The feasibility and performance of using producer gas as a gasifying medium

The gasification process in a 20-kW downdraft gasifier is investigated using Computational Fluid Dynamics. The research aims to examine the feasibility of air and producer gas co-injectionas a gasifying medium on the performance of the whole process. To this end, the effect of varying the injected amount of the producer gas on producer gas composition, emissions, H 2 production, producer gas yield and higher heating value (HHV), the gasification, and carbon conversion efficiencies are reported. The results reveal promising findings for boosting the producer hydrogen amounts, and the corresponding heating values, with lower CO 2 levels. The ratio of the injected producer gas (within the gasifying medium) to the total required amount of air for gasification is defined by z (eqn. (10)), and it is found that optimum z values are around (0.25 – 0.5). For the same working conditions of moisture content (MC), feedstock, and equivalence ratio (ER), the gasifying medium of z = (0.25 – 0.5) has higher HHV, gasification, and carbon conversion efficiencies than air gasification up to 33%, 31%, and 19% respectively. Additionally, for the same optimum range of z, it is found that the producer gas yield, and H 2 production are higher than air gasification by 10%, and 34% respectively, with a reduction in CO 2 by 18%. Consequently, the research presents a significant potential for higher yield, rich H 2 , and CO 2 -free syngas production, while demonstrating a promising novelty and technique that is applicable to any sort of the gasification units ’ type and capacity.


Introduction
The CO 2 production from fossil fuels and the associated environmental impacts are affecting climate change ( [1,2]).Additionally, the increased energy demands and higher rates of fossil fuels depletion around the world are leading to more focus on renewables [3].Biomass, agricultural, bio-waste residues are all renewable sources of energy that could be converted into valuable products through gasification, pyrolysis, and combustion ( [4][5][6]).Additionally, renewable energy sources are considered as neutral-CO 2 sources because of the same amount stored inside is released during burning such materials.
Equilibrium and kinetic models are used widely in the gasification, combustion, and pyrolysis modelling of biomass.However, such models have limitations that restrict their ability to fully simulate the gasification.Zero dimensional kinetic and equilibrium models do not consider geometry changes effect on the process because it only depends on chemical reactions taking place inside the gasifier.Additionally, multiphase fluid dynamics, and gas-solid interactions inside the gasifier cannot be addressed using these models [21].A robust modelling tool that could meet all the previous mentioned restrictions can be assessed using CFD modelling which considers mass, and heat transfer, multi-phase interactions, gasifier geometry, and design limitations [19].The approaches in CFD models vary in design, feedstock characteristics, operating conditions, and gasifying mediums.As a result, CFD models are widely used in the gasification process influenced with wide range of chemical kinetics.Hence, it is expected to predict the gasification process effectively, and reduce the required time in designing a gasifier [22,23].
L.Yu et al. [24] introduced a CFD coal gasification model inside a fluidized bed gasifier.They combined Arrhenius rate reactions for the coal gasification with the kinetic theory of granular flow (KTGF).The results were validated against experimental data and found good agreement.After validation, the model was further used to examine the effect of changing gasifier height on velocity, temperature, and syngas composition.A detailed CFD model was built by Fletcher et al. [25] using CFX4 package for the gasification process in entrained flow gasifiers.The particles entering the gasifier are modelled using Lagrangian approach.They obtained gas species concentration by solving heterogenous reactions with transport equations.The results of producer gas composition and exit temperature meets a good agreement with the experimental data.
Kumar and Paul [9] presented a 2D model for a 20-kW downdraft biomass gasifier using ANSYS Fluent.The model includes the four main zones of a gasifier (i.e., drying, pyrolysis, combustion, and reduction).Euler-Lagrangian approach was applied through the DP model.The results found a good agreement with previous experimental data.Additionally, they used different feedstocks with different equivalence ratios (ER).The model showed reliable results, however, it was not performing well for lower ERs <0.35.Furthermore, the 2D model was later converted into 3D model [19] using rubber wood as a feedstock and found good agreement at the same working parameters of the previous 2D model, and experimental results.
Previous CFD research from literature discusses the effects of changing gasifier type, design, gasifying medium mixtures, amount, ER, and feedstock variations e.g., ([26-29]).In contrast, the current study goal is to examine the effect of using the producer gas with air as a gasifying medium towards the improvement of the whole process including maximizing H 2 and reducing CO 2 yields from the gasification process in a downdraft gasifier.Hence, the literature sheds light on previous and current gasifying mediums used in the gasification process and try to minimize the gaps in this track.The gasifying medium type and amount controls the composition of producer gas, quality, and the desired output power of a gasifier [30].To date, air, steam, oxygen, carbon dioxide, and mixtures of them are used in the gasification process.
Air is the most common gasifying agent because of its availability, cheapness, and readiness to be used all times without further separation (e.g., O 2 , CO 2 ) or prior arrangements ( [31,32]).However, the higher nitrogen content in air dilutes the producer syngas, and hence, lowers the energy content.On the other hand, the use of steam in the gasification process tends to produce higher hydrogen amounts in the producer gas and lower tar content, resulting in higher heating values.It results in producing up to 50 vol% of H 2 in the producer gas [33].Additionally, it is much cheaper than using O 2 , or CO 2 because of the higher separation costs [33].Nonetheless, it requires higher energy to initiate the gasification process because of the endothermic reactions associated with steam gasification e.g.(water-gas, and steam reactions-Table 2), [34].
J. Li et al. [39] recently introduced a CFD model to predict the steam gasification of biomass and waste materials.They introduced wide sensitivity analysis and concluded that lower temperatures (~600 • C), and lower steam/biomass ratio (SB) of 0.1 tend to produce the optimum values of syngas (higher heating values).On the other hand, higher temperatures and SB was found to weaken the (carbon-hydrogen-oxygen bonds) CHO effects on the producer gas and the corresponding heating values.More recent CFD models that work on steam gasification are introducing the same findings ( [40,41]).
Oxygen is a promising gasifying medium that produces higher quality syngas.However, the separation costs of oxygen from air are expensive.Also, oxygen mixtures produce higher heating values for the syngas with lower tar amounts [42,43].A 2D numerical CFD model was presented by Couto et al. [26] along with experimental set-up to examine the effect of using mixtures of oxygen and air on the biomass gasification process.They used Euler-Euler approach in the exchange of energy, mass, and momentum.Kinetic theory of granular flow (KTGF), discrete phase mode (DP), and k-ε model was selected for turbulence.The numerical results found a good agreement with experimental data.Additionally, they examined the effect of oxygen on steam/biomass ratio (SB), composition of producer gas, and temperature along the gasifier.They found that N 2 , and H 2 content decrease while CO 2 amounts increase as a function of O 2 content.Additionally, higher oxygen content boosts the cold gas efficiency.The CO 2 gasification process has gained wide attention recently because of carbon emissions reduction and the utilization of CO 2 e.g., ([44,45]).The producer CO 2 during the gasification process could be recycled back to the combustion zone as a gasifying medium.Several advantages for such technology are stated by Ref. [46], including the reduction of residual char, while increasing the reactive char which leads to higher efficiency of the gasification process, besides the reduction of CO 2 emissions.Additionally, CO 2 is a less corrosive agent compared to steam.
From modelling perspective, Wang et al. [44] designed an equilibrium mode to investigate the CO 2 gasification of biomass.They examined the gasification temperature, and CO 2 /C ratio, and found that the producer gas composition was enhanced in terms of heating value, by higher temperatures and CO 2 /C ratios.On the other hand, a detailed thermodynamic model using ASPEN plus was presented by Chaiwatanodom et al. [45].They examined the use of O 2 , steam, and CO 2 as gasifying mediums.To evaluate the process performance, they calculated energy demand, gasification efficiency, and the emissions of CO 2 .Experimental works for CO 2 gasification were also carried out for different materials (e.g., coal [47], biomass char [48], and wheat straw char [49]).The reaction rates of CO 2 gasification are reported as lower than steam gasification [50].Additionally, the use of CO 2 as a gasifying agent increased the H 2 production, syngas yield, and power generation ( [51,52]).
To the best of the authors' knowledge, and as per the presented literature review, previous studies do not adequately present the effects and significant use of producer gas as a gasifying agent.Additionally, the combined effect on gasifier performance, producer gas yield, heating value, and the gasification and carbon conversion efficiency are not yet explored.Furthermore, one of the major goals of the current research is to examine the possibility of boosting hydrogen yield while simultaneously decreasing the associated CO 2 emissions.As a result, the current research will try to address the aforementioned challenges for better gasification performance.

Numerical model description
A 20-kW downdraft gasifier is designed based on the kinetic model developed by the author [53].The integrated model is composed of three zones as illustrated by Fig. 1-drying and pyrolysis, combustion, and gasification.Each zone is controlled by a set of chemical reactions (Tables 1 and 2) where ANSYS fluent 19.0 [54] is used in the current simulation.The full description of the design is fully covered by ([8,8, 53]), and for brevity it is not repeated in the current research.
Biomass is fed from top of the gasifier through drying and pyrolysis zone (D = 21.8 cm), while producer gas is derived from two nozzles at the bottom of the gasifier (each D = 1.6 cm).The gasifying medium is injected through two nozzles within the combustion zone at the gasifier sides.Each nozzle (D = 1.6 cm) is specified at fixed height (7.8 cm) above the throat diameter (D = 6.2 cm) based on the previous recommendations described in Ref. [53].

Governing equations
Species transport model is initiated to track particles formation along the gasifier.Discrete ordinates (DO), and k-epsilon models are defined for radiation and turbulence respectively for their accuracy and higher predictions in gas-solid interactions [54].Biomass, and gasifying medium are fed at 300 K, and 600 K respectively.The biomass particles are modelled using the discrete phase model (DPM) following the

A.M. Salem and A.R. Abd Elbar
Lagrangian approach.The DPM considers the particles trajectories as a continuous phase of fluid where the interaction between the species/particles takes place considering the heat and mass transfer equations.The conservation equations of energy, mass, momentum, and species transport are solved numerically under the turbulent flow steady-state condition as specified in the model assumptions, with a set of finite rate kinetic reactions.These equations ( 1)-( 6) are presented as following [54,55]: Energy conservation: ∇.
where the parameters 92, and Y m = 0.09.h j is the enthalpy of species (j), S m is the mass added to the phase (kg), λ eff is the effective conductivity, τ the stress tensor (pa), and ε is the turbulent dissipation rate (m 2 /s 3 ).
The species transport equation [56]: where i refers to different species in the simulation, R i is the net rate of the production of different species (i) by the chemical reactions, Sc t is the turbulent Schmidt number and is represented by the ratio of turbulent viscosity to eddy diffusivity.The current research describes the multi-phase modelling with scaling.This scaling includes many parameters and their effect on gasifier hydrodynamics, behaviour, and products distribution.The accurate modelling approach is of a big importance in such cases.The current research will consider the proper scaling with the species transport and chemical kinetics inside the gasifier.This is also specified with the two-equations (kϵ RNG) modelling for specifying the turbulence.Additionally, the multiple non-dimensional parameters have to be considered during the scaling by the transport phenomena e.g., Re, Nu, Pr, …etc.Such parameters scaling is widely discussed in the ANSYS theory guide [54].

Drying and pyrolysis
The gasification process is composed of four main zones/steps.Drying, followed by the volatiles release/pyrolysis, combustion, and reduction/gasification.The heat released during the combustion process drives the biomass drying and decomposition in the pyrolysis zone.The Lee model [57] is the default drying model within the ANSYS directory [54] which accurately predicts the drying and moisture evaporation for mixtures.It showed a good stability for Euler-Lagrangian, and the VOF multi-phase models.Therefore, it will be used in the current simulation.After drying, in the pyrolysis zone, the biomass first decomposes into volatiles and char, followed by further interactions between the species during the pyrolysis zone as illustrated by Eqn (7, and 8) in ( [58,59]).
The volatiles are composed of gases (CO, CO 2 , CH 4 , and H 2 ) and other higher HC components (i.e., tar).Tar conversion kinetics were not included in the model as illustrated by the model assumptions, because the main objective was to investigate the effect of new gasifying agent on the gasifier performance.The producer tar during the gasification process includes hundreds of species with complicated reaction kinetics, and as a result, this is not discussed in the current research.However, it is planned for our future works.
The devolatilization and pyrolysis process takes place after the drying process.Depending on its composition and boundary conditions, biomass is decomposed into char, volatiles, and ash.The model carries out an elemental mass balance for the volatiles to estimate its products.However, the CO concentrations are first calculated using the model proposed by Ref. [60] which calculates the mass fraction of every species based on the pyrolysis temperature.Equation ( 8) describes the process of volatiles break-up based on [60].Then it is further implemented within the ANSYS directory to describe the species release during the biomass devolatilization (equation 7, and 8) based on the feedstock's composition.

Devolatilization and char burnout
The pyrolysis process starts when the combustingbiomassparticle temperature exceeds the vaporization temperature.Particle injection is governed by the discrete phase model which continuously injects uniform particle sizes following the Euler-Lagrangian approach.The current research simulations will consider using the single kinetic rate model for the particle's devolatilization.The model assumes first-order devolatilization depending on the remaining amount of volatiles in the particle [61], eqn.(9).
where m p is the particle mass in kg, f v,0 , and f w,0 are the mass fraction of volatiles and evaporating particle at initial conditions respectively, k is the kinetic rate in s − 1 , and m p,0 is the initial mass of the particle in kg.
The value of f v,0 is defined manually inside the ANSYS Fluent directory from the biomass ultimate analysis (Table 3), while the kinetic rate is calculated following the Arrhenius activation energy and preexponential factor.Devolatilization temperature is slightly varying only between discrete time integration steps.
The char burnout is governed by the combusting of volatiles with the char and oxygen as described through the DPM model by setting the injection properties.Additionally, char reactions are set up using the volumetric and multiple surface reactions model including all kinetic rate reactions in Tables 1 and 2.

Char and gas phase reactions
The different reactions included in the model for combustion and gasification reactions are illustrated by Tables 1 and 2 The detailed kinetic rate data well represent the reactions taking place in the oxidation and reduction zones.All the reactions are implemented inside the ANSYS Fluent code, including the volatiles decomposition (eqn.( 8)) reactions illustrated earlier.

Boundary conditions
Wood pellets and rubber wood feedstocks are used in model validation under different working conditions.Both feedstocks' proximate and ultimate analysis are illustrated in Table 3 where VM is denoted for volatile matter content, FC is the fixed carbon amount, and MC is the moisture content.

Assumptions and model convergence
The model is considering the following assumptions: • Steady-state simulations.
• Tar and other higher hydrocarbons are neglected in the current model.• Uniform spherical particle size.
• Two equations k-epsilon model is specified for turbulence.
• Turbulence intensity and hydraulic diameter where specified for all inlets/exits for uniform distribution of flow inside the gasifier.• All reactions take place under atmospheric pressure.
In addition, the set of models and solution methods, residuals control, and the boundaries used are all concluded in Table 4.
The pressure-velocity coupling algorithm is selected for the current simulations which effectively solves the combined pressure-based, and momentum equations [54].Additionally, the two-phase equations ( 1)-( 8) are numerically solved using the implicit finite volume method under the Fluent environment.The pressure spatial discretization is solved by PREssure STaggering Option (PRESTO) which accurately gives better conversion for the VOF, and multi-phase models.Upwind scheme is used for solving the gas species discretization, energy, and momentum.

Results and discussions
The model is evaluated for its stability through the mesh independency test carried out at different grid sizes.In addition, two different feedstocks are used for model validation at different working conditions.Air and producer gas will be further used as gasifying mediums, while this effect will be elaborated towards increasing the gasifier performance through studying gas composition, temperature, velocity, and species distribution, gasification, and carbon conversion efficiencies, CO 2 , and H 2 production, and producer gas yield.

Evaluation of mesh quality
The air gasification of rubber wood at ER = 0.326 and MC of 18.5% is selected for the mesh independency test with 5 different cell numbers (22526, 45402, 58968, 89956, and 144334).Additionally, the experimental results from Ref. [59] are compared with other values to study the current model validity.The producer gas composition and its corresponding heating value are represented in Fig. 2.
The results of the producer gas (mol %) and heating value in (MJ/ Nm 3 ) show slight variations with the different grid sizes selected in the comparison.Additionally, the results show good agreement with the experimental data [59], which demonstrates the consistency of results throughout the selected mesh sizes.For example, the variations in HHV are found to be <1.5% between all grid sizes.On the other hand, starting from 58968 cells, the results show almost no variation across all gas species and the associated heating value, implying the stability of the predicted results.Consequently, the mesh size of 58968 is selected for the current research simulations since the higher grid sizes requires higher computational cost.

Model validation
The model is validated against two different conditions including the feedstocks stated in Table 3.Air is used as a gasifying medium while different feedstocks, ER, and MC are considered depending on the experimental set-up from the literature.The kinetic model [53] is using the same gasifier thermal capacity (20 kW), however, the experiment [62] is using different scale of the downdraft gasifier.Different gasifier scales/sizes are not affecting the derived producer gas composition and the corresponding heating value.This is because the producer gas is depending only on feedstock, moisture content, and the used air/fuel ratio.Since these factors remain constant in the current model validation, it is expected to have the same producer gas composition (Fig. 3).
Rubber wood (Fig. 3a) is used at ER of 0.3 and MC 18.5, while wood chips (Fig. 3b) are used at ER 0.35, and 10% of MC.The volumetric gas composition, and heating value for both feedstocks are compared with the kinetic model data of [53] for rubber wood, and experimental data from literature for wood chips [62].The HHV variations for rubber wood, and wood chips are 3.3%, and 2.47% respectively, while other species shows slight variations for gas composition.For example, CO and H 2 variations are <3%, and CO 2 and N 2 variations are <10%.As a result, the model demonstrates its ability to replicate the gasification process where the results meet a good agreement between the current model, kinetic [53], and experimental results of [62].Temperature, velocity, and gas species distribution.
Temperature, superficial velocity, and gas species distribution along the gasifier for air gasification will be presented and compared with previous experimental/CFD data to further assesses the model capability to simulate the gasification process.
Fig. 4 depicts the superficial velocity and temperature contours for rubber wood air gasification at ER = 0.3.Maximum achieved temperature around the nozzles is approaching 1700 K, however, it does not represent the actual combustion temperature at the centreline of the gasifier (~1300 K) which is in a fair agreement with previous experimental results for similar working conditions e.g., ( [59,63]).For example, the experimental set up of Barrio and Fossum [63] for a downdraft biomass gasifier using ER of 0.29 showed that the combustion and gasification temperatures are around 1323 K, and 1000 K respectively.The current work average temperature at the centreline showed the combustion and gasification temperatures around 1300 K, and 900 K respectively which finds a fair agreement with the experimental data.However, slight variation are found because of gasifier scale and feedstock conditions.Higher residence time inside the combustion zone due to throttling leads to higher temperatures in this area which drives the whole gasification process.The heat generated inside the combustion zone because of the combustion heterogenous/exothermic reactions (Table 1) provides the required energy for drying, and pyrolysis, as long as the required temperature for gasification reactions (endothermic, Table 2) to take place.The highest velocity is achieved around the air nozzles and at the syngas exits (~0.4 m/s).this is because of smaller cross-sectional areas which under the conservation of mass provides higher velocities.On the other hand, at the combustion zone, throttling generates higher velocities, and hence, higher turbulence.The velocity values at the gasifier exit were previously reported by Ref. [64] during their experimental results by 0.3-0.55m/s.Additionally, the current model results finds a good agreement with previous numerical and experimental works of ( [22,65]) for the same working parameters.
Fig. 5 illustrates the mole fraction of volatiles, and different gas species formed during the air gasification of rubber wood at ER = 0.3.Biomass is fed from top of the gasifier at room temperature (300 K), and while the temperature inside the gasifier rises, the drying process starts followed by the volatiles release.The mole fraction of volatiles starts with 1 at top of the gasifier and decreases within the pyrolysis zone until all the gases and char are released before further react with the oxidant within the combustion zone.Nitrogen concentrations starts with maximum value around the nozzles (combustion zone), then it drops because of the dilution that takes place since other species are reacting forming syngas.The model does not consider any NO x formation or other reactions for N 2 compounds.
CO, H 2 , and CH 4 are following the same trend.During pyrolysis, the devolatilization process takes place forming higher concentrations of CO, H 2 , and CH 4 as depicted by the figure.Later, all volatiles are reacting with the gasifying medium in the oxidation zone leading to the reduction of their concentration during the partial oxidation under the controlled ER, and subsequently convert into H 2 O, and CO 2 .This is clearly seen in the CO 2 concentrations which increase during the combustion zone.However, the CO 2 , and H 2 O further react again with the volatiles, and combustion products within the reduction zone under the gasification reactions forming the producer gas.Highest concentrations are for CO, and H 2 than other species.The results meet strong agreement with the CFD results of [9].

Producer gas injection effect
The producer gas of rubber wood gasification with ER = 0.3 (Fig. 3a) is used in the following simulations to be co-injected with air at different ratios (z).The ratio z represents the ratio of producer gas within the gasifying medium.Air will be compared with the producer gas injection results for z (0-1), where z = 0 represents only air is used as gasifying medium, and z = 1 is when equal amounts of air and producer gas are coinjected.The current research is examining the effect of this injection (z ratios) on the performance of the gasification process.z = producer gas mass in gasifying medium total mass of gasifying medium (air) (10) Fig. 6a illustrates the velocity contours distribution along the gasifier with different z ratios.The z ratio clarifies the amounts of air and producer gas injected within the combustion zone at the fixed ER of 0.3.Higher z values show higher velocities within the combustion zone, throat, and around the gasifier exit nozzles.This is because of more throttling in such areas which in turn increase the velocity.
On the other hand, temperature along the gasifier centreline is presented in Fig. 6b the highest temperatures are achieved within the combustion zone, showing maximum values around 37 cm higher than the gasifier exit, which is below the nozzle's injection line.Additionally, maximum temperature was found to decrease with higher z ratios  because of lower nitrogen, and oxygen amounts.However, it was noted that higher z ratios (e.g., 0.75, and 1) have a lower gasification temperature compared to other ratios.For example, the achieved temperatures which in some simulations are ~1050 K are not favourable for gasification of wood biomass because of higher tar concentrations (e.g., Refs.[66][67][68]) where they report that T > 1400 K are favourable for tar destruction.As a result, the recommended values of z will be in the range of (0.25-0.5) for higher production of syngas.
The volumetric gas composition of different gas species and the corresponding heating value are presented in Fig. 7.The results depict the boosting of CO, H 2 , and CH 4 mole fractions with increasing z values.Additionally, increasing z values leads to a significant increase in the heating value because of the higher amounts of CH 4 , CO, and H 2 .This is due to the increased injection of CO, CH 4 , H 2 , and CO 2 from producer gas inside the combustion zone.Hence, this leads to higher reaction rates for the combustion reactions (Table 1) -althoughsuch partialcombustion is governed by the ER.Thus, resulting in increasing the production of CO, CH 4 , and H 2 .Additionally, it was reported by Ref. [46] that the recycled CO 2 increases the amounts of reactive char, which in turn increases the amounts of CO, and H 2 through the gasification reactions including water-gas, and boudouard reactions (Table 2).As a result, CO 2 amounts are found to decrease slightly while higher z values are injected during the combustion process.Furthermore, higher z ratios mean lower amounts of N 2 in the gasifying medium because it will require higher producer gas in the gasifying medium and lower amounts of air (e.g., see eqn.(10)).This subsequently decreases the dilution of producer gas, and in turn, higher syngas concentration.
While changing z from 0 to 1, it is found that CO, CH 4 , H 2 mole fractions increase by 48.6%, 110%, and 60% respectively.On the other hand, CO 2 emissions drops by 37% mole % under the same conditions.Additionally, the corresponding heating value increases by 63% than air gasification.However, due to the previous reported factors of lower temperatures (Fig. 6), temperature values for z > 0.75 are found to decrease ~ (1050 K i.e., 777 • C) which are not favourable for the gasification process.Such low temperature leads to higher tar formation, and lower reaction rates, resulting in lower syngas composition and lower heating values.Additionally, higher z value means more consumption of the recycled gas which has side economic effects.As a result, and based on the current research findings, the recommended/optimum z value will be 0.5 in terms of producer gas composition and lower emissions.Under z = 0.5, the values of CO, CH 4 , H 2 mole fractions increase by 26%, 50%, and 34% than air gasification respectively.On the other hand, CO 2 emissions drops by 18%, while the heating value increase by 33% than air gasification.

Producer gas yield, carbon conversion, and gasification efficiency
The producer gas composition results showed a good performance of the producer gas injection with air, increasing the heating value of producer gas, and lowering CO 2 emissions.However, it needs more justification to evaluate the performance of the gasification process under these circumstances.As a result, the gasification efficiency, carbon conversion efficiency, and the yield of producer gas will be presented in this section.Because carbon is the dominant element of product gas and feedstock in gasification, it represents the carbon conversion efficiency ɳ cc .Carbon conversion efficiency is defined as the proportion of output/converted carbon into gases to the total amount of carbon in the biomass [69,70].
The gasification efficiency is calculated as follows [71].
where Q g (MJ/Nm 3 ) is the producer gas LHV, Q s (MJ/kg) is the LHV of the injected syngas with air, G p is the producer gas yield in Nm 3 /kg, and Q b is the biomass LHV in MJ/kg and both are estimated as following [72].The carbon conversion efficiency is calculated from Ref. [70].
The effect of changing gasifying medium composition on producer gas yield is presented in Fig. 8. Rubber wood is used as feedstock with ER 0.3.Air shows lower yield of producer gas compared to higher z values.However, this finds a good agreement with previous works of ( [72,73], and [74]).As the value of z increase, higher velocities are achieved (Fig. 6).As a result, this effect leads to higher velocities, and volume flowrates near the gasifier exit, and correspondingly, higher yield of the producer gas.Moreover, higher z values produce higher syngas composition (rich in CO, and H 2 ), and higher heating values (Fig. 7), which in turn, boosts the gasification efficiency.Changing z (0-1) results in an increase for the producer gas yield by 16%.On the other hand, for the suggested optimum z values (0.25-0.5), the increase of yield than air gasification is found to be (5-10) %.Additionally, Fig. 9 depicts the gasification, and carbon conversion efficiencies of rubber wood gasification under different ratios of the injected producer gas (z).Air gasification is presented in all figures for easier comparisons.Additionally, the air gasification results are verified against experimental data from literature to validate the current model results.Air yields moderate values of carbon conversion, and gasification efficiencies which are in a fair agreement with [72][73][74] for the gasification efficiency, and ( [69,70,75]) for the carbon conversion efficiency.Since there were no previous experimental or numerical data regarding syngas yield, gasification and carbon conversion efficiencies for the current set up, the comparison with previous data (e.g., Refs.[69,72]) were only against air gasification considering using same boundary conditions and biomass.For example, Gai and Dong [72] reported that in a downdraft gasifier, air at ER from 0.18 to 0.41 would give a gasification efficiency and gas yield from 21.17 to 73.61% and 1.35-to 2.86 Nm 3 /kg respectively, which finds a good agreement with the current data.Same gasifier type was used (downdraft), however, not the same  gasifier size.It is worth mentioning that the gasification efficiency is not depending on the gasifier size.This is because the syngas LHV, syngas composition, and biomass heating value are only depending on biomass type, gasifying medium, and ER.As a result, both gasification and carbon conversion efficiencies have no dependency on the gasifier scale.
Carbon conversion efficiency is depending on the producer gas yield, and the resulted converted carbon.As a result, increasing the z values would achieve higher carbon conversion.This is because of higher yield of the producer gas while z value increase.Additionally, higher z values are found to consume more char and increase the gasification reactions leading to higher CH 4 and CO as previously described by Fig. 7. Consequently, air gasification depicts lower values compared to higher z values.For the same working conditions, increasing z from 0 to 1 produces higher carbon conversion efficiency by 35%.While for the optimum working parameters z= (0.25-0.5), it tends to producer higher carbon conversion efficiency than air by (10)(11)(12)(13)(14)(15)(16)(17)(18)(19) %.On the other hand, the gasification efficiency found to decrease while increasing z values between (0.1-0.25), then it tends to show a gradual increase for higher z values.The gasification efficiency is highly affected by the producer gas yield and LHV, biomass LHV, and the injected stream of producer gas with the gasifying medium.While higher z values show higher yield of the producer gas, it also requires higher addition of the producer gas with air (i.e., higher energy requirements), resulting in lower efficiencies.However, this principle is applicable to a certain amount of z.Higher values of z are producing higher yield which has a higher influence on the gasification efficiency than the energy introduced by the gasifying medium.Thus, higher gasification efficiencies are achievable.For the same working conditions, higher z values could produce 31% higher gasification efficiency than air gasification.Fig. 10 illustrates the volumetric composition of the producer gas at different ERs for rubber wood gasification.Fixed amount of producer gas is injected with air based on the optimum values discussed earlier, so the current results are based on z = 0.5.Nitrogen yield is lower than air gasification for all simulations since the injected amount with the gasifying medium is reduced while ER increases.Higher ER promotes higher oxygen amounts in the oxidation zone, higher turbulence, and hence higher combustion reactions rates, and correspondingly higher amounts of CO 2 and more consumption of H 2 , CO, and CH 4 .As a result, higher ER produces lower heating values for the producer gas.While reducing ER from 0.4 to 0.2, the values of CO, H 2 , CH 4 , and HHV increase by 15.8%, 21%, 26%, and 21% respectively.On the other hand, CO 2 emissions are found to reduce by 21% for the same range of ER.
For the same working conditions (i.e., downdraft gasifier, ER 0.3, and rubber wood), the current results are compared with other gasifying mediums in terms of produced gas heating value, yield, and gasification efficiency.Oxygen gasification [22], oxygen-steam gasification [76], and steam gasification [77] produces HHV of 12, 11.4, and 6.5 MJ/Nm 3 respectively vs 8 MJ/Nm 3 for the current work.This gives an advantage of the current work which produces 23% higher heating values than steam gasification, however, lower values than oxygen gasification by ~30%.On the other hand, oxy-gasification requires additional cost of oxygen separation, supply, and safety, which makes the current suggested gasifying mediumproducer gasa promising alternative for higher and rich value products from the gasification process.
Furthermore, gasification efficiencies for oxygen gasification [22], and the current work are found to be 52, and 51 respectively, while the carbon conversion efficiencies for the same mediums were found to be 62 and 78% respectively.In addition, for the same conditions, oxygen gasification and the current research yields 1, and 2.04 Nm 3 /kg of producer gas.The current research demonstrates higher carbon conversion, and gas yield than oxy-gasification, resulting in promising findings that could be transferred in the future as a strong alternative to the traditional gasifying mediums.

Carbon intensity
Compared to air gasification, the effect of co-injection of the producer gas with the gasifying medium proves its potential to decrease CO 2 emissions.For example, compared to the optimum conditions of z = 0.5, the air gasification produces higher CO 2 emissions by ~18% (Fig. 5).This has been clarified by more retractive char produced, more combustible species, and higher steam amounts, resulting in CO 2 consumption.Additionally, the carbon conversion of the process enhanced with higher z amounts (Fig. 9).This is also explained in terms of higher yield of the producer gas, and CO 2 consumption, which results in higher carbon conversion efficiencies.
The derived results from (Figs. [8][9][10] proves the model applicability on any range of gasifiers.Although the current research is based on a small-scale industrial gasifier (20 kW), but it shows promising findings on other scales.The results of dry gas composition are derived based on specific working parameters regardless of the gasifier size (e.g., ER, MC, and biomass type).Additionally, the gasifier performance (yield, efficiencies) is independent of the gasifier scale since the efficiencies are dimensionless %, and the yield is presented in (Nm 3 for every kg of biomass).As a result, the represented findings of the current study could be applicable to different scales and multi-applications.As a result, the key findings from the current research of boosting H 2 production, increasing producer gas yield, and HHV, with reducing CO 2 emissions are all giving promising outcomes without the need for any capture technologies, using solvents, catalysts, or complicated design processes.

Conclusion
A CFD model is developed for a 20-kW downdraft gasifier to investigate the effect of using mixtures of air and producer gas as gasifying mediums on the process performance.It was found that increasing the amount of producer gas with air (i.e., higher z ratios) favors the production of CO, CH 4 , H 2 , resulting in higher HHV for the producer gas.Additionally, it also results in boosting the gasifier performance in terms of higher producer gas yield, gasification, and conversion efficiencies than air gasification under the same working conditions.Furthermore, it was found to decrease the CO 2 emissions than air gasification.However, the continuous feed of producer gas with air leads to a drop in the gasification temperature, which is not recommended for tar removal.As a result, the recommended vales for z are advised to be (0.25-0.5) to achieve better performance for the gasification process.
The reported results give a promising outcome in reducing CO 2 emissions without the need of a catalysts, or other removal techniques.Additionally, the results and findings are applicable to any sort of gasification system with different capacities, giving promising outcomes for economic applications.

Fig. 2 .
Fig. 2. Producer gas composition at different cell numbers for rubber wood gasification.

Fig. 4 .
Fig. 4. Temperature and superficial velocity contours along the gasifier for air gasification.

Fig. 8 .
Fig. 8. Producer gas yield for different conditions of air and syngas/air ratios.

Table 1
Oxidation reactions in combustion zone.

Table 2
Reduction zone reactions.

Table 3
Feedstocks data used in validation and testing the model.

Table 4
Solution methods and boundary conditions followed in the CFD modelling.