Evaluation of Hydrogen Yield Evolution in Gaseous Fraction and Biochar Structure Resulting from Walnut Shells Pyrolysis

: Conversion experiments of wet and dry walnut shells were performed, the inﬂuence of moisture content on the hydrogen yield in the gas fraction was estimated and the resulted biochar structure was presented. Measurements of the biochar structures were performed using X-ray di ﬀ raction and scanning electron microscopy methods. The results demonstrate that heating rate played a key role in the pyrolysis process and inﬂuenced the biochar structure. Under fast heating rate, the interactions between the water vapors released and other intermediate products, such as biochar was enhanced and consequently more hydrogen was generated. It could also be observed that both biochar samples, obtained from wet and dry walnut shells, had an approximately smooth surface and are di ﬀ erent from the rough surface of the raw walnut shell, but there are not obvious di ﬀ erences in shape and pores structure between the two biochar samples. The increasing of the biochar surface area versus pyrolysis temperature is due tothe formation of micropores in structure. The biochar shows a surface morphology in the form of particles with rough, compact and porous structure. In addition the biochar structure conﬁrmed that directly pyrolysis of wet walnut shells without predried treatment has enhanced the hydrogen content in the gas fraction. had favored an increasing of hydrogen production. This could be justiﬁed by that the water vapors released from the humidity of the wet walnut shells sample would react with the intermediate compounds such as volatile and biochar, to for much more H 2 .


Introduction
The main organic components in biomass (i.e., including agricultural residue) and their chemical structure have a substantial role in improving of the processes used to produce fuels and chemicals [1][2][3]. The lignocellulose from biomass contains four major components and these are: cellulose, hemicellulose, lignin and extractives [3]. The first three types of components have large molecular weights and constitute the main mass of biomass, while the extractives are compounds of smaller molecular size, and are in a low amount [4,5]. Hydrogen can be obtained by biomass pyrolysis and gasification but these technologies need further improvement and development [6][7][8].Hydrogen production from biomass brings some advantages [9] such as reducing the cost to oil imports, a level of stable price and the balance of CO 2 is stabilized by ≈25-30% [10].
The H 2 production from different biomass sources is based on thermochemical and biological processes. The thermochemical process has a higher efficiency (η ≈ 50%) and the cost is low [11,12].
Thermocatalytic biomass conversion includes pyrolysis, gasification and reforming processes while the biological route refers to the fermentative process [13][14][15][16]. These conversion processes produce gases such as, H 2 , CO 2 , CO, CH 4 and also light gaseous hydrocarbon. The use of catalyst processes together with separation techniques could help to increase the yield of hydrogen in the gaseous fraction [17]. The mixture of H 2 and CO, known as syngas, could be used as an intermediate for the obtaining of gasoline by Fischer-Tropsch synthesis [15,16]. Although, research has been focused on how the different conversion processes could be improved to increase the productivity, there is also high attention on the environmental impact of using the different technical routes for hydrogen-rich syngas production [14,17]. There are extensive review studies on hydrogen production by CH 4 conversion, agricultural residues by fermentation process [18], photochemical conversion of hydrogen sulfide and thermochemical biomass conversion [19][20][21]. None of these reviews referred to the production of hydrogen-rich gas by thermal biomass conversion using their own moisture content.
At temperatures in the range of 500-800 • C, the biomass suffers a thermal degradation and results in a liquid and gaseous fraction that usually contains bio-oil product and CO, H 2 , CO 2 , CH 4 , H 2 O and other hydrocarbons (C 2 -C 7 ) as gases [11].Composition of gaseous product resulted from the biomass gasification depends on the type of reactor, gasifying agent and the feedstock composition [22][23][24]. The conversion of biomass is generally given by the reaction: Biomass + heat + gasifying agent (air, steam)→ CO, CO 2 , H 2 O, H 2 , CH 4 + C 2-7 + tar + char + ash (1) In a pyrolysis process of the biomass, the first step is a thermochemical degradation of the main components (cellulose, hemicellulose and lignin) and results in biochar, liquid (tar) and volatiles [4,25,26]. Further, the biochar and tar gasification reactions could appear and other equilibrium reactions to take place. Figure 1 shows the possible products resulted from a pyrolysis process. During the biomass conversion together with the gaseous phase, it formed tar and its separation and removing is difficult to be achieved using only a physical method [27]. The products distribution and the composition of gaseous fraction depend on some important parameters, such as conversion temperature, the raw biomass used and the type of reactor [28]. The gasification process ensures the possibility to convert biomass feedstock into clean gas fuel/synthesis gas [29,30]. The synthesis gas is mainly formed of H 2 and CO, and it is known as bio-syngas if is obtained from biomass [4]. Bio-syngas can be obtained from biomass by catalytic and non-catalytic processes, and also using a gasification agent (air/steam/CO 2 ) [31].
Energies 2020, 13, x FOR PEER REVIEW 2 of 17 focused on how the different conversion processes could be improved to increase the productivity, there is also high attention on the environmental impact of using the different technical routes for hydrogen-rich syngas production [14,17]. There are extensive review studies on hydrogen production by CH4conversion, agricultural residues by fermentation process [18], photochemical conversion of hydrogen sulfide and thermochemical biomass conversion [19][20][21]. None of these reviews referred to the production of hydrogen-rich gas by thermal biomass conversion using their own moisture content. At temperatures in the range of 500-800 °C, the biomass suffers a thermal degradation and results in a liquid and gaseous fraction that usually contains bio-oil product and CO, H2, CO2, CH4, H2O and other hydrocarbons (C2-C7) as gases [11].Composition of gaseous product resulted from the biomass gasification depends on the type of reactor, gasifying agent and the feedstock composition [22][23][24]. The conversion of biomass is generally given by the reaction: Biomass + heat +gasifying agent (air, steam)→ CO,CO2, H2O, H2, CH4 + C2-7 + tar +char + ash (1) In a pyrolysis process of the biomass, the first step is a thermochemical degradation of the main components (cellulose, hemicellulose and lignin) and results in biochar, liquid (tar) and volatiles [4,25,26]. Further, the biochar and tar gasification reactions could appear and other equilibrium reactions to take place. Figure 1 shows the possible products resulted from a pyrolysis process. During the biomass conversion together with the gaseous phase, it formed tar and its separation and removing is difficult to be achieved using only a physical method [27]. The products distribution and the composition of gaseous fraction depend on some important parameters, such as conversion temperature, the raw biomass used and the type of reactor [28]. The gasification process ensures the possibility to convert biomass feedstock into clean gas fuel/synthesis gas [29,30]. The synthesis gas is mainly formed of H2 and CO, and it is known as bio-syngas if is obtained from biomass [4]. Bio-syngas can be obtained from biomass by catalytic and non-catalytic processes, and also using a gasification agent (air/steam/CO2) [31]. The biomass conversion using steam as a gasification agent is a promising way used for production of gas with high H2 content [25]. In recent years, H2 production by biomass pyrolysis and/or gasification became a subject intensively researched in the domain of biomass utilization [32]. Previous studies [33,34] have reported that the introduction of steam during biomass conversion process lead to an increasing of H2 amount due to the tar and higher molecular weight hydrocarbons The biomass conversion using steam as a gasification agent is a promising way used for production of gas with high H 2 content [25]. In recent years, H 2 production by biomass pyrolysis and/or gasification became a subject intensively researched in the domain of biomass utilization [32]. Previous studies [33,34] have reported that the introduction of steam during biomass conversion process lead to an increasing of H 2 amount due to the tar and higher molecular weight hydrocarbons Energies 2020, 13, 6359 3 of 17 reforming, and the water gas shift reaction [2,5,6].Using steam as an oxidizer, the following chemical reaction can be written: CH 1 (2) with the biomass chemical formula, which was calculated using elemental analysis, Table 1. According to reaction (2), a minimum water (steam)/biomass (S/B) ratio of 0.968 is established for the theoretical maximum H 2 yield. The recent studies refer to modification of the operating parameters, the use of different types of reactors and catalytic processes, etc. [32][33][34]. Unfortunately, no studies have been performed related to the reduction of costs due to the way the steam is supplied [35] even if this consumes a big quantity of energy from the whole process. Usually, the raw biomass contains a big fraction of moisture, which even exceeds 40-50 wt.% (wet basis). In the conventional biomass steam gasification (CBSG), own moisture from biomass is removed before conversion by a predrying step and then the predried biomass is subjected to the gasification/pyrolysis process [36][37][38][39]. The steam used as a gasification agent is usually provided by an auxiliary source such as a generator or boiler. Both the predrying step and steam supply need energy consumption and for this reason the CBSG process is expensive and complex. Considering these aspects, in this work a method of H 2 rich gas production by agricultural residue pyrolysis in an autogenerated moisture atmosphere was experimented. In this method, the agricultural residue, including walnut shells, with high moisture content was used as starting raw materials for a pyrolysis reactor. Walnut shells were used because it is a biomass residue that results in large quantities in the geographical area, contains high moisture content and is not valorized The method has used its own moisture from the initial biomass sample as a reactant to combine with the intermediate products resulted from the pyrolysis and in this way to produce additional gas (hydrogen). Additionally, to obtain some basic useful information, the influence of several parameters such as pyrolysis reactor temperature, heating rate, moisture content in the agricultural residue samples and stripping gas flow on the efficiency of the pyrolysis process was investigated and evaluated. The characteristics of the resulted biochar by the pyrolysis process were also presented.

Materials and Sample Preparation
In this study, the residual biomass, including walnut shells (WSs), was employed as raw materials. The raw walnut shells in hemispherical shape with about 25-35 mm diameter and 2-3 mm thickness were supplied by the Research Station for Fruit Growing Valcea ( Figure 2). The walnut shells were collected from the hill area of Valcea county, approximately 35 km away from the town of Rm. Valcea (located on the parallel 45 • 06 N).  Table 1 shows the composition and elemental analysis for walnut shells. Walnut shells were first crushed, then grinded and sieved to obtain particle size ≤ 1 mm. The proximate analysis of raw walnut shells was done according to ASTM D3174-04 for ash analysis and ASTMD3175-89a for volatile matter. The ultimate analysis to determine the quantity of C, H and N (O calculated from the difference) was made according to ASTM D5291-96 using a FLASH-2000 Elemental Analyzer (Table 1). The scanning electron micrograph (SEM) analysis of the raw walnut shells and biochar samples obtained after pyrolysis was performed by using Jeol JSM-6060.
All chemicals used in this study were of analytical grade. To study the effect of moisture content, the walnut shells were dried naturally, by sun drying to different moisture contents: (a) wet walnut shell sample (WSW), the as-received walnut shell, with a moisture content of ≈48.15 w.% (wet basis); (b) a partially predried fraction of the as received wet walnut shell sample (WSPD), with a moisture content of ≈33.68 wt.% (wet basis) and (c) totally predried walnut shell sample (WSTD) with a moisture content of ≈7.85 wt.% (wet basis).The moisture content was determined according to ASTM D4442-07.
A measurement of specific surface area of the biochar samples has been performed by N2 adsorption at −196.15 °C, using a surface analyzer (Quantachrome Instruments, Corporate Headquarters USA). The total volume of the pore (Vt) was determined from the quantity of N2 vapor adsorbed at relative pressure of 0.95. The total surface area (St) was determined using BET equation in the relative pressure range 0.05-0.35. The limits between micro-and mesopores and also meso-and macropores were taken following the IUPAC nomenclature and the Barrett-Joyner-Halenda (BJH) method was used to establish the mesopore size distribution. XRD analysis was performed using D/max-2200/PC, Rigaku, Japan, and copper KR radiation (40 kV, 20 mA) as the X-ray source. Scanning electron microscopy (SEM) was performed using a microscope JSM-7500 F (JOEL-Japan), operated at 10 kV.

Experimental Pyrolysis Tests
Two types of reactors were used for the pyrolysis experimental tests, one was a stainless steel reactor, used for the slow pyrolysis tests and another one was a quartz tube reactor, used for the fast pyrolysis tests, each depending on the process parameters required. Figure 3a-c shows a schematic diagram of the experimental system and the dimensions for the stainless steel reactor and quartz reactor.  Table 1 shows the composition and elemental analysis for walnut shells. Walnut shells were first crushed, then grinded and sieved to obtain particle size ≤ 1 mm. The proximate analysis of raw walnut shells was done according to ASTM D3174-04 for ash analysis and ASTMD3175-89a for volatile matter. The ultimate analysis to determine the quantity of C, H and N (O calculated from the difference) was made according to ASTM D5291-96 using a FLASH-2000 Elemental Analyzer ( Table 1). The scanning electron micrograph (SEM) analysis of the raw walnut shells and biochar samples obtained after pyrolysis was performed by using Jeol JSM-6060.
All chemicals used in this study were of analytical grade. To study the effect of moisture content, the walnut shells were dried naturally, by sun drying to different moisture contents: (a) wet walnut shell sample (WS W ), the as-received walnut shell, with a moisture content of ≈48.15 w.% (wet basis); (b) a partially predried fraction of the as received wet walnut shell sample (WS PD ), with a moisture content of ≈33.68 wt.% (wet basis) and (c) totally predried walnut shell sample (WS TD ) with a moisture content of ≈7.85 wt.% (wet basis).The moisture content was determined according to ASTM D4442-07.
A measurement of specific surface area of the biochar samples has been performed by N 2 adsorption at −196.15 • C, using a surface analyzer (Quantachrome Instruments, Corporate Headquarters USA). The total volume of the pore (Vt) was determined from the quantity of N 2 vapor adsorbed at relative pressure of 0.95. The total surface area (St) was determined using BET equation in the relative pressure range 0.05-0.35. The limits between micro-and mesopores and also meso-and macropores were taken following the IUPAC nomenclature and the Barrett-Joyner-Halenda (BJH) method was used to establish the mesopore size distribution. XRD analysis was performed using D/max-2200/PC, Rigaku, Japan, and copper KR radiation (40 kV, 20 mA) as the X-ray source. Scanning electron microscopy (SEM) was performed using a microscope JSM-7500 F (JOEL-Japan), operated at 10 kV.

Experimental Pyrolysis Tests
Two types of reactors were used for the pyrolysis experimental tests, one was a stainless steel reactor, used for the slow pyrolysis tests and another one was a quartz tube reactor, used for the fast pyrolysis tests, each depending on the process parameters required. Figure 3a-c shows a schematic diagram of the experimental system and the dimensions for the stainless steel reactor and quartz reactor.
The experimental system consisted of four parts: (I) the stainless steel reactor, (II) the quartz tube reactor, (III) the gas clean and collection system and (IV) the temperature control system. Slow pyrolysis tests were carried out in the stainless steel reactor (4). The walnut shell sample (100 g, air-dry basis) was placed in the reactor (350 mm length × 25.4 mm i.d.). Next, the reactor was stripped with the inert gas (N 2 ) for 5-10 min. The experiments were started by heating the reactor at a temperature in the range of 300-800 • C, at a heating rate of 5 • C/min, using electric furnace of 1100 • C. For temperature value comparison, the temperatures in the middle of the biomass bed and of the reactor wall were measured using two thermocouples, respectively. The volatile products were continuously removed from the reactor by stripping with N 2 gas. The volatile products were passed through the two coolers and then by a glass wool filter to remove the impurities from the volatile products, as shown in Figure 3. The gas fraction was stored into a gas tank, and the volume was determined with a wet gas-meter. After the reactor temperature was reached, the reactor was maintained for 20 min at that temperature, after which the heating was interrupted and the reactor was naturally cooled. The experimental system consisted of four parts: (I) the stainless steel reactor, (II) the quartz tube reactor, (III) the gas clean and collection system and (IV) the temperature control system. Slow pyrolysis tests were carried out in the stainless steel reactor (4). The walnut shell sample (100 g, air-dry basis) was placed in the reactor (350 mm length × 25.4 mm i.d.). Next, the reactor was stripped with the inert gas (N2) for 5-10 min. The experiments were started by heating the reactor at a temperature in the range of 300-800 °C, at a heating rate of 5 °C/min, using electric furnace of 1100 °C. For temperature value comparison, the temperatures in the middle of the biomass bed and of the reactor wall were measured using two thermocouples, respectively. The volatile products were continuously removed from the reactor by stripping with N2 gas. The volatile products were passed through the two coolers and then by a glass wool filter to remove the impurities from the volatile products, as shown in Figure 3. The gas fraction was stored into a gas tank, and the volume was determined with a wet gas-meter. After the reactor temperature was reached, the reactor was maintained for 20 min at that temperature, after which the heating was interrupted and the reactor was naturally cooled.
Fast pyrolysis tests were performed using the reactor made of quartz. To obtain a fast heating rate, the amount of used biomass was only 10 g (air-dry basis). Next, a ceramic boat containing biomass sample was put into the reactor (120 mm length × 60 mm i.d), the reactor was stripped with the inert gas (N2) for 5-10 min and then was heated in a N2 atmosphere using an electric furnace, until the desired temperature (the range of 500-800 °C) at a heating rate of 50 °C/min, and held for 20 min, then the heating was interrupted. The gas was cleaned and then collected in the same way as that in the slow pyrolysis tests. The composition of the gaseous phase (H2, CH4, CO and CO2), was established using a gas chromatograph type Varian CP-3800 with TCD and FID detectors, equipped with three columns: (i) a 5 Å molecular sieve column,(ii) a Porapak N column and (iii) a Chromosorb column, and using argon as carrier gas. The amount of biochar was determined by weighing with an analytical balance having a precision of ±0.1 mg. Fast pyrolysis tests were performed using the reactor made of quartz. To obtain a fast heating rate, the amount of used biomass was only 10 g (air-dry basis). Next, a ceramic boat containing biomass sample was put into the reactor (120 mm length × 60 mm i.d), the reactor was stripped with the inert gas (N 2 ) for 5-10 min and then was heated in a N 2 atmosphere using an electric furnace, until the desired temperature (the range of 500-800 • C) at a heating rate of 50 • C/min, and held for 20 min, then the heating was interrupted. The gas was cleaned and then collected in the same way as that in the slow pyrolysis tests. The composition of the gaseous phase (H 2 , CH 4 , CO and CO 2 ), was established using a gas chromatograph type Varian CP-3800 with TCD and FID detectors, equipped with three columns: (i) a 5 Å molecular sieve column,(ii) a Porapak N column and (iii) a Chromosorb column, and using argon as carrier gas. The amount of biochar was determined by weighing with an analytical balance having a precision of ±0.1 mg.
The carbon conversion quantifies the efficiency of the pyrolysis process. The carbon conversion efficiency (η cc ) was established in terms of the moles of carbon in the walnut shells that are transformed to carbon containing gases [36], in accordance with Equation (3): where M product represent the carbon moles in gas fraction (CO, CO 2 and CH 4 ), and M biomass represent the carbon moles in the walnut shell sample.

The Effect of Reactor Temperature on the Pyrolysis Process
It is known that temperature plays a very important role in the biomass conversion (pyrolysis/gasification) process. In the present study, the reactor temperature was varied in the range of 300-800 • C with a step of 100 • C. The reactor temperature and the obtained results are shown in Figures 4 and 5.
Where Mproduct represent the carbon moles in gas fraction (CO, CO2 and CH4), and Mbiomass represent the carbon moles in the walnut shell sample.

The Effect of Reactor Temperature on the Pyrolysis Process
It is known that temperature plays a very important role in the biomass conversion (pyrolysis/gasification) process. In the present study, the reactor temperature was varied in the range of 300-800 °C with a step of 100°C. The reactor temperature and the obtained results are shown in Figures 4 and 5.  As can be observed in Figure 4, the gas yield increased with the reactor temperature, while the biochar yield decreased. This behavior was due to the cracking reactions as a result of the temperature increasing into the reactor, the consequence was that more gas was produced. The tar yield increased first and then decreased, and a maximum of 76.18 wt.% appeared in the range of 500-600 °C, this temperature value indicating a suitable one for liquid fraction production (water and tar). These results are similar to the previous research [37,38], which used predried biomass. In addition, it was found that 86.05 wt.% of walnut shells put into the reactor was converted in volatile products (gas, tar and water) at 500 °C, and this percent increased slightly with the temperature from a reactor, being 87.08 wt.% at 800 °C. These results show that an important part of the volatile products were released from the biomass sample before 600 °C, and after that, the increasing of the temperature had a little effect on the walnut shells decomposition. Where Mproduct represent the carbon moles in gas fraction (CO, CO2 and CH4), and Mbiomass represent the carbon moles in the walnut shell sample.

The Effect of Reactor Temperature on the Pyrolysis Process
It is known that temperature plays a very important role in the biomass conversion (pyrolysis/gasification) process. In the present study, the reactor temperature was varied in the range of 300-800 °C with a step of 100°C. The reactor temperature and the obtained results are shown in Figures 4 and 5.  As can be observed in Figure 4, the gas yield increased with the reactor temperature, while the biochar yield decreased. This behavior was due to the cracking reactions as a result of the temperature increasing into the reactor, the consequence was that more gas was produced. The tar yield increased first and then decreased, and a maximum of 76.18 wt.% appeared in the range of 500-600 °C, this temperature value indicating a suitable one for liquid fraction production (water and tar). These results are similar to the previous research [37,38], which used predried biomass. In addition, it was found that 86.05 wt.% of walnut shells put into the reactor was converted in volatile products (gas, tar and water) at 500 °C, and this percent increased slightly with the temperature from a reactor, being 87.08 wt.% at 800 °C. These results show that an important part of the volatile products were released from the biomass sample before 600 °C, and after that, the increasing of the temperature had a little effect on the walnut shells decomposition. As can be observed in Figure 4, the gas yield increased with the reactor temperature, while the biochar yield decreased. This behavior was due to the cracking reactions as a result of the temperature increasing into the reactor, the consequence was that more gas was produced. The tar yield increased first and then decreased, and a maximum of 76.18 wt.% appeared in the range of 500-600 • C, this temperature value indicating a suitable one for liquid fraction production (water and tar). These results are similar to the previous research [37,38], which used predried biomass. In addition, it was found that 86.05 wt.% of walnut shells put into the reactor was converted in volatile products (gas, tar and water) at 500 • C, and this percent increased slightly with the temperature from a reactor, being 87.08 wt.% at 800 • C. These results show that an important part of the volatile products were released from the biomass sample before 600 • C, and after that, the increasing of the temperature had a little effect on the walnut shells decomposition. It should be noted that the liquid fraction yield was high in the all ranges of temperature analyzed. At 800 • C, the yield of liquid fraction even reached ≈68.9 wt.%. Such a high liquid yield indicated that the reactions between the intermediate components (volatile and char) and the generated water vapors were limited, possibly due to the low heating rate and also due to the presence of stripping gas. These results are in agreement with data presented in the literature [34,35]. In Figure 5 the effect of the temperature on gas product compositions shown. As can be seen, with the increasing of the temperature, the amounts of H 2 and CH 4 increased from 15.68 to 27.02 vol.% and from 8.57 to 13.37 vol.%, respectively. CO amount had a smaller decreasing from 39.36 to 36.71 vol.% between 500 and 800 • C, while CO 2 content decreased from 48.85 to 23.04 vol.% with the temperature increasing. The increasing of the gas yield and the H 2 content indicates that the H 2 yield increased with the temperature and this can be attributed to the cracking reactions and steam reforming process at a higher temperature to form much more H 2 [35,37].

The Effect of Moisture Content on Hydrogen Yield
In this study, walnut shell samples with different moisture contents (WS W , WS PD and WS TD ) were tested in the pyrolysis process. For this, two types of reactor were used, to provide different heating rates. Figure 6 presents the influence of the humidity on the hydrogen yield. The gas yield and the conversion efficiency of the biochar and the gas composition for each experimental test including CO, H 2 , CO 2 and CH 4 are shown in Figure 6 and Table 2. The pyrolysis of WS W and WS PD samples represents the biomass pyrolysis process in which the wet and predried biomass were used as initial raw materials. In comparison with the WS TD sample pyrolysis, it was found that the hydrogen yield increased with the increasing of the humidity content, this indicating that a direct pyrolysis of the wet biomass sample without a predrying treatment had favored an increasing of hydrogen production. This could be justified by that the water vapors released from the humidity of the wet walnut shells sample would react with the intermediate compounds such as volatile and biochar, to for much more H 2 .
Energies 2020, 13, x FOR PEER REVIEW 7 of 17 It should be noted that the liquid fraction yield was high in the all ranges of temperature analyzed. At 800 °C, the yield of liquid fraction even reached ≈68.9 wt.%. Such a high liquid yield indicated that the reactions between the intermediate components (volatile and char) and the generated water vapors were limited, possibly due to the low heating rate and also due to the presence of stripping gas. These results are in agreement with data presented in the literature [34,35]. In Figure 5 the effect of the temperature on gas product compositions shown. As can be seen, with the increasing of the temperature, the amounts of H2 and CH4 increased from 15.68 to 27.02 vol.% and from 8.57 to 13.37 vol.%, respectively. CO amount had a smaller decreasing from 39.36 to 36.71 vol.% between 500 and 800 °C, while CO2 content decreased from 48.85 to 23.04 vol.% with the temperature increasing. The increasing of the gas yield and the H2 content indicates that the H2 yield increased with the temperature and this can be attributed to the cracking reactions and steam reforming process at a higher temperature to form much more H2 [35,37].

The Effect ofMoisture Content on Hydrogen Yield
In this study, walnut shell samples with different moisture contents (WSW, WSPD and WSTD) were tested in the pyrolysis process. For this, two types of reactor were used, to provide different heating rates. Figure 6 presents the influence of the humidity on the hydrogen yield. The gas yield and the conversion efficiency of the biochar and the gas composition for each experimental test including CO, H2, CO2 and CH4 are shown in Figure 6 and Table 2. The pyrolysis of WSW and WSPD samples represents the biomass pyrolysis process in which the wet and predried biomass were used as initial raw materials. In comparison with the WSTD sample pyrolysis, it was found that the hydrogen yield increased with the increasing of the humidity content, this indicating that a direct pyrolysis of the wet biomass sample without a predrying treatment had favored an increasing of hydrogen production. This could be justified by that the water vapors released from the humidity of the wet walnut shells sample would react with the intermediate compounds such as volatile and biochar, to for much more H2.  Table 2. The influence of the moisture content on the gas yield, conversion efficiency (η) and gas composition (40 L/h striping gas flow rate) and statistical processing of experimental data obtained for the hydrogen content in the gas fraction, depending on the moisture content of the walnut shells samples.  In addition, with the increasing of the humidity content, the water vapors partial pressure increased and thus it improved hydrogen yield. The main reactions that took place during the pyrolysis of wet biomass sample could be as follows:

Parameter
Drying process: Wet Biomass + Heat → Water Vapors + Dry Biomass Generally, the hydrocarbons reforming reaction can be written as: Energies 2020, 13, 6359 8 of 17 The reforming reactions of methane: The water gas shift reaction: As can be seen in Figure 6 and Table 2, under slow pyrolysis conditions, the hydrogen yield and amount increased, while the gas yield and the efficiency (η) showed little variations with the moisture content from the biomass sample. However, under fast pyrolysis conditions, these variables increased more. With the increase of the moisture content from 7.85% to 48.15%, the hydrogen yield increased. The results obtained are similar to other studies [17,19,36]. Since this study investigated the evolution of hydrogen content in the gas fraction obtained, Table 2 shows the standard deviation from the determined average value (each average value representing the arithmetic average of at least three tests), and the dispersion. It can be seen that the standard deviation for the samples with the highest moisture content (48.15 wt.%) was ±0.2455 for slow pyrolysis and ±0.0086 for fast pyrolysis (the standard deviation was much smaller in the case of the fast pyrolysis process). Table 2. The influence of the moisture content on the gas yield, conversion efficiency (η) and gas composition (40 L/h striping gas flow rate) and statistical processing of experimental data obtained for the hydrogen content in the gas fraction, depending on the moisture content of the walnut shells samples. Comparing the results from both the slow and the fast pyrolysis, it could be observed that under fast pyrolysis conditions the influence of the moisture content on the pyrolysis process was stronger than that under slow pyrolysis conditions. This may be related with the interactions that take place between the generated water vapors and the intermediate compounds, such as volatile and biochar, produced from different heating rates. For the slow pyrolysis conditions, the water vapors generated, from moisture content, was not totally removed by striping gas before coming in contact with the intermediate compounds, due to the long time of drying and pyrolysis, thus, determining a poorer influence of moisture content on the subsequent pyrolysis process. In the case of the fast pyrolysis process, however, both the moisture evaporation and intermediate products generation (volatile and biochar) occur in a shorter time, which lead to improving the interactions between water vapors-volatile and the water vapors-biochar. Figure 7 shows the temperature curves of different WS samples during pyrolysis tests using the two types of reactors.

Parameter
Energies 2020, 13, x FOR PEER REVIEW 9 of 17 In the case of the fast pyrolysis process, however, both the moisture evaporation and intermediate products generation (volatile and biochar) occur in a shorter time, which lead to improving the interactions between water vapors-volatile and the water vapors-biochar. Figure 7 shows the temperature curves of different WS samples during pyrolysis tests using the two types of reactors. The temperature profiles show that the time duration of both drying and pyrolysis processes was reduced when the heating rate increased. Consequently, the direct use of the wet biomass sample will determine two different effects on the pyrolysis process: (i) the decreasing of the reactor temperature due to the drying step, which represents a disadvantage to the pyrolysis process, and (ii) the generated water vapors from moisture content can act like a reactant to improve the intermediate products decomposition (volatile and biochar), which represents an advantage to the pyrolysis process. Which of two effects will play a dominant role depends on the heating rate, but the obtained results demonstrated that under fast pyrolysis conditions the second effect dominated the whole process, whereas the first effect could be neglected. For the slow pyrolysis process, it also can be observed that the CO2 content shows a decreasing trend with humidity amount from biomass sample increases (Table 2) and this is an interesting behavior, but the interpretation is not very clear, and further study is required. However, the CO2 decreasing represents a useful parameter for improving the gas quality.

Effects of Stripping Gas Flow Rate on the Hydrogen Yield
In this study, the stripping gas flow rate was varied from 50 to 0 L/h and the hydrogen yield obtained from walnut shells pyrolysis with various flow rates is presented in Figure 8 and Table 3. The temperature profiles show that the time duration of both drying and pyrolysis processes was reduced when the heating rate increased. Consequently, the direct use of the wet biomass sample will determine two different effects on the pyrolysis process: (i) the decreasing of the reactor temperature due to the drying step, which represents a disadvantage to the pyrolysis process, and (ii) the generated water vapors from moisture content can act like a reactant to improve the intermediate products decomposition (volatile and biochar), which represents an advantage to the pyrolysis process. Which of two effects will play a dominant role depends on the heating rate, but the obtained results demonstrated that under fast pyrolysis conditions the second effect dominated the whole process, whereas the first effect could be neglected. For the slow pyrolysis process, it also can be observed that the CO 2 content shows a decreasing trend with humidity amount from biomass sample increases (Table 2) and this is an interesting behavior, but the interpretation is not very clear, and further study is required. However, the CO 2 decreasing represents a useful parameter for improving the gas quality.

Effects of Stripping Gas Flow Rate on the Hydrogen Yield
In this study, the stripping gas flow rate was varied from 50 to 0 L/h and the hydrogen yield obtained from walnut shells pyrolysis with various flow rates is presented in Figure 8 and Table 3.
The stripping gas flow rate presents an important effect on the hydrogen yield, this has increased with the decreasing of the stripping gas flow rate. A possible explanation may be that the higher flow rate of stripping gas increased the transport speed of uncondensed volatiles compounds from the reaction zone, decreasing secondary reactions, such as thermal cracking, repolymerization or recondensation [36,39].
In addition, for the direct pyrolysis of the WS W sample, the higher flow rate of stripping gas also quickly removes the generated water vapors from the reaction zone of the reactor, thus reducing the steam reforming of the intermediate products (volatile and biochar).
Using a lower stripping gas flow (N 2 ), more intermediate products (volatile and biochar) are transformed into gases, resulting in a higher degree of conversion and a higher amount of gaseous Energies 2020, 13, 6359 10 of 17 product. It can be also observed that hydrogen and the CO amount varied with the stripping gas flow rate, while for the rest of gases no important difference was seen.
In Figure 9 is shown the distributions of water and solid fractions for the WS W sample and its pyrolysis products. The high yield of gas related to the solid fraction contained in WS W shows that a lot of part of moisture amount reacted with the intermediate compounds (volatile and biochar).  The stripping gas flow rate presents an important effect on the hydrogen yield, this has increased with the decreasing of the stripping gas flow rate. A possible explanation may be that the higher flow rate of stripping gas increased the transport speed of uncondensed volatiles compounds from the reaction zone, decreasing secondary reactions, such as thermal cracking, repolymerization or recondensation [36,39].
In addition, for the direct pyrolysis of the WSW sample, the higher flow rate of stripping gas also quickly removes the generated water vapors from the reaction zone of the reactor, thus reducing the steam reforming of the intermediate products (volatile and biochar).
Using a lower stripping gas flow (N2), more intermediate products (volatile and biochar) are transformed into gases, resulting in a higher degree of conversion and a higher amount of gaseous product. It can be also observed that hydrogen and the CO amount varied with the stripping gas flow rate, while for the rest of gases no important difference was seen.
In Figure 9 is shown the distributions of water and solid fractions for the WSW sample and its pyrolysis products. The high yield of gas related to the solid fraction contained in WSW shows that a   Figure 9. The distribution of water and solid fractions for WSW and its pyrolysis products. The test was performed under the fast pyrolysis process without N2 stripping gas.
On the other hand, the fast pyrolysis of the WSTD sample, without stripping gas was also performed and the results are also shown in Table 3. The WSW pyrolysis released a higher hydrogen yield of 38.06 vol.%, than those of 29.43 vol.%, respectively, from WSTD, indicating a significant On the other hand, the fast pyrolysis of the WS TD sample, without stripping gas was also performed and the results are also shown in Table 3. The WS W pyrolysis released a higher hydrogen yield of 38.06 vol.%, than those of 29.43 vol.%, respectively, from WS TD , indicating a significant increasing of hydrogen yield. The modification of both hydrogen and carbon monoxide contents maybe due to the water gas shift reaction. The higher H 2 content of 38.06 vol.%, obtained in this study, was similar with that was obtained in other studies [7,10] at the same temperature value.

Biochar Yield and Its Characteristics
In Figure 10 is shown the biochar yield obtained in the slow and fast pyrolysis of WS W ,WS PD and WS TD samples, respectively. It can be seen that the moisture content from samples had also an influence on the biochar yield. Figure 9. The distribution of water and solid fractions for WSW and its pyrolysis products. The test was performed under the fast pyrolysis process without N2 stripping gas.
On the other hand, the fast pyrolysis of the WSTD sample, without stripping gas was also performed and the results are also shown in Table 3. The WSW pyrolysis released a higher hydrogen yield of 38.06 vol.%, than those of 29.43 vol.%, respectively, from WSTD, indicating a significant increasing of hydrogen yield. The modification of both hydrogen and carbon monoxide contents maybe due to the water gas shift reaction. The higher H2 content of 38.06 vol.%, obtained in this study, was similar with that was obtained in other studies [7,10] at the same temperature value.

Biochar Yield and Its Characteristics
In Figure 10 is shown the biochar yield obtained in the slow and fast pyrolysis of WSW,WSPD and WSTD samples, respectively. It can be seen that the moisture content from samples had also an influence on the biochar yield. With the increasing in the moisture content the biochar yield decreased, which is more evidence for the fast pyrolysis process. This behavior is due to the following reasons: (i) the changing of the internal biomass pores before and after drying step. Using conventional drying methods, raw biomass usually suffers a mass reduction process after drying treatment, which leads to a low gas diffusivity. In our tests, the drying rate, even for the slow pyrolysis process, was faster than those from conventional drying methods. This higher drying rate determined the creation of big channels and much more open pores, even if the structure of the biomass tissue was fragmented, and the release from solid particles of the generated organic molecules was facilitated during the pyrolysis/gasification process; (ii) the nascent char gasification with water vapors was amplified because the increase of the water vapors partial pressure as a result of higher moisture content from biomass samples and (iii) the shorter residence time of volatile products into biomass particles, determined from the improving gas With the increasing in the moisture content the biochar yield decreased, which is more evidence for the fast pyrolysis process. This behavior is due to the following reasons: (i) the changing of the internal biomass pores before and after drying step. Using conventional drying methods, raw biomass usually suffers a mass reduction process after drying treatment, which leads to a low gas diffusivity. In our tests, the drying rate, even for the slow pyrolysis process, was faster than those from conventional drying methods. This higher drying rate determined the creation of big channels and much more open pores, even if the structure of the biomass tissue was fragmented, and the release from solid particles of the generated organic molecules was facilitated during the pyrolysis/gasification process; (ii) the nascent char gasification with water vapors was amplified because the increase of the water vapors partial pressure as a result of higher moisture content from biomass samples and (iii) the shorter residence time of volatile products into biomass particles, determined from the improving gas permeability, as above shown, and a fast heating rate, decreased intraparticle volatile cracking and char deposition, the result being a lower biochar yield. The X-ray intensity curves of WS W sample and CWS w (char resulted from the WS W pyrolysis) and CWS TD (char resulted from the WS TD pyrolysis) are given in Figure 11 and SEM images are shown in Figure 12. The pyrolysis tests were carried out under slow heating conditions (slow pyrolysis experiment). permeability, as above shown, and a fast heating rate, decreased intraparticle volatile cracking and char deposition, the result being a lower biochar yield. The X-ray intensity curves of WSW sample and CWSw (char resulted from the WSW pyrolysis) and CWSTD (char resulted from the WSTD pyrolysis) are given in Figure 11 and SEM images are shown in Figure 12. The pyrolysis tests were carried out under slow heating conditions (slow pyrolysis experiment).  For the WSW sample, the X-ray intensity curve shows two peaks in low angle 2θ (range 5-36°), and these corresponded to the γ band (1) and to (002) band (2), respectively (see Figure 11a). After pyrolysis of WSW and WSTD samples, the γ band was not visible, had almost disappeared and the (002) band decreased, as can be seen in Figure 11b,c. The γ band certainly corresponded to aliphatic chains and the (002) band represented the aromatic compounds, this being in agreement with other studies [22,26]. The obtained results show that a lot part of aliphatic chains were decomposed in the pyrolysis process, while only some parts of the aromatic compounds were decomposed. In addition, it also can be seen that the intensity of CWSW curve was lower than that of CWSTD. This lower intensity curve of CWSW highlighted that much more aliphatic and aromatic groups were detached from biomass particles for the WSW pyrolysis than that for the WSTD pyrolysis and this was in agreement with the biochar yield changing, as can be observed in Figure 11. The SEM method was used to analysis the surface physical morphology of the raw walnut shell and biochar obtained after the pyrolysis test. Figure 12a shows the micrograph for the wet walnut shell used as the raw material in the pyrolysis process. The material shows a surface morphology in the form of particles with rough, compact and porous structure. Figure 12b,c shows the micrographs for the biochar obtained after pyrolysis of WSw and WSTD samples. It can be observed that both biochar samples (CWSw and CWSTD, respectively) had an approximately smooth surface and different from the rough surface of the raw walnut shell, but there was not obvious differences in shape and pores structure between the two biochar samples. The biochar yields, elemental analysis, surface area and micropore volume are shown in Table 4. For the WS W sample, the X-ray intensity curve shows two peaks in low angle 2θ (range 5-36 • ), and these corresponded to the γ band (1) and to (002) band (2), respectively (see Figure 11a). After pyrolysis of WS W and WS TD samples, the γ band was not visible, had almost disappeared and the (002) band decreased, as can be seen in Figure 11b,c. The γ band certainly corresponded to aliphatic chains and the (002) band represented the aromatic compounds, this being in agreement with other studies [22,26]. The obtained results show that a lot part of aliphatic chains were decomposed in the pyrolysis process, while only some parts of the aromatic compounds were decomposed. In addition, it also can be seen that the intensity of CWS W curve was lower than that of CWS TD . This lower intensity curve of CWS W highlighted that much more aliphatic and aromatic groups were detached from biomass particles for the WS W pyrolysis than that for the WS TD pyrolysis and this was in agreement with the biochar yield changing, as can be observed in Figure 11. The SEM method was used to analysis the surface physical morphology of the raw walnut shell and biochar obtained after the pyrolysis test. Figure 12a shows the micrograph for the wet walnut shell used as the raw material in the pyrolysis process. The material shows a surface morphology in the form of particles with rough, compact and porous structure. Figure 12b,c shows the micrographs for the biochar obtained after pyrolysis of WS w and WS TD samples. It can be observed that both biochar samples (CWS w and CWS TD , respectively) had an approximately smooth surface and different from the rough surface of the raw walnut shell, Energies 2020, 13, 6359 14 of 17 but there was not obvious differences in shape and pores structure between the two biochar samples. The biochar yields, elemental analysis, surface area and micropore volume are shown in Table 4. Table 4. The biochar yields, elemental analysis, surface area and micropore volume for walnut shell pyrolysis at a low heating rate (5 • C/min) and high heating rate (50 • C/min) and moisture content 7.85 wt.% at 800 • C.

Characteristic
Low In the pyrolysis process at a temperature below 400 • C, the biochar surface area did not change too much, due to an incomplete removal of volatile compounds from biomass. At a pyrolysis temperature in a range 400-500 • C, the biochar surface area began to increase and the increasing continued up to a pyrolysis temperature at least of 800 • C. This increase could be up to 600 m 2 /g. The increasing of the carbon surface area with the increasing of the pyrolysis temperature is due to the formation of the microporous structure [40]. In Table 5 it is shown the effect of pyrolysis temperature on physical properties (surface area and micropore volume) of biochar produced from walnut shells pyrolysis. Table 5. The temperature effect on the physical properties of the biochar (surface area and micropore volume) for walnut shell pyrolysis at a low heating rate (5 • C/min) and high heating rate (50 • C/min) and moisture content 7.85 wt.%, at 800 • C. Additionally, chemical properties of biochar change with biomass type and pyrolysis process conditions, for example, functional groups present in biochar can change by the pyrolysis temperature [40][41][42]. Usually, the H/C and O/C rates for biomass is about 1.2-1.8 and 0.55-0.75, respectively and this indicates that the biomass has low aromaticity content and has a high aliphatic content [43,44]. For walnut shells the H/C and O/C rates were 1.3105 and 0.6863, respectively (see Table 1). Pyrolysis leads to a significant decreasing in the H/C and O/C atomic ratios, the H/C and O/C rates for resulted biochar were in the range 0.098 to 0.2255 for the low heating rate and 0.0883-0.1463 for the high heating rate, respectively ( Table 4).

Characteristic
The changes were more important with the increase of pyrolysis temperature. Below 500 • C, the major reactions were mainly dehydration, decarboxylation and decarbonylation. However, at a temperature above 500 • C, dehydrogenation is the major reaction. These different reactions could affect the characteristics and functionality of the char. This carbon-based material with typical size features in the lower nanometer size range will be attractive for fundamental research and potential new applications, such as for environment applications (sorbent or catalytic material). The results also show the relationships between pyrolysis conditions and product characteristics and the research is intended to use the biomass residues to produce both hydrogen rich gas with the aim to become an important clean energy resource and biochar with a high porous structure.

Conclusions
Residual biomass, such as agricultural residue, represents a valuable renewable energy source and there are different ways by which biomass energy can be transformed into fuels. In this study were used walnut shells as a raw biomass source. Pyrolysis experiments of wet and dry walnut shells were performed and the influence of humidity amount on the hydrogen yield in the gas phase was estimated.
The tests were carried out to analyze the influence of the operating parameters (temperature, humidity content, flow rate of the stripping gas and heating rate) on the pyrolysis process and hydrogen yield in gas fraction. The results lead to some conclusions: (i) The pyrolysis of wet walnut shells without the step of predried favored the production of hydrogen-rich gas. Under fast pyrolysis and without the usingofN 2 stripping gas, the pyrolysis of the WS W sample releases higher H 2 yield than those obtained from the WS TD sample;(ii) In order for the water vapors releasing from its own humidity amount of the walnut shells to be used to obtain more hydrogen, the fast pyrolysis process was required; (iii) The stripping gas flow rate had influence on hydrogen yield, with the decreasing of the stripping gas flow rate the hydrogen yield increased and (iv) This study demonstrated that a new steam supply way is possible by combination of the predrying step with the water vapors generating from wet biomass and thus making this method much more efficient and economical.
At a pyrolysis temperature below 400 • C, the biochar surface area did not change too much, due to an incomplete removal of volatile compounds from biomass, while at a pyrolysis temperature in a range of 400-500 • C, the char surface area began to increase and the increasing continued up to a pyrolysis temperature at least of 800 • C. The increasing of biochar surface area with pyrolysis temperature was due to the development of microporous structure. Pyrolysis led to a significant decreasing in the H/C and O/C atomic ratios, the H/C and O/C rates were in the range 0.098-0.2255 for the biochar resulted by pyrolysis with a low heating rate and 0.0883-0.1463 for biochar resulted by pyrolysis with a high heating rate, respectively.