Inﬂuence of Torrefaction Temperature and Climatic Chamber Operation Time on Hydrophobic Properties of Agri-Food Biomass Investigated Using the EMC Method

: Due to the tendency for excessive moisture adsorption by raw, unprocessed biomass, various methods of biomass valorization are in use, allowing for the improvement of physical– chemical biomass properties, including hydrophobicity. One of the methods is torrefaction, which changes the hydrophilic properties of the biomass to hydrophobic. Therefore, in this study, the inﬂuence of the torrefaction temperature and the exposure time to moisture adsorption conditions on the hydrophobic properties of waste biomass from the agri-food industry (lemon peel, mandarin peel, grapefruit peel, and butternut-squash peel) were analyzed. The torrefaction was carried out at the following temperatures: 200, 220, 240, 260, 280, 300, and 320 ◦ C. The hydrophobic properties were determined by using the EMC (Equilibrium Moisture Content) method, conducting an experiment in the climatic chamber at atmospheric pressure, a temperature of 25 ◦ C, and relative humidity of 80%. The total residence time of the material in the climate chamber was 24 h. It was shown that the torrefaction process signiﬁcantly improves the hydrophobic properties of waste biomass. Concerning dried raw (unprocessed) material, the EMC (24 h) coefﬁcient was 0.202 ± 0.004 for lemon peels, 0.223 ± 0.001 for grapefruit peels, 0.237 ± 0.004 for mandarin peels, and 0.232 ± 0.004 for butternut squash, respectively. After the torrefaction process, the EMC value decreased by 24.14–56.96% in relation to the dried raw material, depending on the type of organic waste. However, no correlation between the improvement of hydrophobic properties and increasing the torrefaction temperature was observed. The lowest values of the EMC coefﬁcient were determined for the temperatures of 260 ◦ C (for lemon peel, EMC = 0.108 ± 0.001; for mandarin peel, EMC = 0.102 ± 0.001), 240 ◦ C (for butternut-squash peel, EMC = 0.176 ± 0.002), and 220 ◦ C (for grapefruit peel, EMC = 0.114 ± 0.008). The experiment also showed a signiﬁcant logarithmic trend in the dependence of the EMC coefﬁcient on the operating time of the climatic chamber. It suggests that there is a limit of water adsorption by the material and that a further increase of the exposure time does not change this balance.


Introduction
The need to limit the use of fossil fuels, dictated by the prospect of their exhaustion, and the growing challenges related to the implementation of a zero-emission economy leave us looking for alternative solutions to meet environmental problems, while also ensuring the demand for electricity and heat [1]. One of the prospective sources of energy that can meet these expectations is waste biomass, which includes, among other things, agri-food residues, wood industry residues (logging and processing), residues from urban green areas, or roadside vegetation [2]. It is estimated that the annual potential of waste biomass the organic matter in biomass can be quickly decomposed, and its storage in inappropriate conditions may result in the decomposition of the material, related to biological degradation, due to the activity of microorganisms [26,27]. For this reason, methods of valorization of waste biomass are sought that will reduce the biomass ability to adsorb moisture from the environment, while improving the physical-chemical properties of the biomass.
One of the promising methods of biomass valorization is torrefaction, a thermochemical process consisting of mild thermal treatment, occurring typically between 200 and 320 • C, at atmospheric pressure, under inert gas flow for several tens of minutes [28]. Torrefaction, in addition to improving properties such as high moisture content, low energy density, and poor flowability, also makes the biomass hydrophobic [29]. So far, studies carried out by many authors have shown that, as a result of the torrefaction process, a significant improvement in the hydrophobic properties of the biomass has been achieved. These results were confirmed, inter alia, by Baronti et al. [30], using the WDPT method; Alvarez et al. [31], using the contact angle test method; or Yan et al. [32], using the EMC method.
It should be mentioned that, during the estimation of the EMC coefficient, the experiment is carried out for variable humidity, thanks to which moisture sorption isotherms are obtained that are used to describe the relationship between water content and equilibrium humidity at equilibrium state [33] by applying different equations [21]. However, little attention is paid to the determination of this parameter, analyzing the time significance when conducting the experiment under conditions of constant humidity. The exposure time of the material to adsorb moisture can be important in this subject. Moreover, there is not too much data in regard to what extent the time significance influences the hydrophobicity of biomass materials subjected to the torrefaction process. Taking these arguments into account (applying the EMC method), this work aimed (i) to evaluate the effect of the torrefaction temperature of waste biomass on the water adsorption propensities, (ii) to investigate the influence of the torrefied material exposure time under test conditions on the degree of water adsorption, and (iii) to determine the Equilibrium Moisture Content Index.

Materials
In the research, the waste biomass from the agri-food industry were used. Four types of organic residues were evaluated ( Figure 1): mandarin peel, lemon peel, grapefruit peel, and butternut-squash peel.

Samples Preparation and Torrefaction Procedure
Before the torrefaction process, all materials were dried for 24 h, at 105 • C, in a Drying Chamber KBC-65 W (WAMED, Warszawa, Poland), in order to obtain the same analytical state. After the drying process was completed, the materials were crushed in an LMN 400 (TESTCHEM, Pszów, Poland) knife mill with a sieve size of 1 mm. For the thermal conversion, the shredded organic waste in the amount of 50 g was put into the muffle furnace SNOL 8.2/1100 (SNOL, Utena, Lithuania). The mass of the samples was determined by using the RADWAG AS 220.R2 (RADWAG, Radom, Poland) scale. The torrefaction of the waste biomass was carried out in the commonly used conditions at the following temperatures: 200, 220, 240, 260, 280, 300, and 320 • C, for a period of 60 min [34,35]. Carbon dioxide from the gas cylinder was used to maintain an inert atmosphere in the reaction chamber.

Physical Properties Analysis
To characterize the basic properties of the alternative fuel from waste biomass, a proximate analysis was carried out. The analysis included the determination of moisture content (MC), ash content (AC), volatile matter content (VMC), fixed carbon content (FCC), and the higher heating value (HHV). The analysis was carried out according to the standards (Table 1). Each measurement was repeated three times, and the statistical analysis was applied.

Hydrophobic Properties Analysis
The evaluation of the hydrophobic properties was performed applying the Equilibrium Moisture Content (EMC) method [36]. In this method, the ability of the material to perform water adsorption (in the defined/controlled conditions of the environment) is measured. The less water is adsorbed by the tested material the more hydrophobic it is. Thus, in terms of hydrophobic properties, the lower value of the EMC coefficient is, the better. The Equilibrium Moisture Content (EMC) assay was calculated by using the following formula: where EMC p is the Equilibrium Moisture Content (EMC p ); m p is an increase in mass of the sample compared to the initial state (after the specified time of climatic chamber operation time) (kg), and m t is the initial mass of the sample in the dry state (kg). The determination of moisture adsorption degree of the tested materials was performed in the Climatic Chamber WK111 340 (WEISS Technik, Liedekerke, Belgium). The technical data of the climatic chamber are presented in Table 2. The test conditions were set at 25 • C, 80% relative humidity, and normal pressure. Before starting the experiment, 1 g of each investigated material was placed in ceramic crucibles and dried in a laboratory drier for 24 h, at the temperature of 105 • C. Then, the crucibles were covered with lids and placed in a desiccator to minimize environmental influences on the sample. Finally, the weighed crucibles with the material (without lids) were placed in the climatic chamber, and the mass increase was measured at appropriate time intervals. Measurements were made at the following time intervals: 4 measurements every 15 min, 4 measurements every 30 min, 5 measurements every one hour, and one measurement after 24 h from the beginning of the tests. Each measurement was repeated three times. Due to the tendency of the EMC coefficient to the equilibrium value (using the Solver add-in), the values of the coefficients of determination (R 2 ) and the coefficients a and b were determined in accordance with the logarithmic trend line expressed by the following formula: where CCOT is the climatic chamber operation time (h), and a and b are constant values of functions (-).

Statistical Analysis
The obtained research data were analyzed by descriptive statistics, taking into account the mean values and standard deviation for all tests. Moreover, a statistical analysis (at p-value < 0.05) involving a two-way analysis of variance (ANOVA) was performed. The test was focused on the elaboration of statistical significance of the influence of the torrefaction temperature and the climatic chamber operation time on the hydrophobic properties of the materials under analysis. Additionally, for the considered materials, the interaction between these parameters was investigated as well. The differences between the levels of factors were determined by a post hoc test, using a Tukey (HSD) test. The statistical analysis was developed in statistical software STATISTICA (StatSoft-DELL Software, TX, USA).

Proximate Analysis of the Torrefied Waste Biomass
The analysis of the moisture content in the tested materials did not show any dependencies between materials and temperature. The torrefied waste biomass was characterized by moisture content from 1 to 10%.
When analyzing the ash content ( Figure 2) in the torrefied fruit waste biomass, we observed the increase of AC as the torrefaction temperature increased. The average ash content in the mandarin, lemon, and grapefruit peels ranged from 4 to 12% with increasing torrefaction temperature. The highest content of the ash was observed for butternut-squash peels and ranged from 8 to 20%. Similar changes of ash content in waste biomass after the torrefaction process were observed for other organic materials [37,38]. During the torrefaction process, a significant amount of volatile matter devolatilized from the waste biomass. As the torrefaction temperature increases, the volatile matter loss is greater ( Figure 3). For the untreated waste biomass, the VMC was approximately 80%. However, for the torrefied biomass at 320 • C, the VMC was ca. 40%. No significant differences in VMC between the materials were observed. One of the basic fuel properties is the content of the fixed carbon, which indicates the carbonization rate of the material and affects the caloric value. The fixed carbon content was calculated as the difference between unity and MC, AC, and VMC. For the tested materials, it was from ca. 10% (for dried raw material) up to 50% (for torrefied biomass at 320 • C) ( Figure 4). Considering the heating value of the dried raw and torrefied waste biomass, the HHV of the untreated waste biomass was similar to the other biomasses, such as straw or hay, and it was approximately 16 MJ·kg −1 ( Figure 5). As the torrefaction temperature increased, the HHV also increased, and for the biomass torrefied at 320 • C, it was approximately 26 MJ·kg −1 , which can be comprised of the HHV of the coal [39]. The details of the proximate analysis for the investigated materials are shown in Supplementary Materials Table S1.

Impact of the Torrefaction Temperature and Climatic Chamber Operation Time on Hydrophobic Properties
Tables 3-6 show the influence of the climatic chamber operation time and the torrefaction temperature on the value of the EMC coefficient for the organic waste used in the tests. Table 3. Influence of the climatic chamber operation time (CCOT) and the torrefaction temperature on the value of the EMC coefficient for lemon peel.  Table 4. Influence of the climatic chamber operation time (CCOT) and the torrefaction temperature on the value of the EMC coefficient for grapefruit peel.  Table 5. Influence of the climatic chamber operation time and the torrefaction temperature on the value of the EMC coefficient for mandarin peel. In the case of lemon peel, after the first measurement (CCOT = 15 min), the highest moisture increase was observed for the material dried at 105 • C, non-torrefied (EMC = 0.026 ± 0.003). Slightly lower values were observed for the torrefied material in 260 • C (EMC = 0.025 ± 0.002) and 280 • C (EMC = 0.024 ± 0.006). The highest hydrophobicity was observed for materials torrefied at 320 and 240 • C (EMC = 0.02 ± 0.003). Thus, it can be seen that the EMC coefficient values after the first measurement cycle did not differ significantly from each other. However, along with the extension of the climatic chamber operation time, higher differences in the EMC coefficient were noted between the torrefied and non-torrefied material. After the last measurement cycle (CCOT = 24 h), the increase in moisture of the dried lemon peel was at the level of EMC = 0.202 ± 0.004. The material torrefied at 260 • C (EMC = 0.108 ± 0.001) was characterized by the highest hydrophobicity, achieving nearly half the value of moisture increase. Slightly higher increases were also recorded for torrefied materials at 280 • C (EMC = 0.113 ± 0.003), 300 • C (EMC = 0.117 ± 0.001), and 320 • C (EMC = 0.118 ± 0.002). Slightly greater increases were seen in torrefied lemon peel at 200 • C (EMC = 0.164 ± 0.001), 220 • C (EMC = 0.145 ± 0.006) and 240 • C (EMC = 0.128 ± 0.002).
The initial level of moisture adsorption for mandarin peels was similar to that of lemon and grapefruit peels. Only in one case-for the material torrefied at 240 • C-a much higher value of moisture adsorption (EMC = 0.054 ± 0.031) was recorded than for other temperatures. The non-torrefied material dried at 105 • C was characterized by EMC = 0.032 ± 0.002. Similar values of the index were also obtained by torrefied materials (EMC varied from 0.030 to 0.036). Extension of the CCOT resulted in higher differences in hydrophobicity between non-torrefied and torrefied material. The mandarin peel dried at 105 • C, after the last measurement cycle (CCOT = 24 h), was characterized by the EMC coefficient at the level of 0.237 ± 0.004. The highest hydrophobicity was observed for materials torrefied at 260 • C (EMC = 0.102 ± 0.001), 280 • C (EMC = 0.107 ± 0.003), and 300 • C (EMC = 0.108 ± 0.001), where the moisture adsorption decreased by more than 50%, as compared to non-torrefied material. Materials torrefied at other temperatures were also characterized by much lower moisture adsorption coefficients: for 320 • C, the EMC = 0.12 ± 0.004; for 240 • C, the EMC = 0.134 ± 0.034; for 220 • C, the EMC = 0.145 ± 0.002; and for 200 • C, the EMC = 0.156 ± 0.013.
In the case of testing the increase in moisture adsorption after the first measurement cycle (CCOT = 0.25 h) for butternut-squash peel, a slightly different situation than for other organic waste was observed. The dried material adsorbed the least amount of moisture and was characterized by the EMC coefficient at the level of 0.022 ± 0.002, while the highest moisture adsorption capacity was noted for the material torrefied at 300 • C (EMC = 0.036 ± 0.001). High increases in humidity also applied to the temperatures of 320 • C (EMC = 0.032 ± 0.005), 260 • C (EMC = 0.032 ± 0.001), 280 • C (EMC = 0.031 ± 0.002), and 240 • C (EMC = 0.030 ± 0.001). Slightly higher increases in the amount of water concerned materials torrefied at the two lowest temperatures: 200 • C (EMC = 0.023 ± 0.002) and 220 • C (EMC = 0.023 ± 0.002). As the operating time of the climatic chamber was extended, more and more moisture was detected for the non-torrefied material, while the torrefied materials were characterized by a more hydrophobic structure. After the last measurement cycle (CCOT = 24 h), the dried butternut-squash peel was characterized by the value of the coefficient EMC = 0.232 ± 0.004. For materials torrefied at temperatures of 220, 240, 260, 280, and 320 • C, a much smaller increase in moisture adsorption was noted (EMC = 0.176 − 0.178 ± 0.002). A slightly greater increase was recorded at the temperature of 300 • C (EMC = 0.184 ± 0.003) and 200 • C (EMC = 0.190 ± 0.002). Therefore, it should be noted that the butternut-squash peel was characterized by smaller differences between the hydrophobic properties in relation to the torrefied and non-torrefied material, as compared to other organic waste.
The research showed that non-torrefied materials were characterized by a much higher ability to adsorb moisture from the environment. Meanwhile, the torrefied waste biomass obtained hydrophobic properties, which significantly increased the resistance of materials to water adsorption. However, it should be marked that the most important factor is to perform the torrefaction process, as it causes significant changes in the abilities of water adsorption. The temperature of the torrefaction, in fact, is not as critical as the performance of the thermal treatment of the waste biomass (torrefaction) itself. The differences caused by the temperature (in the examined range) in the resistance to water adsorption are smaller. Moreover, the results clearly indicated that the increase of the torrefaction temperature is not always in line with the best hydrophobic propensities. It depends on the type of biomass and probably on its many other properties.
In terms of the CCOT, the initial water adsorption by all the samples results rather in surface wetting. The water, in the case of torrefied biomass, does not bond internally with a material. The results should be considered as consistent with other literature reported. In the previous work [37], by performing an alternative test (the Water Drop Penetration Time-WDPT), it was shown that the torrefaction of materials changes the hydrophilic properties of biomass to extremely hydrophobic, preventing water droplets from penetrating the surface of the material. The experiment carried out in a climatic chamber confirms these results. After the last measurement cycle, the increase in the moisture content (water adsorption) of the torrefied material, as compared to the nontorrefied material, was lower at 46.53% (for lemon peel), 48.88% (for grapefruit peel), 56.96% (for mandarin peel), and 24.14% (for butternut-squash peel). However, it was noticed that the materials torrefied at 260 • C (lemon peel, mandarin peel), 220 • C (grapefruit peel), and 240 • C (butternut-squash peel) showed the lowest increase in water adsorption for specified materials. So far, mainly the improvement of hydrophobic properties of biomass with increasing torrefaction temperature has been reported [37]. Similar conclusions were observed by Acharjee et al. [40], who obtained a decrease in the EMC coefficient with an increase in the torrefaction temperature, conducting tests at a constant temperature (30 • C) and humidity (11.3% and 83.6%). Materials subjected to the torrefaction process at higher temperatures and a longer residence time in the reaction chamber were characterized by a lower moisture ratio uptake [41].
The mechanism of improving hydrophobic properties, along with increasing the temperature of the torrefaction process, is associated with the degradation of hydroxyl groups, which is responsible for binding moisture in the material. This is because of the availability of hydroxyl groups in hemicelluloses and the amorphous reign of the cellulose chains. Their presence makes the biomass hygroscopic and makes it much more susceptible to water attraction. As a result of the thermal processing of materials (torrefaction and pyrolysis), these compounds undergo thermal decomposition. Thereby, the capture of the water molecules is impossible in order to prevent the formation of hydrogen bonds [42]. It was found that the dehydration reaction is common during the torrefaction process, and the formation of carbon dioxide is mainly due to the decarboxylation of unstable carboxyl groups in the hemicellulose structure [43]. These factors were the basis for recognizing the relationship between the presence of hydroxyl groups and the thermal treatment of the material, which was described by Chen et al. [44]. As a result of increasing the temperature of the torrefaction process, more O-H bonds are dissolved and dehydrated, thanks to which the material becomes hydrophobic [45,46]. Additionally, it is also worth noting that the lower saturated moisture content of the torrefied biomass may, to some extent, be the result of tar condensation in the pores of the thermally treated biomass. Such a phenomenon significantly hinders the transfer of moist air through a solid, which prevents water vapor condensation [42]. The presence of condensed apolar tar on a solid, such as torrefied biomass, prevents condensation of water vapor inside the pores [47].
The obtained results are therefore the basis for the search for further correlations between the chemical and physical properties of the material and its hydrophobicity. Recently, Korshunov et al. [48] noticed that the hydrophobic properties of biomass depend on its porosity. They found that the contact angle of the droplet with a material's surface decreases with increasing measurement time, due to the penetration of water droplets inside the dried material. This speed decreases if torrefaction was applied.
From an economic point of view, obtaining a higher degree of hydrophobicity at a lower torrefaction temperature is highly desirable in order to minimize the production costs resulting from the higher electricity/energy consumption. Hence, further elucidation of the determinants influencing the hydrophobic behavior of materials seems to be crucial. Figure 6 shows the kinetics of EMC growth as a function of the climatic chamber operation time (CCOT), expressed as logarithmic trend lines. It can be seen that the value of the EMC coefficient tends to stabilize (the curve flattens). For all organic waste, in the initial period, the increase in moisture adsorption is dynamic, and then it decreases, reaching the limit, meaning that it reaches the thermodynamic equilibrium. This phe-nomenon is quite common in the literature that studies water adsorption by materials of various types. Chen et al. [49], examining the water adsorption capacity of bamboo and low-density polyethylene composites with bamboo charcoal addition in the immersion test, observed that the samples rapidly adsorbed water in the early stages of immersion. This process slowed down until the equilibrium conditions were reached. Similar conclusions were defined by others: Tiebie et al. [50], examining the kinetics of water adsorption by coffee powder (test duration time was 90 min); Mu et al. [51], investigating the characteristics of biocomposites (water adsorption study exceeding 50 days); and Tamrakar and Lopez-Anido [52], examining water-adsorption behavior and durability of extruded wood polypropylene composite (WPC) material used in Z-section sheet piles. The shape of the water-adsorption curves is usually constant, regardless of the test time (it is also dependent on some other factors, such as material shape, structure, type, and sample size).  Table 7 shows the evaluated parameters of the a and b coefficients for the modeled logarithmic trend lines. The highest matches with respect to the experimental data concerned butternut-squash peel (average R 2 = 0.9907), and the lowest concerned mandarin peels (average R 2 = 0.9456). In the other cases, the coefficient of determination (R 2 ) was 0.9594 (average) for grapefruit peel and 0.9834 (average) lemon peel. Therefore, it can be concluded that the logarithmic model used indicates a high degree of matching the estimated parameters to the experimental data.

Main Results of Statistical Analysis
The conducted statistical analysis (applying two-way analysis of variance ANOVA) confirmed that there is a strong dependence (p-value < 0.05 for all organic waste) between the climatic chamber operation time (CCOT) and the Equilibrium Moisture Content assay (EMC). Likewise, changing the torrefaction temperature (TT) of torrefied materials significantly affects the EMC as well.
Appendix A shows significant differences for the EMC variable for the effect of torrefaction temperature (TT), according to Tukey's test (HSD), for biomass waste. The conducted analysis showed a statistically strong relationship between dried raw and torrefied waste for all groups in terms of moisture adsorption. It was noted that the impact of the torrefaction temperature itself is smaller, as evidenced by statistically insignificant differences in a large proportion of these cases.
Appendix B shows significant differences for the EMC variable for the effect of climatic chamber operation time (CCOT), according to Tukey's test (HSD), for biomass waste. In the vast majority of cases, a significant time relationship was demonstrated, with the exception of a few statistically insignificant cases.
With the combination of the torrefaction temperature parameters and the operating time of the climatic chamber (TT·CCOT), a significant influence (p-value < 0.05 for all organic waste) on the EMC index increase was observed. Detailed data for the analysis of variance are presented in Table 8.

Conclusions
Due to the unfavorable physical, chemical, and hydrophilic properties of raw biomass, different low-energy-consuming solutions are offered to obtain a valuable stable and highquality fuel. One of the proposed alternatives is biomass torrefaction, which improves the energy density, as well as the hydrophobicity, of the material. The hydrophobic properties are of special importance, as they enable the safe and long-term storage of torrefied biomass, thus preventing its degradation.
The article analyzed the impact of the torrefaction temperature and the operation time of the climatic chamber on the hydrophobic properties of selected agri-food waste (lemon peel, mandarin peel, grapefruit peel, and butternut-squash peel). It was shown that torrefaction significantly improves the hydrophobic properties of biomass. With regard to dried non-torrefied material, the EMC coefficient decreased by a maximum of 56.96-24.14%, depending on the type of organic waste.
There was also no evidence of an increase in hydrophobic properties (lowering of the EMC coefficient) with increasing torrefaction temperature. The lowest EMC index was obtained at different temperatures for different waste biomass. It is an important observation influencing the economics of torrefaction and optimization of production processes, guaranteeing a reduction in energy expenditures, and thus lowering the costs of torreficates production. Significant differences in the kinetics of adsorbing moisture were also observed, along with increasing the operation time of the climate chamber.
The results also create a space for further research on the determinants of improving the hydrophobic properties of biomass during torrefaction, especially in terms of the physical and chemical properties of materials.

Conflicts of Interest:
The authors declare no conflict of interest.

Appendix A
Statistically significant differences are marked in bold.