Biochar from Delonix regia pod: consideration of an updraft retort carbonisation process

Abstract Retort carbonisation is a novel technology especially suited to the sub-Saharan Africa energy conversion challenges. This study aims to produce and characterise biochar from Delonix Regia pod (DRP) via the retort carbonisation process. The process achieved a Delonix Regia pod biochar (DRPB) yield of 29.48 wt% at a peak temperature of 375 °C and a process time of 150 min. The average pore volume, specific surface area, and average pore size of DRPB were 0.0352 cm3/g, 88.03 m2/g, and 1.6 nm, respectively. Morphological analysis revealed that DRPB had a heterogeneous surface morphology with an average roughness of 12.96 × 103 µm. Functional groups such as C-O, N-O, O-H, C = O, CO-O, and C-H are present in the biochar. DRPB compares well with other retort carbonisation biochar. Potential applications were also discussed based on the biochar properties and the product can be tested for water treatment applications and as an additive for improving the tribological and rheological properties of lubricating oils.


Introduction
Biomass is a renewable agricultural waste that is low-cost, readily available, and possesses a reduced production time [1].Biomass wastes gotten from agricultural byproducts are one of the most predominant universal environmental problems [2][3][4][5].The sustainable transformation of these wastes into better quality useful products is essential.One such transformation is based on using waste as a precursor for producing biochar and energy pellets [6].Therefore, there is a need to devise a means of producing biochar from cheaper and readily available materials [7,8].
Delonix regia is popularly known as the 'flame of forest' or royal poinciana.It belongs to the family Fabaceae [9] and is well known for its flamboyant display of flowers [10] and fern-like leaves [1].It is distributed throughout Madagascar, Africa, India, and Northern Australia [11].Trees of Delonix regia are grown essentially for an ornamental purpose [12].The pods are brownish when ripe and green when young [13].The length can be about 60 cm and approximately 5 cm in width with very small individual seeds [14].Different parts of the plant are customarily used for medicinal purposes [15].Delonix regia pod (DRP) is in general considered agricultural waste but the leaves and pods could serve as a good precursor for the production of biochar [13].Exploring the use of biomass material to produce efficient and low-cost biochar may play a part in environmental sustainability and proffer benefits for future industrial applications [16].The pods steadily fall off the parent trees when matured.Due to their free accessibility, the pods are a very useful precursor for the production of low-cost biochar [1].
Biochar is a microcrystalline, non-hazardous solid, nongraphitic form of black carbonaceous material with a closely packed permeable structure obtained from thermochemical processes [17][18][19][20][21].In recent times, researchers discovered that biochar gotten from high carbon agricultural residues had good adsorbent characteristics [22] and were suitable for the adsorption of hazardous gases, air pollution control, solvent recovery, food and chemical industries, wastewater treatment and recovery of gold and silver in hydrometallurgy [22,23].Some biochar is usually manufactured or produced at an increased temperature that ranges between 600 and 900 � C with inert media (nitrogen) [24].Biochar produced under these preconditions intensifies the cost of production and the energy needed.Therefore, it is necessary to reduce the cost of production and energy requirement by producing the biochar at low to medium temperatures [25].Retort carbonisation is one such biomass production technology without an electricity requirement [26,27].
The use of the thermochemical conversion process option for municipal solid waste management is gaining solid research interest due to its adaptability to various biomass resources [28,29].Carbonization is a process whereby the amount of elemental carbon in biomass resources is increased by heating the organic matter in an oxygen-lean or oxygen-free environment [30].It has varieties based on the nature of the process.When there is no oxygen in the reacting system, it is described as pyrolysis [31].When a gasifying agent (such as air, steam, or water) is allowed to flow through the feed during the process (to produce syngas) it is termed gasification [5].When some water/steam is introduced into the feedstock (not as a flowing stream) it is termed hydrothermal carbonisation/liquefaction [3,32].When the system is designed such as the flue gas generated in the retort process is re-combusted in the heating zone to generate more thermal energy for the process, it is termed retort carbonisation [33].If the heating temperature is quite low and biomass conversion is incomplete, it is termed torrefaction [34].Pyrolysis and hydrothermal carbonisation/liquefaction tend to produce more solid and liquid phase products while gasification tends to produce more gas phase products and less of the others.Processes like retort carbonisation, torrefaction, and gasification do not/rarely produce any liquid products.Retort carbonisation is a thermochemical process focused on biochar production.Besides the heating by controlled biomass combustion, the combustible gas released from the carbonisation of the biomass is burnt as a secondary heat source which constitutes a retort [35,36].One of the earliest conceptualisation of the process was reported by Adam [37].Chandrasekaran et al. [38] investigated the retort carbonisation of Prosposis juliflora.They observed a 29.6 wt% biochar yield at 500 � C. Ighalo et al. [39] observed a 28.57wt% biochar yield at 494 � C for dry almond (Terminalia catappa) leaves.Similar investigations have also been conducted for some other biomasses [26,35,40].
Based on literature, the process performance and yield of Delonix regia pods (DRP) in the retort carbonisation process is not understood.Besides, the retort carbonisation technology is a novel technique especially suited to the peculiarities of the sub-Saharan Africa energy conversion challenges.There is a need to explore the potential of the current feedstock in producing high-quality biochar from the low-temperature retort carbonisation process.This can help justify the appropriateness of a low-temperature process for producing biochar with properties that could make it agreeable for multiple energy and environmental applications.The aim of this study was to investigate the production performance and characterisation of biochar from DRP via the retort carbonisation process.The study also analytically evaluated the characteristics of the product obtained from the process.

Sourcing and preparation of DRP
The matured dried DRP were collected from different locations within the University of Ilorin, Kwara State, Nigeria.Before sun-drying, the pods were split and the inner seeds were removed.The collected materials were thoroughly cleaned to remove dust and other adhering impurities.The cleansed DRP were dried to constant weight at 105 � C for 3 hrs using an electric oven.The dried pods were trimmed to 5-10 cm in length.The size reduction was done so that the DRP feedstock can fit in the reactor chamber as well as allowing natural convective air flow.The combustion fuels used for the study were the stems of bamboo (Bambusa vulgaris) and African balsam (Daniellia olivieri) stalks.They were obtained within the University of Ilorin community and were chopped into pieces of average length 3 -5 cm for easy combustion.The proximate and ultimate analysis of the fuel and feedstock are reported in Table 1.The methods for their determination is reported in the supplementary material.

Experimental set-up and batch procedure
An updraft biomass retort carboniser made of stainless steel was used in this study for the production of biochar and described in Figure 1.The reactor is 23 cm high with a radius of 14.5 cm while the inner carbonisation unit is 18.5 cm high with a radius of 12 cm.Full details of the dimensioning of the reactor is presented in Table S1.The points of temperature measurements were at the sides (T a , T b , T c ) and Top (T d ) using the incident infrared beam thermometer.A CASON CA380 infrared thermometer (accuracy ± 0.1 � C) was used for temperature measurement during the process.In taking readings, the thermometer was set to read 'within body' measurements with incident beams from the side of the reactor.However, inner reactor temperature readings were taken using an incident from the top of the reactor through the opening of the exhaust.The reactor contains a small, centrally placed carbonisation chamber, usually covered with a lid but with associated airholes (Figure 1).The carbonisation chamber contained the DRP for the carbonisation while the stems of bamboo (Bambusa vulgaris) and African balsam (Daniellia olivieri) stalks for the production of heat occupies the heating gap (the space between the inner carbonisation chamber and the reactor).For each experimental run, 100 g of the feedstock was placed in the carbonisation chamber.For the mixture of bamboo (Bambusa vulgaris) and African balsam (Daniellia olivieri) stalks, a 1:1 (by weight) mixture was used till the carbonisation chamber was full.
A vertical exhaust enabled the elutriation of product gases being emitted.The holes at the base of the retort carboniser allow for the air updraft by natural convection to the ringed combustion front (Figure 1).The combustion zone is the ringed region between the carbonisation chamber and the outer reactor vessel.The reactor uses the toplit updraft approach that enables the combustion fuel when ignited at the top and burn slowly downwards (with bottom to top air draft) until the combustion fuel is spent.

Determination of yield and characterisation of DRPB
The product obtained was named Delonix regia pods biochar (DRPB).The experiments were conducted twice.The percentage yield DRPB (Yield DRPB ) was calculated using Equation ( 1) where is the mass M DRP is the mass of DRP biomass fed into the reactor at the start of the process and M DRPB is the mass of biochar at the end of the process.Gas yield can be determined by material balance.

Temperature profile and product yield
The peculiarity of the retort carbonisation process is that there are no temperature controls and it is self-regulating based on the nature of the combustion feed.Hence, the temperature profile needs to be monitored and evaluated to determine the performance of the system [38].The performance of the reactor was evaluated based on the process time, peak temperatures, and biochar yield.The process time and peak temperatures were explicated from the temperature profile.The temperature profile of the retort carbonisation process for the conversion of DRP into DRPB is shown in Figure 2. Based on the reading at all regions of the reactor (T a , T b , T c and T d ), the temperature can be observed to increase steadily from ambient conditions (25 ± 2 � C) till peak temperature was attained after 30 min.This was then followed by a gradual drop in temperature until the combustion fuel was used up and ambient condition was achieved.Similar temperature profiles have been reported for retort carbonisation albeit for oil palm fibers [43] and sugarcane bagasse [44].The rising temperature at the beginning of the process is expected as the controlled combustion in the combustion gap leads to heat transfer into the carbonisation chamber.There is a natural convection draft of air from the bottom air holes upwards towards the controlled combustion zone (updraft) to sustain the process.This combustion zone slowly consumed the heating fuel in the gap and proceeds from the upward region downwards toward the bottom of the reactor.In the carbonisation chamber, the feedstock is being thermally converted without any fresh oxygen supply.The partially burnt exhaust which is still rich in CO escapes from the carbonisation chamber into the combustion gap.This is then combusted immediately it enters the combustion gap thereby providing extra thermal energy from the process.The concept of 'retort' in the process is this secondary combustion of escaped gaseous product from the carbonisation chamber.This has also been implemented in other designs by Chandrasekaran et al. [38], Abdelhafez et al. [45] and Padakan [40].The current study uses a cylindrical design of the reactor which helps to develop a synergistic heating effect on the central carbonisation chamber.Retort heating helps the reactor hit high temperatures for a sustained period using only a very small amount of heating biomass.The profile in Figure 2 reveals a gradual drop in temperature till ambient conditions are achieved after 150 min of starting the process.A yield of 29.48 wt% was obtained.This value is observed to be one of the highest when compared to other pure biomass feedstock for the retort carbonisation process (Table 3).Only feedstock with biomass plastic blends achieved significantly higher yields than for the current study.Besides the achievement higher temperatures, it is important that these are burnt-off to avoid the greenhouse effect.CO has a high emission factor [27] and could be deleterious to the environment if not harnessed.

Textural properties of DRPB
The textural properties of DRPB were determined by BET and summarised in Table 2.The specific surface area and average pore size were 88.03 m 2 /g and 0.0352 cm 3 /g respectively.These are important properties that determine how well the material can function as an adsorbent [46,47].These serve as the surface onto which adsorbate species are attached and removed from the aqueous phase [48].However, the surface area is comparatively poorer than those of other biomass feedstock (see Table 3).This is due to the compositional nature of the biomass feedstock.The BET average pore size was < 2 nm suggesting that the obtained biochar was nanoporous.This is the first reported nanoporous biochar from the retort carbonisation process.N 2 adsorption-desorption isotherms for DRPB (Figure S1).The figure shows type II isotherm and confirms that micropores are responsible for adsorption at a relative pressure of 0 to 0.17, the multilayer adsorption by the mesoporous content of the biochar started around a relative pressure of 0.2, up to 0.8 [49,50].The Barrett, Joyner, Halenda (BJH) pore size distribution for DRPB is shown in Figure S2.The BJH pore size distribution confirms that the majority of the pores which participated in adsorption were less than 1000 Å, correlating with type H1 hysteresis loop.In addition, the desorption profile of the prepared biochar occurred at a relatively higher pressure, this indicates that the biochar requires more pressure to desorb its content, hence it can be recommended for long-term adsorption and quantification applications.

Morphological properties of DRPB
The morphological analysis of DRPB was conducted using SEM and 3D reconstruction.The SEM micrograph is shown in Figure 3(a,b) while the 3D reconstruction is shown in Figure S3.The SEM reveals that DRPB has a surface with deep round gorges and crack lines where flakes have broken off.There are therefore corresponding flaky particles  scattered around the biochar surface.3D reconstruction was done to help visualise the undulating nature of the surface morphology at an angled perspective so that the comparative depth and height of features can be deduced [42], though it cannot fully detect minute particles on the biochar surface.SEM gives a direct vertical view that sometimes makes it difficult to appreciate these features.For DRPB, Figure S3 reveals that the surface has a heterogeneous nature.The 3D reconstruction software is also used to compute the roughness based on the micrograph.An average roughness (R a ) value of 12.96 � 10 3 mm was obtained.48.99 pixel (PX) is the original value which was converted by the factor 1 PX ¼ 264.5833 mm.This R a value shows that the biochar surface is quite heterogeneous as it is higher than any conventional standard N-rating for smooth surfaces.

Functional group properties of DRPB
FTIR was used to analyse the functional groups of DRPB.
The spectrum (presented in Figure 4) shows a wide variety of peaks showing that numerous functional groups are present [51,52].This is characteristic of materials obtained from thermochemical processing [53].The group frequencies are characteristic of small groups of atoms or functional groups and are observed at the higher frequencies of the spectrum.The sharp peak at 3677 cm −1 is due to the stretching vibrations of the O-H bond in alcohols.The presence of free -OH in DRPB.The sharp peaks at 2901 and 2973 cm −1 are due to the stretching vibrations of C-H.The peak at 1740 cm −1 can be assigned to the stretching vibrations of C ¼ O in carboxylic acid and esters.The peak at 1584 cm −1 can be assigned to the stretching vibrations of N-O in the nitro group.The fingerprint frequencies are highly characteristic of molecules as a whole and are observed at the lower frequencies in the FTIR spectrum (below 1500 cm −1 ).The peak at 1394 cm −1 can be assigned to the stretching vibrations of S ¼ O.The peak at 1229 cm −1 can be assigned to the stretching vibrations of C-O.The peak at 1022 cm −1 can be assigned to the stretching of CO-O.The peak at 879 cm −1 can be assigned to the bending vibrations of C-H.These fingerprint frequencies suggest that the main elements present in the biochar are carbon, hydrogen, and oxygen, it also reveals the major bonds present between these species.

Comparison with other retort carbonisation biochar
To evaluate the potential of the biomass as a feedstock for the retort carbonisation process, the quality of DRPB obtained is compared with those of other biomasses also from retort carbonisation.This is summarised in Table 3 and sorted in order of decreasing biochar yield.It should be noted it is impossible to make a comparison at similar temperatures and times for retort carbonisation because the process is self-regulating, and these are dependent on the nature and volume of the biomass heating source and design of the controlled combustion chamber.It can be observed from the comparison that DRP gives one of the highest yields when compared to other pure biomass feedstock for the retort carbonisation process.After Prosposis Juliflora, it has the highest yield for pure biomass for the process.Only feedstock with biomass plastic blends achieved significantly higher yields than for the current study.From Table 3, it can be observed that the BET specific surface area is comparatively low in light of other biomass materials.However, it has a BET average pore size in the nanoporous range (< 2 nm) based on standard classification.This is the lowest BET average pore diameter reported in the literature for retort carbonisation biochar.There are several potential reasons for the intermediate/low surface area observed in this study.The nature of the feedstock plays an important role.For very dense feedstock, the eventual surface area of the product can be lower as the migration of the gaseous phase from the organic matter could be limited during heating [54].Also, the ash content of the feedstock can reduce the surface area as the fusion of ash can fill up the pores in the biochar [55].The low surface area could be due to the loss of secondary volatiles during the heating process [56] which escape into the carbonisation chamber and are used up.

Relevance of the study in Sub-Saharan Africa
Several practical implications are derived from the study which are discussed herein.Furthermore, potential applications of the product (based on the deduced properties) are also discussed.DRPB, as introduced in the paper does not have a competitive use at the current time.This means the current investigation is a viable solid waste management technique, at least in the context of sub-Saharan Africa.With the recent drive for improved waste-to-wealth initiatives in Nigeria and other sub-Saharan African countries [62], such technologies could be interesting to potential investors.Other technical advantages of retort heating technology such as no electrical requirement and ease and inherent safety of operation are discussed elsewhere [44,57].Also, since the retort process helps in combusting the CO-rich carbonisation off-gases, it can be considered an environmentally friendly and clean technology [63,64].
Based on the biochar properties it is suggested that DRPB can be explored as an adsorbent.The level of specific surface area and average pore size suggest it can accommodate a surface phenomenon like adsorption [65].Heterogeneous surface morphology and high roughness are also observations that reinforce this idea.Functional groups like O-H, C ¼ O, C-O, N-O, and CO-O possess heteroatoms hence adsorptive interactions by hydrogen bonds are possible with most species.Though this is a weak physical interactive force, chemical interactions are also possible via other mechanisms (though these require further investigations).The positive performance of retort carbonisation biochar has been investigated in other studies [66,67] albeit using other feedstock.Other investigations have also confirmed that retort carbonisation biochar is also usable for the improvement of the tribological and rheological properties of lubricating oils [68].Studies have shown that the addition of a very small quantity of biochar to lube oil helps it retain its kinematic viscosity for longer hence improving operational performance in vehicles [68].This is because biochar seems to have the right quantity of inorganics complemented by an unwillingness to contaminate the base oil (relative inertness).Future studies could attempt to model the relationship between the process parameters and some of the quantitative product quality metrics.There is a recent flurry of the application of intelligent system [69][70][71][72] and this could also be considered for retort canonisation.

Conclusion
Based on the analysis of results, several key conclusions were derived from the study.The retort carbonisation process achieved a biochar yield of 29.48 wt% at a peak temperature of 375 � C and process time of 150 min.The specific surface area and average pore volume of DRPB were 88.03 m 2 /g and 0.0352 cm 3 /g respectively.The BET average pore size was < 2 nm suggesting that the obtained biochar was nanoporous.Morphological analysis revealed that the biochar had a heterogeneous surface morphology and average roughness (R a ) value of 12.96 � 10 3 mm.FTIR analysis revealed functional groups such as C-O, N-O, O-H, C ¼ O, CO-O, and C-H present in DRPB.The current feedstock gave a nano-porous of biochar from retort carbonisation.For future investigations, attempts can be made to explore its adsorbent properties for water treatment applications.It could also be attempted as an additive for improving the tribological and rheological properties of lubricating oils.

Figure 1 .
Figure 1.Description of the DRP sample, reactor design and general scheme of the process.

Figure 2 .
Figure 2. The temperature profile of the retort carbonisation of DRP into DRPB.

Table 3 .
Comparison of yield and properties of retort carbonisation biochar.