or water addition , a 100 g single portion would contain a total of 0.6–0.7 g of phytic acid in EPea and EHemp5, and 1.8 g of phytic acid in EHemp3. Remarkably, the consumption of EPea and EHemp5 would not necessarily increase the daily intake of phytic acid in consumers following a plant-based diet, reported as >1000 mg . On the other hand, the consumption of EHemp3 could potentially result in a higher average daily dietary phytic acid intake in humans following a plant-based diet living in certain European countries, such as Sweden . It is noteworthy that the inhibition of the intestinal mineral absorption can be counteracted by many dietary compounds such as organic acids and complexing agent like ascorbic acid, competing with phytic acid for mineral binding. Thus, little evidence exists that in well-nourished population groups dietary phytate may seriously affect the status of iron, zinc, and calcium . On the other hand, phytate may lead to serious deficiencies with unbalanced nutrition or undernourishmentin developing countries. In this case, dry-fractionated protein concentrates from different botanical original than hemp, or wet-fractionated hemp samples, would be preferred. At any event, the consumption of dry-fractionated hemp HMMAs would result in average daily dietary phytic acid intakes within the range found for adults living in developing countries . Industrial hemp is an annual, bushy, dioeciously or monoecious herbaceous plant which belongs to the family of Cannabaceae, genus Cannabis.
The industrial and medical importance of Cannabis sativa L. is reflected in the increasing market interest for consumables, fiber, medicine and miscellaneous products derived from raw or dried cannabis plant parts. The moisture removal of the harvested goods is one of the most important unit operation, in which the moisture is reduced to a certain balance level that the deterioration and microbial spoilage are minimized. In order for efficient, feasible and properly dimensioned convective drying systems to be deployed, accurate modeling of each raw material behavior during the drying process is required. Several mathematical thin-layer drying models are proposed in the literature for describing the drying kinetics in a range of agricultural products. More sophisticated approaches of drying scheme implementations arise in the scope of improving drying time, cost and energy, since the convective drying is an energy-intensive and slow process when low temperature levels are applied. Non-isothermal drying, implementing a variable mode of drying conditions by modifying as a function of time, one or more drying parameters is a technique followed by many researchers, also applied for herbaceous materials, as a nonstandard drying process aiming to reduce energy consumption, improve the product quality or reduce the required drying duration. In a study presented by Ozguven et al., peppermint was dried at two constant drying temperatures of 35 and 55 ◦C and two incremental temperature rises of 5 and 2.5 ◦C⋅h-1. Both non-stationary temperature profiles had a shorter drying duration a lower final moisture content compared with the lowest constant temperature level applied. A similar study was performed by Tarhan et al. forlemon balm .