Relationship between Odor Adsorption Ability and Physical–Hydraulic Properties of Torrefied Biomass: Initial Study

Abstract

Various techniques are implemented to reduce odor emission due to their potential multi-source nature. One modern approach is the use of thermochemically processed biomass to eliminate odors. Compared with raw biomass, processed biomass is characterized by greater porosity and an expanded specific surface. In these laboratory experiments, adsorption tests for a mixture of indole, 2,3-dimethylpyrazine, and 2,3,5-trimethylpyrazine are carried out using torreficates produced from biomass from the agri-food industry (walnut shells, orange peels, peach stones, and apple wood chips). This research is focused on the determination of the correlation between the physical-hydraulic properties of the torreficates and their ability to reduce the odors simulated by the selected compounds. The results indicate that 2,3-dimethylpyrazine and 2,3,5-trimethylpyrazine are not detected in any of the investigated low-temperature biochars. However, indole is detected in most materials, and its most significant quantities are adsorbed on torreficates made of orange peels (45.64 µg·mL⁻¹ ± 40.02 µg·mL⁻¹) and peach stones (61.26 µg·mL⁻¹ ± 49.55 µg·mL⁻¹). The performed analysis reveals that the highest correlation with the ability to adsorb indole is found for the average pore size (r = 0.66) and specific density (r = −0.63) as well as the content of fixed carbon (r = 0.66), which may prove the importance of physical-hydraulic properties in odor sorption by low-temperature torreficates.

1. Introduction

The human sense of smell is essential to life functions such as warning and protection against environmental hazards, eating behavior and nutrition, and social communication [1]. The sensation of smell can evoke pleasant or unpleasant emotions (depending on subjective preferences) and even lead to physical responses involving the trigeminal nerve and other levels in the brain [2]. Unpleasant odors that put a person in a state of subjective discomfort in the physical and mental sphere are called odor nuisances (odors) [3]. The degree of irritation caused by odorous air depends on five main parameters (FIDOL) [4], which are [5,6,7]: odor frequency (F), odor intensity (I), the duration of exposure to odor (D), odor offensiveness (O), and the location of the odor (L). It is assumed that the impact of odors is associated with a negative impact on health and quality of life, mainly due to their harmful effects on mental health [3]. The conducted research showed that prolonged exposure to odors can cause depression, mood changes, fatigue, respiratory problems, headaches, nausea, chest tightness, nasal congestion and eye and throat irritation [3,8,9]. The classification of fragrances, aromas, and odorants is highly varied and is the source of much scientific debate. As commented by [10], the fundamentals of classification schemes vary widely, as each is established with specific goals and objectives in mind. Classification schemes may differ in terms of theoretical basis; thus, special care should be taken when comparing them [10].

Regardless of the sources of odor classification, their destructive effects, and the multi-source nature of emissions (they may cover most scopes of economic and industrial activities but also occur as a result of the common use of the environment), different elimination techniques are being explored, allowing them to be cost-effective and practical and thus reducing emissions of these compounds [3,11]. According to [12], the methods of eliminating odorous substances can be divided into primary and secondary. Primary methods primarily comprise appropriate spatial planning [13], while secondary methods include physicochemical technologies (mainly filtration, adsorption, ionization, ultraviolet disinfection, and UV-photocatalytic oxidation) and biological technologies [14]. Due to the multidirectional and diverse emissions of odorous pollutants, it is not easy to define universal methods for preventing and treating odorous gases [12]. In the case of primary methods, one of the main factor determining an appropriate preventive measure selection is the economic aspect [15]. The situation is different concerning secondary methods, where the choice of odor cleaning technique is determined by the properties of the treated gases, investment costs, operating costs, and expected effects [12].

A novel approach to eliminating odor emissions is using thermochemically processed biomass (biochar/torreficates)—substances with physical–chemical properties similar to charcoal and for the production of which organic waste of agricultural and food origins is used, among other substances. According to an FAO report [16], it has been estimated that 1.6 Gtons of “primary product equivalents” are wasted globally in terms of food waste. The advantage of such an approach is therefore the management of bio-waste, which is also essential from the point of view of circular economy implementation. Preliminary studies have shown that biochar may be an effective substrate in absorbing odors from various industries [17]. It can be dosed directly onto the surface layer of the odor-emitting material and constitute a kind of filter (or a filter bed, if there is the additional application of microorganisms), which becomes a barrier to the stream of volatile aromatic compounds flowing through a given installation [18,19,20]. Although it is known that thermochemically processed biomass is a good source of odor reduction, the mechanisms and quantitative descriptions of the effects of biochar on some odor-causing compounds are still not fully understood.

Furthermore, the correlations between the specific properties of biochar and the effectiveness of the odor-reduction process are still unknown in many cases. It should be noted that most experiments in the study of processed biomass for odor reduction are focused on the evaluation of biochar, formed mainly at temperatures of 400–700 °C. The study of torreficates, formed at temperatures of 200–320 °C, is often overlooked. The choice of biochar to counteract odor phenomena over torrefacted products is highly rational since higher processing temperatures increase the microporous structure and expand the specific surface area to a greater extent [21], which may influence sorption values. Additionally, Deng et al. [22] explain that the high adsorption potential of biochar is due to its complex pore structure and numerous surface-active functional groups. However, on the other hand, the requirement of higher temperatures causes biochar production costs to be significantly higher than those of torreficates. Thus, supplementing the current literature with issues related to the use of torreficates seems to represent a valuable research niche and may shed new light on the issues of using organic materials to eliminate odor nuisance. This article can also help determine the key properties that may affect the adsorption characteristics of materials produced at low temperatures.

As mentioned earlier, some of the mechanisms responsible for the effectiveness of odor sorption are not fully understood. According to Moradi-Choghamarani et al. [23], the thorough knowledge of the physical–chemical properties of valorized biomass and its molecular composition is of crucial importance and conditions its application in environmental management systems. These aspects are mostly well known; however, not enough attention is paid to the physical–hydraulic properties (such as wettability, real density, porosity, etc.) before its practical application [23]. Thus, this work aims to connect two research gaps to accomplish the following goals: (i) determine whether torreficates produced from agri-food industry biomass are able to adsorb odors; (ii) determine whether physical-hydraulic properties may play a role in the effectiveness of odor sorption by torreficates; (iii) provide the literature with new information about the physical-hydraulic characteristics of raw and torrefied biomass.

This article is structured as follows: Section 2 details the materials and methodology related to the torrefaction process, odor adsorption experiment, and physical–hydraulic property determination. Section 3 presents the results analysis, discussion, and future recommendations. Finally, Section 4 includes conclusions based on the conducted experiments.

2. Materials and Methods

2.1. Feedstock

Biomass residues from agri-food processing were used for the production of torreficates. Four types of organic waste were selected—walnut shells (WS), orange peels (OP), peach stones (PS), and apple wood chips (AWC). These are typical types of waste biomass from the Polish agri-food sector.

2.2. Initial Preparation and Torrefaction Process

The collected waste biomass was dried in a KBC-65 W drying chamber (WAMED, Warszawa, Poland) according to PN-EN ISO 18134-2:2017-03 standards. The material was dried for 24 h at a temperature of 105 °C. Then, the substrates were comminuted in an LMN 400 knife mill (TESTCHEM, Pszów, Poland) using a sieve diameter of 1 mm. Then, a 60 g sample of the material was prepared and measured with an AS 220.R2 (RADWAG, Radom, Poland) balance.

The torrefaction process was carried out in the SNOL 8.2/1100 muffle furnace (SNOL, Utena, Lithuania) at a temperature of 200 °C. The residence time was 60 min. In order to ensure the non-oxidation character of the process, 90 mL∙min⁻¹ inert gas was fed into the muffle furnace from a cylinder (CO₂). The heating rate was set to 10 °C∙min⁻¹.

2.3. Proximate Analysis and Torrefaction Process Performance

In order to determine the properties of waste biomass and produced torreficates, proximate analysis (evaluating ash content, volatile matter content, fixed carbon content, and higher heating value) was carried out (each parameter was checked with 3 repetitions). These experiments were performed according to ISO standards. Details on the torrefaction process and determination of the parameters included in the proximate analysis were described in the previous paper [24]. Additionally, parameters of the torrefaction process were determined, such as degree of torrefaction (df), energy density (ED), and energy yield (EY).

2.4. Physical–Hydraulic Properties

The analysis of the physical-hydraulic properties of the torreficates consisted of determining the mass loss (M_L), bulk density (ρ_D), specific density (ρ_S), porosity (ε), pH, electroconductivity (EC), specific surface area (S_BET), total pore volume (V_T), average pore size (L_S), water-holding capacity (WHC), and hydrophobicity (WDPT).

2.4.1. Mass Loss and Density Determination

The mass loss (M_L) was determined on the basis of the difference in mass of the sample before and after thermochemical processing of biomass according to the formula:

$${\mathrm{M}}_{\mathrm{L}}=\frac{{\mathrm{m}}_1-{\mathrm{m}}_2}{{\mathrm{m}}_1}$$

(1)

where M_L—mass loss (%); m₁—sample mass before torrefaction (g); and m₂—sample mass after torrefaction (g).

The bulk density (ρ_D) was determined using a 100 cm³ vessel. After filling the vessel with a specified type of biomass and biochar, the bulk density was determined according to the equation:

$${\mathsf{\rho }}_{\mathrm{D}}=\frac{{\mathrm{m}}_{\mathrm{T}}}{{\mathrm{V}}_{\mathrm{N}}}$$

(2)

where ρ_D—bulk density (kg∙m⁻³); m_T—sample mass after filling vessel (kg); and V_N—volume of the used vessel (m³).

The specific density (ρ_S) was determined according to the method explained and described in [25] using pycnometer and a vacuum pump (liquid immersion method). The pycnometer was filled with ethanol at 20 °C at the end of the capillary and weighed on an analytical balance. Then, 10 g of torreficate/dried material was weighed on an analytical balance. About 2/3 of the ethanol was poured from the previously filled pycnometer, and a portion of the material was poured into it so that it would not remain on the internal walls. The sample was vented by connecting the pycnometer to a vacuum pump and pumping out air for 20 min. After this process, the pycnometer was filled up with ethanol and weighed. Specific density was determined according to the formula:

$${\mathsf{\rho }}_{\mathrm{S}}=\frac{{\mathrm{m}}_{\mathrm{B}}}{\mathrm{Q}+{\mathrm{m}}_{\mathrm{B}}+\mathrm{P}}·{\mathsf{\rho }}_{\mathrm{E}}$$

(3)

where ρ_S—specific density (kg∙m⁻³); m_B—sample mass in pycnometer (kg); Q—pycnometer mass after being filled up with ethanol (kg); P—pycnometer mass with a weight of material after deaeration and pouring ethanol to the end of the capillary (kg); and ρ_E—density of ethanol (adopted ρ_E = 789 kg∙m⁻³).

The porosity of torreficate and the dried product was determined using the specific density and bulk density of materials according to the equation [26]:

$$\mathsf{\epsilon }=1-\frac{{\mathsf{\rho }}_{\mathrm{D}}}{{\mathsf{\rho }}_{\mathrm{S}}}$$

(4)

where ε—porosity (-).

2.4.2. pH and Electroconductivity

The dried materials and the produced torreficates were characterized in terms of pH and electroconductivity (EC). Measurements were made in a measured solution: 1 g of dry weight per 10 mL of distilled water. The measurement was performed 30 min after mixing the solution [27] using the Elmetron CPC-411 (CPC-411, Elmetron, Zabrze, Poland).

2.4.3. Sorption Parameters

Analysis of the porosity of solids by the gas sorption method was performed using a volumetric method on the precise sorption meter ASAP 2020 (Micromeritics Inc., Norcross, GA). N₂ sorption was performed in 77K. The p/p₀ range was 0–0.96. In this study, the specific surface area was determined by the BET method (S_BET), the total pore volume (V_T) was less than 50 nm, and the mean pore size (L_S) was smaller than 50 nm, assuming the slit shape of the pores. The analysis was performed by an external laboratory.

2.4.4. Water-Holding Capacity and Hydrophobicity

The hydrophobic properties of torreficates and dried materials were determined by measuring the water droplet penetration time (WDPT). The test materials were evenly distributed over the Petri dish. Then, using a micropipette, a few drops of demineralized water were placed on the layer and the time after which the water soaked into the material was measured (with a stopwatch). The hydrophobic properties were determined on the basis of the water droplet penetration time according to the classification presented in

Table 1. Classification criterion/range of hydrophobic properties [28,29].

Time of Drop of Water Penetration	Hydrophobic Properties
below 5 s	Hydrophilic
from 5 s to 60 s	Slightly hydrophobic
from 60 s to 600 s	Strongly hydrophobic
from 600 s to 3600 s	Severely hydrophobic
above 3600 s	Extremely hydrophobic

.

The water-holding capacity (WHC) was determined by saturating the biochar and water-dried materials. A total of 5 g of material was poured into a glass vessel; then, water was slowly added until excess was observed. The mixture prepared in this way was left for 24 h to ensure homogeneity of the water content in the sample. After this time, the material was placed in the filter for 2 h to remove excess water. Then, the saturated material was weighed and dried for 24 h at 105 °C (used method slightly modified than described in [30,31]). The water-holding capacity was calculated according to the equation:

$$\mathrm{WHC}=\frac{\left({\mathrm{M}}_{\mathrm{W}}-{\mathrm{M}}_{\mathrm{D}}\right)}{{\mathrm{M}}_{\mathrm{D}}}·100\%$$

(5)

where WHC—water-holding capacity (%); M_W—mass of the sample after filtration (g); and M_D—mass of the sample after drying (g).

2.5. Odor Adsorption Experiment

2.5.1. Odor Mixture

The odor used in this study was a prepared liquid mixture of a 10% solution of 2.5-dimethylpyrazine, a 10% solution of 2,3,5-trimethylpyrazine, and a 7.5% solution of indole (volume percent). The characteristics of these chemical compounds are presented in

Table 2. Odors used in the experiment [32,33].

Compound	Structural Formula *	Odor Description
Indole		Pungent, floral, slightly naphtha- and mothball-like with a fecal and animalic musty character
2,3-dimethyl pyrazine		Nutty, nut skin, cocoa, peanut butter, coffee, walnut, caramelly, roasted
2,3,5-trimethyl pyrazine		Nutty, nut skin, earthy, powdery, cocoa, baked potato, roasted peanut/hazelnut, musty

* Structural formulas were prepared using Chemispider.com software.

.

2.5.2. Laboratory Stand

Figure 1 shows the laboratory stand used to determine the odor absorption capacity of torreficates. The stand consists of a gas cylinder with carbon dioxide, a pressure regulator, throttle-check valves, flow meters, paper fragrance carriers, torreficates, and cotton wool.

2.5.3. Experimental Procedure

Paper aroma carrier (aroma test paper) was sprinkled with 100 µL of the odor mixture using a micropipette, and the filter column was connected to a silicone tube through which an inert gas (CO₂) carrying the evaporating aromatics was passed. Carbon dioxide, along with aromatic compounds, flowed through a layer of 65 mg of biochar, which was an anti-odor filter, and then went to the outlet of the installation. The duration of the process was set at 60 min and the flow of inert gas was set at 4 dm³∙min⁻¹ (1 dm³ ∙ min⁻¹ for each filter column) in accordance with the recommendations of Hwang et al. [34]. After 60 min, the filter columns were removed and the paper aroma carrier and torreficate were transferred to two separate Eppendorf tubes. Then, 1950 µL of dichloromethane and 50 µL of β-caryophyllene at a concentration of 1 mg·ml⁻¹ were added to the test tubes as an internal standard. After 15 min of washing, the paper aroma carrier/torreficate was withdrawn from the tube. The tubes prepared in this way were centrifuged for 12 min using a Scilogex D1008 laboratory centrifuge with a speed of 7000 rpm∙min⁻¹. After this time, the contents of the tubes were filtered through a cotton–celite filter into 912P vials, which were protected with parafilm. The test was performed three times.

The substances prepared in this way were analyzed by gas chromatography combined with mass spectrometry. A capillary column ZB 5plus (Phenomenex, Torrance, CA, USA) with a length of 30 m was used for the analysis: inside diameter 0.25 mm and layer thickness 0.25 µm. The samples were injected at 250 °C with volume of 1 µL and split 10. The initial column temperature was started at 50 °C for 2 min, then was increased to 180 °C at a rate of 10 °C·min⁻¹, then was increased to 270 °C at a rate of 20 °C·min⁻¹ and held there for 5 min. The dispenser was set to 250 °C. Helium with a linear flow rate of 35 cm·s⁻¹ was used as the carrier gas. Mass spectrometer parameters were as follows: ion source temperature 220 °C and scan mode in the range of 35–300 m/z.

2.6. Statistical Analysis

To check the statistically significant differences between the materials, one-way analysis of variance (ANOVA) was performed with Tukey’s post-hoc test (HSD). Additionally, in order to check the values of the correlation between the parameters, the linear correlation coefficient r was determined using the mean results of the parameters. Statistical analysis was performed in Statistica (StatSoft—DELL Software, Tulsa, OK, USA) software at the significance level of p = 0.05.

3. Results and Discussion

3.1. Proximate Analysis of Materials

Table 3. Proximate analysis of dried materials and produced torreficates.

Biomass Residue	Temperature	AC	VMC	FCC	HHV
	°C	%	%	%	kJ∙kg−1
Walnut shells	105	0.99 a ± 0.03	82.11 ab ± 0.76	16.90 abc ± 0.73	19,327 ab ± 269
	200	1.18 a ± 0.07	80.27 a ± 0.10	18.56 b ± 0.17	19,794 a ± 159
Orange peels	105	3.11 d ± 0.05	80.91 ab ± 0.26	15.98 ac ± 0.29	16,210 ± 127
	200	4.55 e ± 0.14	68.35 c ± 0.86	27.11 e ± 0.92	20,692 ± 93
Peach stones	105	0.68 b ± 0.06	80.89 ab ± 1.48	18.43 ab ± 1.44	19,794 ± 159
	200	0.62 b ± 0.07	77.32 d ± 1.24	22.06 d ± 1.31	21,640 ± 333
Apple wood chips	105	1.62 c ± 0.06	83.06 b ± 0.46	15.32 c ± 0.44	19,113 b ± 198
	200	1.52 c ± 0.10	80.41 a ± 1.14	18.07 ab ± 1.04	19,321 ab ± 98

The same signs (a–e) in a column mean no statistically significant changes.

shows the results of the proximate analysis of the dried materials and torreficates. In the case of the ash content (AC) for the raw biomass from the agri-food industry, it was characterized by the range of 0.68–3.11%, with the lowest results for peach stones and the highest for orange peels. It should be considered that these results are relatively low compared with other types of biomass, with the exception of wood and woody biomass, according to results obtained by Zając et al. [35]. In examining the ash content for waste from the agri-food industry, Hills et al. [36], obtained 0.31–1.38% ash for shells (cobnut, coconut, walnut, almond, peanut), 0.80–5.08% ash for fibers (jute, flax, straw, hay, rice husk, coconut husk, sugarcane husk), and 1.45–6.36% ash for soft peel (sweet lime, banana, yam, cassava, potato, pomegranate, orange). For walnut shells and orange shells, the torrefaction process increased the ash content, while for peach stone and apple wood chips, it decreased slightly. However, in general, the process of conventional torrefaction increases the ash content [37], but the decrease in this parameter due to torrefaction may result from the variability of samples as evidenced by the value of the standard deviation. Nevertheless, in these two cases, there were no statistically significant changes in ash content due to thermal processing.

As expected, due to torrefaction in all the types of biomass, the volatile substances (VMC) content decreased. The low degradation levels of these compounds result from the fact that the process of the degradation of volatile materials begins at about 200 °C [38]. A relatively high reduction of VMC was obtained only for orange peels, for which torrefaction caused a decrease in VMC from 80.91% ± 0.26% to 68.35% ± 0.86%. This may be due to the fact that this substrate contains a high amount of hemicellulose [39], which degrades around the temperatures for which torrefaction was carried out [40]. The remaining substrates were characterized by VMC in the range of 80.41–82.11%, which is the typical range of this parameter for biomass from the agri-food industry [41]. The decrease in volatile matter content in the remaining substrates was small (approx. 2–3%), which is often the case for torrefaction processes carried out at low temperatures [42]. The decrease in this parameter resulted in an increase in fixed carbon because the substrates are characterized by solid carbon in the biomass that remains after devolatilization [43]. For raw biomass, the value of this parameter ranged from 15.98% to 18.43%, with the lowest FCC values obtained for orange peels and the highest for peach stones. The high reduction of VMC caused an increase in FCC for orange peels to 27.11% ± 0.92% and for other substrates to approx. 2–3%.

The evaluated biomass differed in terms of the higher heating value. The lowest HHV was noted for orange peels (16,209.7 kJ∙kg⁻¹ ± 127.27 kJ∙kg⁻¹). The remaining substrates were characterized by statistically significantly higher HHV (19,112.7–19,793.7 kJ∙kg⁻¹). These results are confirmed by the research presented in [44], which showed that biomass of similar origin may have different HHV values and energy parameters due to its diversified chemical composition [45]. As a result of the torrefaction process, HHV increased in all materials, which mainly resulted from the decrease in hydrogen and oxygen content and the increase in carbon content during the decomposition of biomass in the torrefaction process [46,47]. The highest increases were noted for orange peels and peach stones, in which the value of the parameter increased to 20,692.0 kJ∙kg⁻¹ ± 93.18 kJ∙kg⁻¹ and 21,640.3 kJ∙kg⁻¹ ± 332.68 kJ∙kg⁻¹, respectively.

3.2. Torrefaction Performance Determination

Table 4. Torrefaction performance coefficients.

Torreficate	ML	dF	ED	EY
	%	-	-	-
Walnut shells	11.70 a ± 0.51	0.98 a ± 0.01	1.02 a ± 0.02	0.90 a ± 0.01
Orange peels	29.17 c ± 3.71	0.84 b ± 0.01	1.28 c ± 0.01	0.90 a ± 0.04
Peach stones	9.66 a ± 2.23	0.96 a ± 0.01	1.09 b ± 0.03	0.99 b ± 0.03
Apple wood chips	1.26 b ± 0.12	0.97 a ± 0.02	1.01 a ± 0.01	1.00 b ± 0.01

The same signs (a–c) in a column mean no statistically significant changes.

presents the coefficients (mass loss, degree of torrefaction, energy densification ratio, energy yield) proving the performance torrefaction process. The mass loss was characterized by high differentiation—the highest values of this parameter were obtained for orange peels. The torrefaction produced from this material lost almost 30% of its weight in relation to the raw biomass. The high weight loss mainly resulted from the removal of volatile yields [48] due to thermochemical processing, which was noted in the subsection above. This result should be considered high because the optimal torrefaction conditions with a 30–60 min holding time responsible for 30% weight loss are temperatures of 250–350 °C [49,50]. In the remaining cases, the weight loss was much lower and amounted to 9.66% ± 2.23% for peach stones and 11.70% ± 0.51% for walnut shells—these differences were not statistically significant. In turn, for apple wood chips, the weight loss was very small and amounted to only 1.26% ± 0.12%. A very low weight loss coefficient for a temperature of 200 °C was also noted in the experiment conducted by Ramos-Carmona et al. [51] where woody biomass was also evaluated.

With regard to the degree of torrefaction, the results showed that the range of this parameter was 0.84–0.98, which indicates that the torrefaction process performed at a relatively low temperature of 200 °C did not significantly improve the stability of the sample. However, this parameter is usually used to determine the combustion characteristics [52], but it can also be an important supplement to the physical characteristics of the torreficates. In the case of ED, the highest energy concentration was observed for orange peels (ED = 1.28 ± 0.01) and peach stone (ED = 1.09 ± 0.03), which may have been due to the largest increase in fixed carbon occurring in these substrates [53]. The energy density in the remaining cases was statistically significantly lower (ED = 1.01–1.02) due to the small amount of released volatiles [24]. In turn, the EY coefficient for walnut shells and orange peels was identical and amounted to 0.90, while for peach stones and apple wood chips it was also similar and amounted to 0.99–1.00. These results should be considered similar to those for lignocellulosic biomass [54].

3.3. Physical Parameter Determination

Table 5. Bulk densities, specific densities, and porosities of torreficates and dried materials.

Biomass Residue	Temp.	ρD	ρS	ε
	°C	kg∙m−3	kg∙m−3	-
Walnut shells	105	653 b ± 24	1340 ab ± 35	0.512 b ± 0.013
	200	650 b ± 12	1337 ab ± 12	0.513 b ± 0.004
Orange peels	105	286 a ± 5	1427 ac ± 51	0.799 de ± 0.007
	200	200 d ± 6	1102 d ± 25	0.819 e ± 0.004
Peach stones	105	720 c ± 32	1367 ac ± 15	0.458 a ± 0.025
	200	703 c ± 15	1252 b ± 12	0.438 a ± 0.005
Apple wood chips	105	325 a ± 8	1449 c ± 57	0.775 c ± 0.009
	200	322 a ± 6	1352 a ± 34	0.762 c ± 0.006

The same signs (a–e) in a column mean no statistically significant changes.

presents the bulk densities, specific densities, and porosities of the dried materials and the produced torreficates. The bulk densities of all the raw samples were highly variable and depended on the consistency of the material. The highest bulk density results were obtained for the peach stone (720 kg∙m⁻³ ± 32 kg∙m⁻³) and for walnut shells (653 kg∙m⁻³ ± 24 kg∙m⁻³). The remaining types of biomass had a significantly lower bulk density—in the case of orange peels, this parameter was 286 kg∙m⁻³ ± 5 kg∙m⁻³, and in the case of apple wood chips, it was 325 kg∙m⁻³ ± 8 kg∙m⁻³. Depending on the type and origin, biomass is characterized by very different values of bulk density, which can range from 50 to 670 kg∙m⁻³ [55]; this is consistent with the results of this experiment. As a result of the torrefaction process in all the materials, the bulk density was reduced. This change was not statistically significant in all cases except for the orange peel (decrease of 86 kg∙m⁻³). Such an action results from the fact that the reduction of bulk density is directly related to the decrease in mass and CO₂ release [56], hence the highest reduction concerned the material in which the most mass was lost.

In the case of specific density, the variation of the parameter was much lower than in the case of bulk density. The raw biomass was characterized by a specific density in the range of 1340–1449 kg∙m⁻³, with the highest specified density for apple wood chips and the lowest for walnut shells. Similar results for ground, dried biomass from forest waste were obtained by Dyjakon and Noszczyk [24], wherein the value of this parameter was in the range of 1442–1604 kg∙m⁻³. The torrefaction process caused a reduction in the specific density in all cases. In 3 out of 4 cases, this change was statistically significant except for walnut shells, where the change was minor and resulted in a 3 kg∙m⁻³ change. The decrease in specific density due to thermal processing in the inert atmosphere in 200 °C was also noted in the case of exotic fruit seed torrefaction [57]. It is likely that this phenomenon results from the fact that the torrefaction of the biomass caused a loss in the mass of the materials while slightly affecting their particle size and pore characteristics. The material fraction has a significant influence on its specific density [58].

The porosity of the materials was varied and was greater for materials characterized by low bulk density—ε = 0.799 ± 0.007 and ε = 0.775 ± 0.009 for orange peels and apple wood chips, respectively. In turn, with respect to materials with a higher bulk density, ε = 0.799 ± 0.007 and ε = 0.775 ± 0.009 were obtained for peach stones and walnut shells, respectively. Torrefaction slightly affected the porosity of the materials—no statistically significant differences were observed in any of the cases. As a rule, the porosity of biochar is at the level of 0.55–0.86 depending on the pyrolysis temperature [26]. Similar results were obtained in the conducted experiment.

Table 6. Results of the analysis of the N₂ isotherm in 77 K for torreficates and dried materials.

Biomass Residue	Temp.	SBET	VT	LS
	°C	m2∙g−1	cm3∙g−1	nm
Walnut shells	105	0.23	0.0003	2.91
	200	0.41	0.0008	3.92
Orange peels	105	0.27	0.0003	2.43
	200	0.85	0.0012	2.84
Peach stones	105	0.08	0.0001	2.71
	200	0.23	0.0003	2.58
Apple wood chips	105	0.57	0.0009	3.23
	200	0.73	0.0012	3.34

presents the results of the N₂ sorption analysis in 77K for torreficates and dried materials. The unprocessed biomass was characterized by a specific surface area (S_BET) in the range of 0.08–0.57 m²∙g⁻¹. Peach stones had the lowest S_BET while apple wood chips had the highest. These results are in line with the literature data, according to which thermally or chemically unprocessed biomass is characterized by very low specific surface area values. When examining S_BET of virgin wood (spruce, birch), Bergna et al. [59] obtained a value of 0.4–0.6 m²∙g⁻¹ depending on the species of wood. Depending on the method of mechanical processing in studies by Victorin et al. [60], wheat straw was characterized by a specific surface area in the range of 0.072–0.114 m²∙g⁻¹. Torrefaction, as expected, did not significantly improve the value of the specific surface area. All the torreficates were characterized by the S_BET parameter below 1 m²∙g⁻¹. The highest increase was observed for orange peels, which, after torrefaction, increased from 0.27 m²∙g⁻¹ to 0.85 m²∙g⁻¹. This is due to the fact that only temperatures in the range of 400–700 °C are suitable for producing a material with high porosity and a high specific surface area from biomass [61]. In general, the temperatures used in the torrefaction process (200–300 °C) are too low to significantly improve the S_BET of raw, unprocessed biomass; however, there are reports in the literature that it is possible to achieve a S_BET level close to 200 m²∙g⁻¹ at torrefaction limits [61,62].

Table 6 also shows the V_T and L_S of dried materials and torrefacted materials. In the case of V_T, it can be observed that torrefaction increased the parameter for all the materials. The total pore volume increased the most in the case of orange peels, for which a four-fold higher pore volume was observed compared with that of the dried material. As a result, the apple wood chips and orange peels had the same pore volume (<50 nm). A slightly different situation was noted in the case of L_S. As a result of the torrefaction process, in most (3 out of 4) materials, the average pore size <50 nm increased. Walnut shells showed the greatest pore expansion.

3.4. pH and Electroconductivity

Table 7. pH and electroconductivity of torreficates and dried materials.

Biomass Residue	Temp.	pH	EC
	°C	-	mS∙cm−1
Walnut shells	105	4.82 b ± 0.04	0.558 ab ± 0.006
	200	5.16 a ± 0.05	0.665 b ± 0.008
Orange peels	105	4.61 d ± 0.02	1.291 e ± 0.130
	200	5.24 a ± 0.02	1.491 f ± 0.051
Peach stones	105	4.92 b ± 0.09	0.434 ac ± 0.056
	200	5.81 c ± 0.08	0.514 ab ± 0.017
Apple wood chips	105	5.27 a ± 0.02	0.848 d ± 0.026
	200	5.67 c ± 0.14	0.356 c ± 0.011

The same signs (a–f) in a column mean no statistically significant changes.

shows the pH and EC of the evaluated materials. Raw biomass differed significantly with the pH value, which was in the range 4.61–5.27. Conducting the torrefaction process increased the pH in each of the materials, as a result of which the parameter range changed to 5.16–5.81; the lowest value was obtained for walnut shells and the highest for peach stones. This situation resulted from the fact that, due to the torrefaction and further increase in process temperature, the pH of the resulting char increased [63] because of the higher ash content in the torreficates after thermochemical processing [64,65]. Admittedly, likely due to sample heterogeneity (despite homogenization), for peach stones and apple wood chips, the ash level decreased and the pH level increased. There is also a chance that the cellulose, hemicellulose, and lignin fractions may suggest correlations between the pH of the raw material and the product thermochemically processed at low temperatures; however, due to the lack of research in this area, this is still unverified [66].

A similar situation was noted for EC, which appears to be an important, albeit underrated, parameter in the context of odorant adsorption, as adsorption mechanisms can be regulated by chemical changes, particularly by the alteration of electrical conductivity or reduction reactions [67]. In general, increasing the temperature of the thermochemical treatment of biomass increases the electrical conductivity of chars [68]. This trend was observed in three of the tested materials—walnut shells, where an increase from 0.558 mS∙cm⁻¹ to 0.665 mS∙cm⁻¹ was observed; orange peels, with an increase from 1.291 mS∙cm⁻¹ to 1.491 mS∙cm⁻¹; and peach stones, with an increase from 0.434 mS∙cm⁻¹ to 0.514 mS∙cm⁻¹. EC for apple wood chips decreased from 0.848 mS∙cm⁻¹ to 0.356 mS∙cm⁻¹. This was probably due to the low degree of carbonization for this sample (measured by the carbon content in relation to the raw biomass) [69], which was also indicated by the low degree of torrefaction. The phenomenon of large fluctuations in this parameter during the thermochemical treatment of biomass is also present in the literature. Zhang et al. [70] obtained different values of electrical conductivity for biochars produced from agricultural substrates during pyrolysis at twelve temperatures. For the highest temperatures, however, the trend was increasing.

3.5. Hydraulic Parameter Determination

Table 8. Wettability of torreficates and dried materials.

Biomass Residue	Temp.	WDPT	Properties	WHC
	°C	s	-	%
Walnut shells	105	25 a ± 2	slightly hydrophobic	129 b ± 11
	200	7020 b ± 635	extremely hydrophobic	79 ab ± 28
Orange peels	105	3 a ± 1	hydrophilic	819 e ± 36
	200	3500 ab ± 173	severely hydrophobic	495 d ± 22
Peach stones	105	39 a ± 3	slightly hydrophobic	73 ab ± 10
	200	1800 a ± 520	severely hydrophobic	55 a ± 2
Apple wood chips	105	34 a ± 3	slightly hydrophobic	288 c ± 26
	200	15,200 c ± 4257	extremely hydrophobic	245 c ± 15

The same signs (a–e) in a column mean no statistically significant changes.

shows the tests for materials’ hydrophobicity and water-holding capacities. The thermochemical processing of the materials at 200 °C improved the nature of the materials from hydrophilic or slightly hydrophobic to extremely or strongly hydrophobic based on WDPT. This phenomenon is commonly observed during the processes of torrefaction and low-temperature pyrolysis. It has been proven that as the temperature of the torrefaction process increases, an increasing number of hydroxyl bonds are dissolved and dehydrated; thus, the material becomes hydrophobic [71,72,73,74,75].

Torrefaction decreased the water-holding capacity parameter in all cases; however, the values of this parameter were highly variable depending on the substrate and even differed several times. The highest WHC was observed for unprocessed orange peel (819% ± 36%) and the lowest was observed for torrefied peach stones (55% ± 2%). There are reports in the literature that water retention capacity decreases with increasing pyrolysis temperature (including dried materials) [76], but there are also contrary reports [31]. The results obtained in the literature and in this experiment may suggest that the value of the WHC parameter varies depending on the type of substrate, its chemical composition, and the degradation of specific compounds by thermal decomposition, but this phenomenon requires further analysis. Hien et al. [77] suggest that changes in the WHC parameter are caused by changes in the physicochemical properties of biochar by thermochemical treatment—in particular parameters such as the specific surface area, total pore volume, and cation exchange capacity. In this experiment, high values of WHC correlation with AC, ε, and EC were obtained, which may also suggest their interdependence.

3.6. Odor Adsorption Test

Table 9. Share of the profiles of aromatic organic compounds in the extracted paper aroma carriers.

Biomass Residue	Temp.	2,3-Dimethylpyrazine	2,3,5-Trimethylpyrazine	Indole
	°C	%	%	%
Walnut shells	105	1.40 ab ± 0.40	20.80 a ± 7.75	77.74 a ± 8.14
	200	1.95 b ± 0.48	21.99 a ± 6.84	76.06 a ± 6.37
Orange peels	105	0.50 ab ± 0.65	25.55 a ± 7.11	73.95 a ± 7.58
	200	5.14 c ± 1.15	24.23 a ± 6.22	70.63 a ± 5.23
Peach stones	105	0.86 ab ± 0.46	29.15 a ± 3.31	72.50 a ± 4.58
	200	1.09 ab ± 0.84	26.42 a ± 4.35	68.62 a ± 5.82
Apple wood chips	105	n.d. a	23.67 a ± 3.95	76.33 a ± 3.95
	200	n.d. a	21.16 a ± 3.08	78.84 a ± 3.08

n.d.—not detected, the same signs (a–c) in a column mean no statistically significant changes.

shows the share of the profiles of tested odors in the extracted paper aroma carriers. The percentage of aromatic organic compound profiles in the extracted paper aroma carriers did not change significantly with respect to 2,3,5-trimethylopyrazine and indole; only for the 2,3-dimethylpyrazine content were statistically significant differences between the samples noted. It was observed that the concentrations of compounds increased with evaporation temperatures according to the hierarchy: 2,3-dimethylpyrazine (evaporation temperature: 156 °C) < 2,3,5-trimethylpyrazine (evaporation temperature: 172 °C) < indole (evaporating temperature: 254 °C).

Table 10. Content of aromatic compounds in the extracted torreficates and dried materials.

Biomass Residue	Temp.	2,3-Dimethylpyrazine	2,3,5-Trimethylpyrazine	Indole
	°C	µg∙mL−1	µg∙mL−1	µg∙mL−1
Walnut shells	105	n.d.	n.d.	14.87 a ± 3.55
	200	n.d.	n.d.	n.d. a
Orange peels	105	n.d.	n.d.	15.92 a ± 14.57
	200	n.d.	n.d.	45.64 a ± 40.02
Peach stones	105	n.d.	n.d.	27.18 a ± 11.61
	200	n.d.	n.d.	61.26 a ± 49.55
Apple wood chips	105	n.d.	n.d.	24.50 a ± 10.98
	200	n.d.	n.d.	n.d. a

n.d.—not detected. the same sign (a) in a column mean no statistically significant changes.

shows the content of aromatic compounds in the extracted torreficates and dried materials after the adsorption test. Supplementary Material S1 also includes the indole adsorption removal rate. The only compound found in the materials was indole. Unfortunately, 2,3-dimethylpyrazine and 2,3,5-trimethylpyrazine were not detected in any case, which may suggest that low-temperature torreficates and raw biomass from the agri-food industry are not suitable adsorbents for these compounds. In two cases, torrefaction significantly improved the indole adsorption property (orange peels and peach stones), and in these two cases, indole was lowered to a level under the detection limit. It is worth noting that the highest indole concentrations were noticeable for torrefied biomass. These results may suggest that in the case of limiting indole emissions, it is not necessary to process the material under pyrolytic range temperatures because the filter bed may be dried materials and low-temperature biochar/torreficate. However, the type of biomass from which biochar/torreficate may be produced should be selected with care because the conducted research showed several instances of differences between the evaluated materials. Additionally, due to the initial character of the research in this area, the aim was to check and answer the question of whether torreficates can retain odor substances. In order to better define this phenomenon in the future, however, it would be necessary to determine the differences in sorption efficiencies between low-temperature biochars/torreficates and commercially used activated carbons/zeolites.

To determine the role and importance of physical–hydraulic parameters in the odor adsorption test on low-temperature torrefaction, a heatmap for the linear correlation between the parameters is shown in Figure 2. The highest correlations with indole adsorption ability were observed for VMC (volatile matter content), L_S (mean pore size), FCC (fixed carbon content), and ρ_S (specific density). These results were slightly different from those of Hwang et al. [34], who, while examining the effectiveness of various biochars in the removal of odorous volatile organic compounds (including indole) emitted from pig manure, suggested that the properties of biochar such as volatile substances, the fixed carbon content, ash content, pH, and BET surface do not play a significant role in the adsorption of these compounds. However, the research conducted so far has focused primarily on determining the adsorption capacity of odors by biochars generated at higher temperatures [78,79]. The obtained results may suggest that physical properties such as L_S and ρ_S in particular may determine the indole adsorption ability of torreficates. This may also be an interesting addition to the commonly accepted mechanism for biochars formed at high temperatures whereby increasing the temperature of the process increases the surface area and pore distribution, making adsorption more effective [80]. However, at low temperatures, as can be seen, the BET surface does not change significantly and may even sometimes be smaller compared with that of raw biomass, which is likely due to pore plugging by melt phases inside the material during degassing as was demonstrated and explained by Granados et al. in their experiment [81].

It should also be mentioned that in this experiment, carbon dioxide was used as a carrier for the evaporating aromatics. Thus, the adsorption of carbon dioxide in the pores of torreficates and dried materials could disturb the deposition of aromatic compounds on their surfaces, blocking the availability of pores for odorants. To confirm this hypothesis, future experiments should be carried out using an inert gas as a carrier of evaporating aromatic compounds. Our analysis showed that if indole odors and carbon dioxide are present in the waste gas stream, low-temperature biochars/torreficates can be used as adsorbents to reduce the odor. However, it is also worth noting that according to [82], the characteristics of molecules, in particular their polarity, hydrophobicity, size, and aromaticity, may determine the adsorption of pollutants on biochar. Process duration and the dimensions of the filter bed are also recommended for evaluation in further studies. In the present experiment, 60 min was used; however, the high concentrations of odorants present on the extracted fragrance paper may suggest that the process duration time was too short for all the aromatics to evaporate. Additionally, the filtration bed used in the study was composed of materials of the same weight. The ability to absorb odors can be influenced by both the dose of biochar used and the volume of the bed; thus, the key in further research would be to use: (a) several of the same doses of biochar (by weight) and (b) different weight doses of biochar, which would have the exact dimensions and volume of the filter bed.

4. Conclusions

This study presented the results of research on the ability to absorb odors (indole, 2,3-dimethylpyrazine, and 2,3,5-trimethylpyrazine) by dried materials and torreficates produced at temperatures of 200 °C. The physical–chemical and physical–hydraulic parameters of the products were also determined, including the volatile matter content, ash content, fixed carbon content, pH, electrical conductivity, higher heating value, energy densification ratio, energy yield, hydrophobicity, water-holding capacity, specific density, porosity, bulk density, specific surface area, pore size, and pore volume. Unfortunately, 2-3-dimethylpyrazine and 2,3,5-trimethylpyrazine were not detected in any of the materials after the adsorption test was performed.

In the case of indole, a chromatographic analysis showed that the greatest sorption of this compound was achieved by low-temperature torrefied products (200 °C) produced from orange peels and peach stones. A high indole content was also noted in the dried materials of all the tested types of biomass. A statistical analysis showed that the greatest correlations between indole retention capacity and the physical-hydraulic parameters of biochar were found between the physical properties, especially the average pore size and specific density. This may suggest directions for further research in the context of the isolation of key properties that may determine the suitability of torreficates for reducing odor emissions and improving their sorption characteristics, making solutions more economically effective. However, it seems necessary to continue research in this area to explain the mechanisms of adsorption and the correlation between the structure of torreficates, their physical–chemical properties, and the parameters of the thermal treatment of raw material.