**4. Discussion**

In this study, pure and blended fuel pellets made from Norway spruce and pea starch were analyzed using advanced analytical instruments able to provide information beyond what is visible to the naked eye in order to establish di fferences in particle bonding mechanism relevant to quality. The results obtained were comprehensively interpreted from a structural chemistry perspective and used to answer questions relating to how particles of pure and blended pellets made from two di fferent biomass materials combine to form pellets, and the source of inter-particle bonding. As previously stated, assessing biomass for any conversion or pretreatment process requires an understanding of its basic composition [1,3,32]. Therefore, compositional analysis data determined that pure and blended samples of NSP and PSP assessed in this study contained various proportions of major organic and elemental constituents, which were vital to understanding the nature of the materials that were pelletized, and which were equally significant in comprehending how the major components combine to form pellets.

From the FT-IR data, it was established that the pure and blended samples of NSP and PSP contained varying concentrations of polar and non-polar functional groups that played a significant role in providing information related to di fferences in the mechanism of bonding and the type of attraction forces between individual particles relevant to quality in terms of strength only. It was determined that bonding in NSP (100%) was mostly due to van der Waals attraction forces, while, for PSP (100%), bonding was attributed to a combination of dipole–dipole and hydrogen bonding. In NSP/PSP (50%/50%), particles were held together by a combination of forces that included hydrogen bonding, dipole–dipole interaction, and slight van der Waals attraction forces. From previous studies [7,41,73,74], it was determined that the quality of biomass pellets, among other factors, is dependent upon the type and strength of attraction forces between individual particles; hence, FT-IR analysis aided the determination of the quality of the pellets in terms of strength based on the theory of functional groups and the strength of the forces acting between individual particles, for which the order of strength of the pellets was given in the last paragraph of Section 3.2. In this study, therefore, spectral deconvolution from FT-IR analysis provided significant information not just about the type of attraction forces between individual particles of the pellets, but also bond strength, which helped to establish the quality of the pellets without having to rely on special quality assessment tools.

Thermal analysis data from TGA and DSC of the pure and blended samples of NSP and PSP suggested that the behavior of the materials (Norway spruce and pea starch) under pelleting temperature was controlled by major components of the pellets, and that temperature di fferences between the pellets played a key role in determining when molecular species relevant to bonding were released. This means that how particles of the materials combined to form pellets during pelleting was determined by temperature, as noted by the degradation patterns of the resulting thermograms from TGA and DSC analyses. In relation to the quality of the pellets, which was defined in terms of burning rate, heat flow rate, and combustion e fficiency, NSP (100%) dominated because of reasons given earlier in Sections 3.3 and 3.4. Di fferences in the modification temperatures between the pure and blended pellet samples also suggested that pelleting pure pea starch would require more energy than pelleting pure Norway spruce, and a 50/50 blend of these two materials would most likely show a balance in energy consumption during pelleting. Nevertheless, a comparative energy consumption analysis of pelleting pure and blended Norway spruce and pea starch must be conducted to establish the energy needed. Judging by the strength of the forces holding its particles together, PSP (100%) was equally a good-quality pellet in terms of strength; however, its poor thermal conductivity as a result of its high oxygen to carbon (O–C) ratio with the ability to retain moisture renders the pellet unsuitable as feedstock in thermal conversion systems for energy production purposes. Biomass feedstock with a high O–C ratio reduces the heating value of the biomass and causes an increased amount of smoke with greater formation of water vapor, as well as significant energy losses when such biomass is used as feedstock in thermal energy production systems [35,52,75–78].

From SEM analysis, it was established that the pattern of surface roughness of the pellets and the mode of interconnectivity of particles, observed from the SEM images of the pellets, o ffered a strong indication of di fferences in bonding mechanisms between the pure and blended pellet samples, which basically showed that major components from two or more di fferent materials behave di fferently under certain pelleting conditions such as temperature and compression force, and that their behavior is affected by the content and concentrations of polar functional groups. According to a previous study [7], functional groups confer specific properties and act di fferently under certain pelleting conditions such as those presented in Table 1. With respect to the quality of the pellets, however, morphological features from SEM analysis could not establish any observable structural changes of importance to quality between the pellets, no matter how quality was defined.

Much still remains to be understood about di fferences in the mechanism of bonding between pure and blended pellets made from di fferent biomass materials and the chemistry of not just the strength of the pellets, but also of their combustion. Although there are many similarities in the composition of biomass, there are equally huge di fferences; hence, understanding how components of di fferent biomass combine to form good-quality pellets is made di fficult by the variability in the composition of biomass and the multiple events that occur as biomass undergoes pelleting. Given this di fficulty, however, it is vital to generate a full understanding of the structural and chemical properties of pelletized biomass relevant to good quality in all ramifications. As such, further studies are required on the bonding characteristics of pure and blended biomass before pelleting, employing contemporary analytical techniques so that a comparison can be made with this study in order to achieve better understanding of how milled biomass is transformed to pellet, as well as the source of inter-particle bonding in pure and blended biomass pellets made from di fferent biomass resources. The ratio of the blended materials should also be widely varied and pelletized in order to conclusively establish di fferences in the mechanism of bonding relevant to quality between the varied biomass pellets. To corroborate evidence of quality of the pure and blended pellets provided by the state-of-the-art analytical instruments used in this study, it is recommended that standard quality assessment tools be used.

In light of what was investigated in this study, it is fair to allude that the significance of the study is related to the information it adds to di fferences in the mechanism of bonding between pure and blended biomass pellets from two di fferent materials and how this correlates to the quality of pellets.
