CHEMICAL ENGINEERING TRANSACTIONS 

VOL. 81, 2020 

A publication of 

The Italian Association 
of Chemical Engineering 
Online at www.cetjournal.it 

Guest Editors: Petar S. Varbanov, Qiuwang Wang, Min Zeng, Panos Seferlis, Ting Ma, JiΕ™Γ­ J. KlemeΕ‘ 
Copyright Β© 2020, AIDIC Servizi S.r.l. 

ISBN 978-88-95608-79-2; ISSN 2283-9216 

Kinetic Model of Steam Reforming for Heavy Tar 

Decomposition in Biomass Gasification 

Weiliang Wanga, Qiang Lib, Qian Wanga,*,Hui Denga 

aEnergy and Electricity Research Center, Jinan University, No. 206 Qianshan Road, Zhuhai, Guangdong 519070, China 
bKey Laboratory for Thermal Science and Power Engineering of the Ministry of Education, Department of Thermal  

 Engineering, Tsinghua University, Beijing, 100084 China 

 qianwang@jnu.edu.cn 

Gasification technology as a feasible clean and efficient utilization technology has been developed to convert 

the low-grade energy resources like biomass into high value syngas.The generation of syngas is often 

accompanied by tar formation, which causes blockage in the downstream processing as tar condenses. The 

elimination of naphthalene and pyrene as heavy tar model compounds by steam  reforming was studied by 

experiments in a horizontal tube reactor and simulation with CHEMKIN. A good agreement between the 

measured and calculated results of light gas products and soot was obsvered, and one can found that the 

reactivity of naphthalene is higher than that of pyrene in the presnece of steam and most of carbon of both 

hydrocarbons convert into soot instead of light gaseous products. The reaction pathways of steam reforming 

were developed by sensitivity analysis and rate of production. Benzene and naphthalene, represented as 

precursor of light gas product, are dominated intermediate components of naphthalene and pyrene. The 

simplified reaction schemes including the reaction pathway as well as the associated kinetics were derived by 

CHEMKIN. 

1. Introduction

To meet the rapidly growing demand of energy and deal with the consequent environmental issues, the 

gasification technology as a feasible clean combustion technology has been developed to convert  biomass into 

high value syngas (Siwal et al., 2020). During gasification, many by-products are generated such as variable 

amounts of fly ash, volatile alkali metals and tar (Palma et al., 2013). In particular, tar formation often occurs in 

the syngas, causing blockage in the downstream processing as it condenses. It is desire to decompose the tar 

into gaseous products as much as possible during the gasification process. 

Tar is a complex mixture of condensable hydrocarbons. According to its appearance, tars can be classified as 

the primary, secondary and tertiary ones. The primary tar mainly consists of oxygenated compounds and its 

formation takes place at a low temperature (673 - 873 K), like acetol and acetic acid (Fitzpatrick et al., 2009). 

The secondary tar containing phenolics and olephines, such as phenol, benzene and toluene, may be produced 

in a medium temperature range (873 - 1,073 K), while the tertiary tar comprising aromatic compounds is formed 

at high temperature range (1,073 - 1,273 K) (Ren et, al., 2020). Heavy tars like large molecular PAHs are 

generally considered as the main tertiary tar (Apicella et al., 2017). Further effort is required to explore the 

decomposition of heavy tar. 

Although many methods have been investigated to remove the tar from product gas (Ashok et al., 2020), steam 

forming of tar seems to be the best way that can not only eliminate tar but also increase hydrogen yield. In case 

of steam reforming, the effects of operational conditions such as temperature, steam/carbon molar ratio and 

residence time on model compounds of the secondary such as benzene, toluene, naphthalene and phenol have 

been partly studied (Artetxe et al., 2017). However, the mechanisms and kinetics of their decomposition were 

often ignored. Very few data related to the kinetic parameters and reaction pathways were obtained. Given the 

complexity of reactions, the majority of the kinetic models obtained by experimental works were overall reaction, 

which were often determined by the overall weight loss of the tar or the light gaseous products such as CO, 

CH4, and H2. The work of Jess (1996) was one of the few that availably presented an reaction scheme of 

 

  DOI: 10.3303/CET2081209 

 

 

 
 
 

 
 

 
 
 

 
 
 

 
 
 

 

 

 
 
 

 
 
 

 
 
 

 
 
 

 
 

 
 
 

 
 
 

 
 
 

Paper Received: 03/04/2020; Revised: 04/07/2020; Accepted: 05/07/2020 
Please cite this article as: Wang W., Li Q., Wang Q., Deng H., 2020, Kinetic model of steam reforming for heavy tar decomposition in biomass 
gasification, Chemical Engineering Transactions, 81, 1249-1254  DOI:10.3303/CET2081209 

1249



aromatic hydrocarbons (naphthalene, toluene and benzene) removal in the presence of steam and hydrogen 

was developed. In the scheme, naphthalene reforming was expressed as the light gaseous formation by 

cracking reactions and soot precursors formation by condensation reactions. The kinetic parameters of the 

formation of soot and cracking products have been calculated.  

In this paper, naphthalene and pyrene, representing two-ring PAH and four-ring and above PAH, are used as 

model compounds of heavy tar in the steam reforming process. The comparison between the experimental 

results and the numerical predictions by CHEMKIN will be performed. Combined the results of experiments and 

simulations, a kinetic model including the reaction pathway as well as the associated kinetics is obtained for 

steam reforming of these model compounds. 

2. Experimental approach

2.1 Experimental system 

A schematic diagram of experiment system is shown in Figure 1. A horizontal corundum tube with inner diameter 

of 30 mm and length of 800 mm was placed inside an electric furnace heating by silicon molybdenum rods. The 

temperatures along the tube reactor were measured by a Pt-Rh thermocouple. The steam was vaporized in a 

steam generator and mixed with hydrocarbons (naphthalene, pyrene) evaporated by electric tracing band. Argon 

as a carrier gas flowed through the electric tracing band and pushed the mixture through the reaction zone with 

length of 300 mm when operation temperature reached the preset value. The mass flow rates of argon and 

steam were well controlled. The product gases were collected by gas storage bag after they passed through a 

condensing and drying tube. The gases were analyzed offline using gas chromatograph (GC). The soot was 

absorbed by quartz wool placed in the outlet of reactor and burned by oxygen in reaction zone. The content of 

soot was calculated by the combustion products analyzed by GC. 

Reaction conditions of the experiments are listed in Table 1. Experiments were carried out at atmospheric 

pressure within a range of 1,273 - 1,673 K in reaction zone. Naphthalene as typical two ring polyaromatic 

hydrocarbons and pyrene as typical four ring polyaromatic hydrocarbons were used as model compound 

representing heavy tars.  

Figure 1: Experimental system of heavy tar steam reforming 

Table 1: Reaction conditions of the experiments 

Model compounds Sample weight (mg) Temperature (K) Steam (vol%) Residence time(s) 

Naphthalene 10 1,273 - 1,673 30 2 

Pyrene 10 1,373 - 1,673 30 2 

2.2 Conversion 

According to the previous work of Jess (1996), the reaction of aromatic hydrocarbons reforming can summarized 

two reaction pathways: (a) Cracking reactions of PAH leading to hydrocarbons with small carbon numbers; 

(b)Condensation reactions leading to soot precursors. The conversion of model compound j with carbon number 

NC,j cracking into light hydrocarbons (CO, CO2, CH4) is defined as: 

𝑋𝑐,𝑗(%) = (𝑛𝐢𝑂,π‘œπ‘’π‘‘ + 𝑛𝐢𝑂2,π‘œπ‘’π‘‘ + 𝑛𝐢𝐻4,π‘œπ‘’π‘‘) (𝑛𝑗,𝑖𝑛 βˆ™ 𝑁𝐢,𝑗)⁄ Γ— 100 (1) 

The yield of soot is calculated based on the amounts of CO and CO2 when the collected sample is fully burnt. 

The conversion of model compound j forming soot precursors can be expressed as: 

1250



𝑋𝑠,𝑗(%) = π‘›π‘ π‘œπ‘œπ‘‘,π‘œπ‘’π‘‘ (𝑛𝑗,𝑖𝑛 βˆ™ 𝑁𝐢,𝑗)⁄ Γ— 100 = (𝑛𝐢𝑂,π‘ π‘œπ‘œπ‘‘ + 𝑛𝐢𝑂2,π‘ π‘œπ‘œπ‘‘) (𝑛𝑗,𝑖𝑛 βˆ™ 𝑁𝐢,𝑗)⁄ Γ— 100 (2) 

The conversion of the model compound j can be calculated by:  

𝑋𝑗(%) = 𝑋𝑐,𝑗 + 𝑋𝑠,𝑗 (3) 

3. Simulation calculation

Numerical simulation for heavy tar steam reforming was conducted using the Plug-flow Reactor module of 

CHEMKIN. A detailed chemical kinetic model for hydrocarbon combustion and polycyclic aromatic hydrocarbon 

developed by Richter and Howard (2002) was chosen. It was used to successfully predicted the thermal 

reactions of aromatic hydrocarbons from pyrolysis of solid fuels (Norinaga et al., 2011) and characteristics of 

partial oxidation of gas emitted from metallurgical coke ovens (Norinaga et al., 2010). The reaction conditions 

used in the simulation were consistent with the experimental conditions. 

4. Results and discussion

4.1 Experiment and simulation analysis 

Figure 2 shows the experimental conversion of PAHs, the cracking gas products and condensable deposited 

soot as a function of reference temperature TR under the reaction conditions. It is found that polymerization and 

condensation lead to approximate 50 % carbons in PAH convert into soot precursors when temperature is up 

to 1673 K. Most of carbon of both of them convert into soot instead of light gaseous products. The main gaseous 

products were CO, CO2 and CH4. The content of soot was calculated by the amounts of combustion products 

(CO and CO2) of the deposited soot precursors. With increasing temperature, the contents of soot, generated 

by naphthalene steam reforming, increases at first then decreases after 1,573 K. It is unraveled that the reaction 

of soot with steam at elevated temperature caused concentration of soot to decrease. The contents of soot of 

pyrene steam reforming rapidly increased in the temperature range from 1,373 K to 1,473 K, but only a slight 

increase was observed when temperature is up to 1,473 K. The possible reason is also the secondary cracking 

of soot with steam, while the reaction rate of soot formation is faster than decomposition rate of soot. Figure 2 

also compares the conversion derived from the numerical simulation with those obtained by experiments in 

different temperature. Good agreements are found for both conversion of light gas products and soot. The 

overall conversion of pyrene is depended on the above two conversions, which is also predicted well. 

（aοΌ‰Conversion of PAHs （bοΌ‰Cracking reactions （cοΌ‰condensation reactions 

Figure 2: Conversion of naphthalene and pyrene at different temperatures(line-simulation,dot-experiment) 

4.2 Reaction scheme 

The steam reforming of heavy tar involves a set of complex reactions among PAH, product gases, steam, soot 

and intermediate products. However, no detailed kinetic modelling is available because of lack of information of 

intermediate products. In this section, the main reaction pathway for tar steam reforming was obtained by 

distributions of main components through CHEMKIN. 

4.2.1 Naphthalene 

Figure 3 gives the predicted distributions of main components in naphthalene steam reforming as a function of 

resident time (length of reactor) at 1,573 K. The reaction of aromatic hydrocarbons reforming are summarized 

into two reaction pathways. The first reaction pathway is the cracking reactions leading to hydrocarbons with 

1251



small carbon numbers, and the other one is condensation reactions accompanying with the growth of larger 

PAHs (soot precursors) (Figure 3b). As shown in Figures 3a and 3c, the consumption of naphthalene finishes 

in 0.5 s. Indene and benzene are two dominating intermediates of cleavage of naphthalene. Indene is a stable 

intermediate of naphthalene in gasification condition. The initial step of decomposition of naphthalene is 

naphthol formation by substitution reaction with steam. After that, naphthol is converted to indene through 

intermediate naphthoxy radical. Benzene as a key component of degradation of aromatic hydrocarbons is 

subsequently formed by deprivation carbons of indene. Consecutive reactions of benzene generate light 

gaseous products. The other direction of conversion of naphthalene is toward soot formation. In addition to the 

substitution reaction with OH radicals, the reaction of attack of H radicals that supplies naphthyl radical is also 

an important initial reaction for naphthalene steam reforming, leading to the formation of larger PAHs. The rate 

of naphthyl radical formation by attack of H radical proposed by Norinaga et al. (2011) is higher than that of 

naphthol formation by substitution reaction with OH radicals beyond 1,273 K. It means that more carbons in 

aromatic hydrocarbon are converted to soot than that in light gas products. This is strongly supported by the 

experimental results in Section 4.1. The concentrations of benzofluoranthene and coronene shown in Figure 3d 

are two largest ones among those of thirty-three PAHs, while their formation pathways are different. The 

combination of naphthalene with naphthyl radical results in the benzofluoranthene formation, whereby hydrogen 

is liberated. Coronene formation is based on the HACA (Hydrogen Abstraction and Acetylene Addition), which 

is a dominant pathway for the growth of PAHs up to the formation of soot. The mechanism of HACA means that 

one mole of hydrogen is removed while adding one mole of acetylene to form aromatic ring. The secondary 

cracking of soot precursors as PAHs with steam after 0.5 s is similar to process of naphthalene steam reforming. 

According to the experimental and stimulative results, a simplified reaction scheme for naphthalene steam 

reforming is developed, as shown in Figure 4. In the scheme, the detail of soot formation and decomposition is 

simplified, then indene and benzene are the dominated intermediates. 

Figure 3: The concentration of main components in naphthalene steam reforming as a function of resident time 

(length of reactor) at 1573 K in numerical simulation 

Figure 4: Simplified reaction paths for naphthalene steam reforming

4.2.2 Pyrene 

Figure 5 gives the predicted distributions of main components in pyrene steam reforming as a function of 

resident time (length of reactor) at 1673 K. As shown in Figure 5a, there is still a small amount of residue at 2.0 

s, which validates that the decomposition rate of pyrene is slower than that of naphthalene. Like decomposition 

1252



of naphthalene with steam, the steam reforming of pyrene is a complex process. The simulation results of 

important intermediates and soot are presented in Figure 5b and Figure 5d. Phenanthryl is a primary reaction 

product, resulting from the substitution reaction with OH radicals, whereby ethenone is liberated. Ethenone 

subsequently converts to CO and H2. Phenanthryl can undergo hydrogenation reaction to phenanthrene, which 

is a stable intermediate in the first stage as shown in Figure 5c. Naphthalene is the second stage intermediate, 

formed by the reverse process of HACA reaction of phenanthryl, namely hydrogen addition and acetylene 

abstraction. The decomposition of naphthalene leads to light gaseous product, which is same with the 

explanation in Section 4.2.1. Likewise, the other direction of conversion of pyrene is toward soot formation. The 

presence of pyrenyl, resulting from the reaction of attack of H radicals, is assumed to be a precondition for soot 

formation. In case of pyrenyl, a reversible reaction occurs, namely formation of cyclopenta[cd]pyrene in the first 

0.25 s and decomposition of that after as shown in Figure 5d. The reversible reaction slows down the reduction 

tendency of pyrene after 0.25 seconds. Similar to naphthalene, the main PAH of soot precursors in pyerne 

steam reforming is coronene based on results shown in Figure 5d. Naphthalene is the dominated intermediate 

and the detail of soot formation and decomposition is simplified. A simplified reaction scheme for pyrene steam 

reforming is shown in Figure 6. 

Figure 5: The concentration of main components in pyrene steam reforming as a function of resident time (length 

of reactor) at 1,673 K in numerical simulation 

Figure 6: Simplified reaction paths for naphthalene steam reforming

4.3 Kinetic model 

According to the simplified reaction scheme above, the primary and consecutive reactions have been given. 

However, due to lack of the concentration of intermediate products, it is hard to present the kinetic parameters 

of the primary reactions involving formation of the precursor of light gas product and soot in in steam reforming. 

To obtain the complete the kinetic model of reaction scheme, the reaction rate of formation of precursor of light 

gas product rc and the rate of soot formation rs were determined from the simulations along length of tube (varied 

residence time) with varied reaction temperature. Benzene and naphthalene were represented as precursor of 

light gas product of naphthalene and pyrene. The kinetic parameters were calculated by the conversion rate of 

the primary products in different resident time and reaction temperature. The kinetic parameters of primary 

reactions of naphthalene and pyrene are summarized in Table 2. The reaction rates of consecutive reaction of 

decomposition of benzene, methane and soot are derived by the work of Jess (1996).  

1253



Table 2: Reaction conditions of the experiments 

Reaction rate Pre-exponential factor Activation energy (kJ mol-1) 

π‘Ÿπ‘,𝑛 = 𝑑𝑐𝐢6𝐻6 π‘‘πœβ„ |𝐢10𝐻8→𝑏𝑒𝑛𝑧𝑒𝑛𝑒 2.72Γ—10
13 m0.3mol-0.1s-1 360 

π‘Ÿπ‘ ,𝑛 = π‘‘π‘πΆπ‘ π‘œπ‘œπ‘‘ π‘‘πœβ„ |𝐢10𝐻8β†’π‘ π‘œπ‘œπ‘‘ 2.66Γ—10
9 m0.9mol-0.3s-1 296 

π‘Ÿπ‘,𝑝 = 𝑑𝑐𝐢10𝐻8 π‘‘πœβ„ |𝐢16𝐻10β†’π‘›π‘Žπ‘β„Žπ‘‘β„Žπ‘Žπ‘™π‘’π‘›π‘’ 1.56Γ—10
8 m-0.3mol0.1s-1 250 

π‘Ÿπ‘ ,𝑝 = π‘‘π‘πΆπ‘ π‘œπ‘œπ‘‘ π‘‘πœβ„ |𝐢16𝐻10β†’π‘ π‘œπ‘œπ‘‘ 4.09Γ—10
14 m-2.1mol0.7s-1 476 

5. Conclusions

The steam reforming of naphthalene and pyrene in a horizontal tube reactor was carried out. The results showed 

that the reactivity of naphthalene is higher than that of pyrene of the steam reforming. Most of their carbon is 

converted into soot instead of light gaseous products. The numerical results predicted by CHEMKIN agree well 

with the experimental data under the given reaction conditions.The simplified reaction pathways of naphthalene 

and pyrene are developed by means of simulation. The reaction pathways of naphthalene and pyrene are 

similar. The substitution reaction with OH radicals is regarded as the first step of PAH steam reforming leading 

to precursor of light gas product, while the reaction of attack of H radicals that supplies the naphthyl or pyrenyl 

radical is a relatively important initial reaction for the formation of soot precursor. Benzene and naphthalene, 

represented as precursor of light gas product, are dominated intermediate components of naphthalene and 

pyrene. Consecutive reactions of these intermediates subsequently generate light gaseous products. Coronene 

is found to be the largest amount PAH among the thirty-three PAHs substitute of soot precursor. The simplified 

reaction schemes for naphthalene and pyrene steam reforming are developed and kinetic parameters of the 

primary reactions are derived. 

Acknowledgements 

This work was supported by the National Key Research and Development Program of China 

(2016YFB0600205) 

References 

Apicella B., Tregrossi A., Stazione F., Ciajolo A., Russo C., 2017, Analysis of petroleum and coal tar pitches as 

large PAH, Chemical Engineering Transactions, 57, 775-780. 

Artetxe M., Alvarez J., Nahil M.A., Olazar M., Williams P.T., 2017, Steam reforming of different biomass tar 

model compounds over Ni/Al2O3 catalysts, Energy Conversion and Management, 136, 119-126. 

Ashok J., Dewangan N., Das S., Hongmanorom P., Wai M.H., Tomishige K., Kawi S., 2020. Recent progress in 

the development of catalysts for steam reforming of biomass tar model reaction, Fuel Processing 

Technology, 199, 106252. 

Fitzpatrick E.M., Bartle K.D., Kubacki M.L., Jones J.M., Pourkashanian M., Ross A.B., Williams A., Kubica K., 

2009, The mechanism of the formation of soot and other pollutants during the co-firing of coal and pine wood 

in a fixed bed combustor, Fuel, 88(12), 2409-2417. 

Garcia X.A., HΓΌttinger K.J., 1989, Steam gasification of naphthalene as a model reaction of homogeneous 

gas/gas reactions during coal gasification, Fuel, 68(10), 1300-1310. 

Norinaga K., Yatabe H., Matsuoka M., Hayashi J.I., 2010, Application of an existing detailed chemical kinetic 

model to a practical system of hot coke oven gas reforming by noncatalytic partial oxidation, Industrial & 

Engineering Chemistry Research, 49(21), 10565-10571. 

Norinaga K., Sakurai Y., Sato R., Hayashi J.I., 2011, Numerical simulation of thermal conversion of aromatic 

hydrocarbons in the presence of hydrogen and steam using a detailed chemical kinetic model, Chemical 

Engineering Journal, 178, 282-290. 

Palma, C.F., 2013, Modelling of tar formation and evolution for biomass gasification: A review, Applied Energy, 

111, 129-141. 

Ren J., Liu Y.L., Zhao X.Y., Cao J.P., 2020, Biomass thermochemical conversion: A review on tar elimination 

from biomass catalytic gasification, Journal of the Energy Institute, 93(3), 1083-1098. 

Richter H., Howard J. B., 2002, Formation and consumption of single-ring aromatic hydrocarbons and their 

precursors in premixed acetylene, ethylene and benzene flames, Physical Chemistry Chemical Physics, 

4(11), 2038-2055. 

Siwal S.S., Zhang Q., Sun C., Thakur S., Gupta V.K., Thakur V.K., 2020, Energy production from steam 

gasification processes and parameters that contemplate in biomass gasifier-A review, Bioresource 

Technology, 297, 122481 

1254