Influence of Bed Movement and Amount of Supplied Air on Updraft Gasification of Hardwood Pellet

This work presents the results of hardwood pellet gasifi cation with different amounts of air as a gasifi cation medium. The effects of bed movement and the equivalence ratio (ER) on the temperature profi le, gas composition, carbon conversion effi ciency and the energy balance were taken into account. Slow movement of the bed promotes high combustion and reduction zones, while fast bed movement leads to high pyrolysis zones and higher caloric values of syngas. When the amount of air increased from 12 to 23 Nm3/h, the gas yield increased from 1.4 to 1.6 Nm/kgbiomass for slow bed movement, and from 1.0 to 1.3 Nm/kgbiomass for fast bed movement. These results show that in both Cases similar specifi c energy values were obtained. However, in Case 1 lower fuel consumption was reached. Chemical energy in the syngas represents 80 % of the output energy for slow bed movement (265 MJ/h) and 75 % for fast bed movement (295 MJ/h). A signifi cant effect of bed movement in the reactor suggests that the gasifi er could be considered as a fl ow reactor, and additionally the fast movement of the bed with 20 Nm3/h of supplied air yielded the highest-quality gasifi cation process. Moreover, fast bed movement in the reactor leads to a high amount of generated char with high energy potential.

DRVNA INDUSTRIJA 69 (4) 339-347 (2018) temperatures in the gasifi er.These aspects were also presented by Ayyadurai et al. (2017), who studied large (1 m in length and 0.06 m in diameter) woody biomass gasifi cation in a 60 kW th updraft gasifi er.These authors found that the gasifi cation process with ER=0.6 producer syngas attained a heating value of 4.5 MJ/Nm 3 and that the gasifi cation temperature reached a temperature of 955 °C in the oxidation zone.
Another interesting issue related to updraft biomass gasifi cation has been presented by Huang et al. (2017).These authors studied the characteristics of residual carbon in biomass, including structure and gasifi cation activity.These aspects were investigated using Raman spectroscopy.Residual carbon in the ash is the result of char particles spending only a short amount of time in the gasifi er.
Based on current literature, there are no available works that provide detailed information about the infl uence of time that the fuel spends in the gasifi er or rate of bed movement and the amount of supplied air on the gasifi cation temperature and syngas characteristics.The primary objective of this work was to investigate an alternative way to stimulate the updraft gasifi cation process by forcing different rates of bed movement at different amounts of supplied air.The main parameters presented in this work include the temperature profi le along the gasifi er (e.g., the characteristics of the temperature zones, gas composition and caloric value, carbon conversion rate and energy balance).

Proximate and ultimate analysis of fuel 2.1. Neposredna i krajnja analiza goriva
The hardwood pellets particles are a cylinder with diameter of 8 mm and length of up to 15 mm.The bulk density of hardwood pellets during experiments was 646 kg/m 3 .To conduct the proximate and ultimate analyses, the hardwood samples were fi rst dried using a moisture analyzer (RADWAG) to determine their moisture content.The analysis of elementary composition was carried out using a CHNS-O Flash 2000 analyzer.The determination and calculation of the calorifi c value was executed using a KL-11 calorimeter.The results of the proximate and ultimate analyses are presented in Table 1.

UVOD
The gasifi cation process, which is recognized as one of the most effective in terms of thermal utilization of biomass and municipal waste, is a very popular topic in global research (Arena, 2012).This process is typically classifi ed based on the type of gasifi cation system.A common construction is a fi xed-bed gasifi er working in the downdraft or updraft confi guration (McKendry, 2002).Downdraft gasifi cation systems generate gas with a lower tar content than updraft gasifi ers.Therefore, many studies focused on different types of downdraft gasifi ers can be found in the literature (Balu and Chung, 2012;Phuphuakrat et al., 2010;Nisamaneenate et al., 2012;Zainal et al., 2002;Erlich et al., 2011).In the case of updraft reactors, the high tar content in the produced gas has signifi cantly slowed the development of this technology (Dudyński et al., 2012).However, studies based on updraft gasifi cation are described in the literature.Blasi et al. (1999) used beechwood biomass in an updraft fi xed-bed gasifi cation process, which resulted in gas with a content of 28 % CO, 7 % CO 2 , 7 % H 2 and 2 % CH 4 .
Chen et al. (2011) studied the operating conditions of updraft gasifi cation of mesquite and juniper and found that the heating value of syngas increased from 3.5 to 3.9 MJ/Nm 3 for juniper and from 2.4 to 3.5 MJ/Nm 3 for mesquite when the equivalence ratio (ER) decreased from 0.37 to 0.22.For both types of fuel, the maximum temperature in the combustion zone was above 1000 °C.Pedroso et al. (2013) studied wood chip gasifi cation in bottom-feed updraft gasifi cation and showed that the gas produced contained 27 % CO and 4 % CH 4 and, relative to a typical updraft gasifi cation system, a lower concentration of H 2 (6 %).The temperature of the bed in the reactor decreased from 885 °C in the combustion zone to 100 °C in the drying zone.
Recent research has focused on methods to clean the producer gas, optimize the updraft gasifi cation process to decrease the high content of tar in the syngas, and obtain high-quality biochar (Taupe et al., 2016).Ismail and El-Sala (2017) carried out numerical simulations and experimental studies of the infl uence of temperature in the gasifi er and ER on syngas composition and tar yields during updraft gasifi cation of wood pellets.It was reported than the ER had a signifi cant impact on bed temperature and gas quality.A higher ER ratio corresponds to a larger amount of oxygen in the combustion zone, which leads to a lower concentration of carbon monoxide and hydrogen and also increases the carbon dioxide content of the syngas.On the other hand, more oxygen supplied to the gasifi er promotes oxidization reactions and leads to more heat generation and higher The initial level of the feedstock was maintained at the level of the air inlet nozzles until a high bed temperature was obtained.After a high bed temperature was obtained, the gasifi er was fi lled to the level of the indicator.In each experiment, the height of the fuel bed was kept at the same level using a rotary fuel level indicator (i.e.112 cm from the bottom).The amount of supplied air was set at 12, 17, 20 and 23 Nm 3 /h, and the duration of each experiment was 120 minutes from the point when the gasifi cation process reached a steadystate condition (i.e. a constant temperature in each zone and a steady amount of received char).The gasifi cation process was performed at atmospheric pressure in the range of 900-1100 °C.The solid residue (char and ash) was gathered from the bottom of the reactor using an inverter coupled with a cyclical screw at a frequency of 5 sec/5min (Case 1) and 10 sec/2.5 min (Case 2), which stimulated slow and fast bed movement in the reactor (char outlet velocity).

Measurements of gas composition 2.4. Mjerenje sastava plina
After steady-state conditions of gasifi cation were achieved, the samples of syngas were sampled using tedlar bags.The analysis of the syngas content was performed using a SRI Instruments 310 gas chromatograph with a ShinCarbon ST 80-100 packed column and a thermal conductivity detector.The gas analyzer was pre-calibrated using a standard mixture of gas (CO, CO 2 , CH 4 and H 2 ), and argon was used as a carrying medium.

Temperature profi le in the gasifi er 3.1. Temperaturni profi l u rasplinjaču
Experimental investigations of hardwood pellet updraft gasifi cation revealed a similar temperature profi le trend in the gasifi er, which is consistent with the literature (Chen et al., 2012;Joseph et al., 2016).The results indicated a signifi cant impact of bed movement velocity in the gasifi er on the height of individual pro-nozzles with a diameter of 8 mm) was installed 52 cm from the bottom of the reactor, ended with a funnel with 43 cm and it was connected with the char outlet tube with a diameter of 104 mm.The syngas outlet was installed 111 cm from the bottom.
During the experimental investigation, fuel from the hopper was loaded by a screw feeder, and tube with a diameter of 104 mm, to the reactor 10 cm below the syngas outlet.Six thermocouples (Type N and S) were installed in order to measure temperature within the gasifi er.These thermocouples were placed 30, 55, 72, 88, 90 and 119 cm from the bottom of the reactor.The air was supplied to the gasifi er via an electric blower and controlled using an inverter and a thermal mass fl ow meter.The syngas left the reactor and passed to the combustion chamber, through an outlet tube with a diameter of 54 mm.On the outlet tube, part of the produced gas was directed to the gas sampling system.

Experimental procedure 2.3. Postupak istraživanja
Prior to starting gasifi cation, a batch of 3 kg feedstock was loaded into the gasifi er in each experiment.tion) zone attained a maximum of 16 cm.Case 2 promoted the pyrolysis process with a zone height of 17 cm.Lower temperatures in pyrolysis zone in Case 2 are associated with the intensifi cation of supplied fresh fuel and consumed heat, supplied by oxidation of biomass at the lower part of the reactor, in endothermic pyrolysis reactions (high temperature gradient between 35 and 40 cm from the air nozzles, Figure 3).The results of the hardwood updraft gasifi cation also indicated that the gradient of the bed temperature was about 6 °C/cm for 12 Nm 3 /h of supplied air and 10 °C/cm for all other amounts of air in Case 1.In Case 2, cess zones (Figure 2 and Figure 3).Based on the literature, the boundary of the combustion and reduction zone was established at 1000 °C and 750 °C (Chen et al., 2012;Sircar et al., 2014;Mani et al., 2011).Our results show that, in Case 1 (slow bed movement in the reactor), the combustion zone attained a height of about 32 cm from the air nozzles and the reduction zone reached a maximum height of 7 cm.
Since the level of the fuel in the gasifi er was set at 50 cm from the air nozzles, the pyrolysis zone attained a height of 11 cm.In Case 2 (fast bed movement in the reactor), the height of the combustion (reduc- the corresponding values were 12 °C/cm and 13 °C/ cm.Faster bed movement in the reactor resulted in a slightly steeper temperature gradient.

Performance of the gasifi cation process 3.2. Obilježja procesa rasplinjavanja
Experimental investigation showed the importance of comparing the amount of air supplied, fuel consumption, equivalence ratio and temperature.The results presented in Table 2 show similar temperature and fuel consumption trends as a function of the amount of air.However, in Case 1 the same value of ER was rated (0.16) in each experiment.In Case 2 ER value increased from 0.09 to 0.13, which is related to similar fuel consumption with the simultaneous increase in the amount of supplied air.It is noteworthy that, in Case 2, increasing of the amount of air supplied to 23 Nm 3 /h led to a signifi cant increase in fuel consumption (from 28.8 to 33.7 kg/h), which may be caused by the increase of the intensity of oxidation/reduction processes.

Gas calorifi c value 3.3. Ogrjevna vrijednost plina
In an updraft gasifi er, carbon dioxide is generated via the oxidation of wood pellets in the lower part of the reactor, and carbon monoxide is produced by a char reduction reaction (Boudouard reaction) in the reduction zone between the bottom and the middle parts of the reactor.Pyrolytic gas is produced in the pyrolityc zone at the lower-middle part of the gasifi er (McKendry, 2002).
Diff erent rates of bed movement, apart from variable amounts of supplied air, also aff ected signifi cantly the generated zone heights and syngas composition.Fast bed movement in the gasifi er (Case 2) promoted pyrolysis process with the height of the zone (Figure 3), which led to a slightly reduced production of pyrolytic gas and higher calorifi c value of the syngas (Table 3).In Case 1, slow movement of fi xed bed resulted in the formation of a high zone of oxidation and reduction (Figure 2) that infl uenced the lower content of carbon dioxide, hydrogen and methane and increased the amount of generated syngas.
Based on the literature, the amount of produced gas was calculated using the nitrogen tracer method (Chen et al., 2012;Thanapal, 2010), which relies on knowing the amount of nitrogen in the supplied air.The results revealed that the slow bed movement in the gasifi er led to a larger gas yield per kg of biomass; the peak value was 1.6 Nm 3 /kg biomass for 23 Nm 3 /h of supplied air.At fast bed movement in the gasifi er, the corresponding value was 1.3 Nm 3 /kg biomass .Furthermore, an increase in the amount of air supplied to the gasifi er resulted in an increase in the amount of gas yield per kg of biomass for both cases, which is consistent with the literature (Chen et al., 2012;, Ismail and El-Sala, 2017).Specifi c gas energy (MJ/kg) is an additional parameter representing the amount of energy obtained from a kilogram per hour of solid fuel in the form of gaseous fuel (Taupe et al., 2016) and allows to determine the energy effi ciency of the gasifi cation process.For both Cases, experimental results showed that the specifi c gas energy reached similar value (Table 3).However, in Case 1 these values were reached at lower fuel consumption (Table 2).It is caused by lower heating values of produced gas (4.8 -6.2 MJ/Nm 3 ) with a  4).In Case 2 (Figure 5) maximum gas yield was reached with a fuel consumption of 28.8 kg/h.This aspect indicates the potentially optimal process conditions for the presented Cases (char outlet velocity).

Char characteristic 3.4. Svojstva pougljenjenog materijala
Experimental investigations of hardwood pellet gasifi cation revealed the impact of stimulated bed movement in the reactor on fuel consumption (Table 4).As the amount of supplied air increased from 12 to 23 Nm 3 /h, the fuel consumption increased from 14.6 kg/h to 27 kg/h in Case 1 and from 26.2 kg/h to 33.7 kg/h in Case 2. This relation was also presented by Pedroso et al. (2013), who showed that the increase in the amount of air from 21 to 28 Nm 3 /h caused an increase in fuel consumption from 10 to 14 kg/h during updraft gasifi cation of woodchips.Furthermore, a larger amount of oxygen supplied to the gasifi er intensifi ed the oxidization reaction, which led to an increase in the carbon conversion rate.In Case 1, an increase in the amount of supplied air from 12 to 23 Nm 3 /h resulted in increased carbon conversion from 70 to 94 %.In Case 2, the corresponding data were 78 and 83 %, respectively.A larger amount of oxygen supplied to the gasifi er promoted oxidization reactions and led to increased CO 2 generation.These aspects were presented by Kalström et al. (2015).These authors found that an increase in carbon dioxide concentration from 13 to 60 % signifi cantly affected the char conversion rate during CO 2 gasifi cation of char particles from torrefi ed fuels like pine shell, olive stones and straw.
The carbon conversion was calculated as the ratio of the carbon difference in biomass and char after gasifi cation compared with the amount of carbon in the biomass.The theoretical amount of char was based on the difference in the mass of carbon in the fuel and the resulting syngas.These calculations did not take into account the tars.The experimental results revealed that optimum conditions were achieved for fast movement of fi xed bed in the gasifi er (Case 2) for 20 Nm 3 /h, where the smallest difference between the received and calculated amount of produced char was obtained, which may indicate the high fuel conversion rate, lowest accumulation of carbon (char) in the reactor and low tar content.

Mass and energy balance 3.5. Masa i energetska bilanca
An important parameter for comparing the impact of the rate of bed movement in the gasifi er and the amount of supplied air on the updraft gasifi cation process is a mass and energy balance.The energy input   fl ow was defi ned as the product of fuel consumption per unit time (Table 4) and the higher heating value of hardwood pellets (Table 1): The energy output fl ow associated with chemical energy in the syngas was calculated from the volumetric composition of the combustible species contained in the syngas and their caloric value (Table 2): The syngas produced at the outlet of the reactor was very hot and also carried a certain amount of heat (sensible heat), which can be defi ned as a product of the specifi c heat capacity, the molar fl ow of each component and the outlet temperature of the syngas: The specifi c heat capacity c p,i at different temperatures is well known in the literature (Nederlandse Gasunie, 1988), and the molar fl ow of each component was calculated from the data given in Table 4.The syngas temperature at the outlet of the gasifi er was no more than 230 °C.
A certain amount of char was collected during gasifi cation of hardwood pellets in each experimental series (from 0.8 to 3.5 kg/h; Table 4).For this reason, the energy output contained in the char P out,char was also taken into account and was defi ned as the product of char received per unit time (Table 3) and its higher heating value: Our experimental results revealed that the energy input contained in the hardwood pellets reached, at 23 Nm 3 /h of supplied air, 530 MJ/h in Case 1 and 660 MJ/h in Case 2 (Table 4).For both cases, the largest fraction of the calculated output energy was chemical energy in the syngas (Figure 4 and Figure 5), from 94 MJ/h (at 12 Nm 3 /h of supplied air) to 530 MJ/h (at 23 Nm 3 /h of supplied air) in Case 1 (Figure 4) and from 513 to 660 MJ/h, respectively, in Case 2. The higher output energy in Case 2 is associated with the promotion of the pyrolysis process and the production of syngas with a higher calorifi c value.
The analysis of heat input and output during updraft gasifi cation of hardwood pellets revealed a deviation of 33-47 % in the heat balance (Table 5), which might be caused by our not taking into account the energy output of tars and the heat loss of the installation.
According to the fi ndings of Ayyadurai et al. (2017), it is diffi cult to heat balance the gasifi cation process.These authors showed that the deviation in the heat balance reached 22 %.They also found that the chemical energy contained in the syngas represented nearly 60 % of the output energy.In our study, the work chemical energy was balanced from 57-88 % in Case 1 and from 67-74 % in Case 2.
The char was gathered from the bottom of the reactor using an inverter coupled with a cyclical screw at a frequency of 5 sec/5min (Case 1) and 10 sec/2.5 min (Case 2), which stimulated bed movement in the reactor.The energy output contained in the char P out is also an important part of the output energy.As shown in Table 6, in Case 2, the thermal output reached 100 MJ/h with a fuel consumption of 28.8 kg/h (20 Nm3/h of air) and remained at a similar level (85 MJ/h) for the other cases.An increase in air supplied to the gasifi er for the slow movement of the bed (Case 1) caused a decrease in the quantity of received char (from 2.6 to 0.8 kg/h) and a decrease in the char output energy from 66 to 23 MJ/h.This aspect shows that fast bed movement in the reactor, stimulated by char outlet velocity, leads to a high amount of the char with a high energy potential, which can react with steam and form clean energy gas.
Our experimental investigation revealed that sensible heat contained in the producer gas did not constitute a signifi cant infraction of the output energy balance; the sensible heat reached 11 MJ/h in Case 1 and 10 MJ/h in Case 2 (Figure 4 and Figure 5).

ZAKLJUČAK
Updraft gasifi cation of hardwood pellets with different bed movement in the reactor and for 12, 17, 20 and 23 Nm 3 /h of supplied air has been developed and evaluated.The following conclusions were drawn: 1. Th e results indicated a signifi cant eff ect of bed movement in the reactor suggesting that the gasifi er could be considered as a fl ow reactor.2. Slow bed movement promotes a higher combustion and reduction zone, which leads to syngas maximum calorifi c value of about 6 MJ/Nm 3 , while fast movement of the bed promotes a higher pyrolysis zone and generates more calorifi c gas (6-7 MJ/Nm 3 ).3. Th e results indicate that slow bed movement, in comparison to fast bed movement, generated a higher syngas yield, with its lower caloric value.Th ese results showed that in both Cases similar specifi c energy values were obtained.However, lower fuel consumption was reached in Case 1. 4. Fast bed movement in the reactor leads to a high amount of generated char with high energy potential.5.An energy balance analysis was carried out for all considered cases.Th e total calculated input energy fl ow attained a maximum of 560 MJ/h for slow bed movement and 660 MJ/h for fast bed movement (at 23 Nm 3 /h of supplied air), and the output energy fl ow was 303 and 400 MJ/h, respectively.Th e devia-tion of 30-40 % in heat balance was caused by not taking into account the tars and heat losses.Furthermore, it was noted that the chemical energy in the syngas represented 60-80 % of the output energy.6.Both cases of updraft hardwood pellet gasifi cation generated high-quality gas.However, based on the temperature characteristics, performance data, carbon conversion effi ciency and energy balance analysis, it was found that the fast movement of the bed with 20 Nm 3 /h of supplied air yielded the highestquality gasifi cation process.

Table 4
Performance data of gasifi cation process