Considerable effort and resources are currently being channelled into the development of alternative energy from biological and natural sources. Chief among them is the production of biodiesel from the transesterification of oils. Biodiesel is bio-degradable, non-toxic and environmentally friendly . The by-product of the process is glycerol (C3H8O3). In converting vegetable oils into biodiesel, approximately 10 % (w/w) of glycerol is produced as a by-product .
With the continual increase in biodiesel production, a glut of glycerol is expected in the world market. Currently, there are many applications of glycerol and these include personal care products, food, oral care, tobacco, and polyurethane production . Most by-product glycerol is sent to water treatment for digestion but this process is slow, expensive and demonstrates low yield. It also can be purified by distillation. However, this is an expensive process and the low market cost of glycerol makes it uneconomic . The glycerol price dropped by two-thirds within the last decade  because of surplus amounts in the market and is still dropping in the current decade. There is a current search for more alternative uses of glycerol as Increasing production of biodiesel will lead to glycerol supply exceeding demand.
One viable and proven possibility is using it as a source for the production of hydrogen. Gasification and aqueous phase reforming are also other processes of producing hydrogen from bio-based sources . Bio-resources based processes for Hydrogen gas synthesis are seen as viable options for the future due to their theoretically carbon-neutral nature. Hydrogen poses no environmental problem and also does not contribute to atmospheric carbon dioxide emissions . Alongside glycerol, pyrolysis oil (another bio-resource) also interestingly has been proved to be okay for steam reforming for hydrogen production [7, 8]. At present, almost 95 % of the hydrogen produced is obtained from fossil fuel based feedstock  and it is used majorly as a chemical ingredient in petrochemical, metallurgical, food, and electronics processing industries. Also, the global demand for hydrogen gas, is growing due to the technological advancements in fuel cell industry . Hydrogen gas possesses the highest energy content per unit of weight (120.7 kJ/g), compared to any of the known fuels .
The steam reforming of Glycerol has been extensively studied experimentally over the years with much focus on energy and exergy analysis , kinetic modelling and thermodynamic analysis [3, 12, 13, 14, 15, 16, 17, 18, 19], sorption enhancement techniques [14, 20], the minimisation of carbon deposition and the trials and effectiveness of different catalyst [2, 5, 9, 12, 21, 22, 23, 24, 25, 26, 27, 28, 29].
Several simulation studies has also been carried out on the steam reforming of glycerol [30, 31] but the focus was majorly on gas flow patterns and velocity. Several factors affect the reactions involved in the steam reforming of glycerol. Among them are reaction temperature, system pressure and steam to glycerol ratio . This study is essentially and in-depth investigation of the interaction of these key factors and their effect on the selectivity of Hydrogen from the process. The basis of the investigation will be a simulated model of the steam reforming process using ASPEN plus V8.8. The model will serve as a reference system  to identify reaction responses to the different factors and how they interact.
Steam reforming is a high temperature process whereby Glycerol is converted to a Hydrogen rich gaseous product known as synthesis gas. The chemical reactions involved in the steam reforming of glycerol into synthesis gas are well understood and have been properly studied [24, 27, 33]. eqs (1–3) give a sequence of the reaction.
Glycerol decomposition; (1)
Water-Gas Shift reaction; (2)
Methanation reaction; (3)
The combination of reactions (1) and (2) gives the overall equation of reaction as considered in this study.
2.1 Performance parameters
The key performance parameters of the system are the reaction conversion, the yield of hydrogen, and the extent of carbon deposition. Studying the system of eqs (1–3), it will be noticed that reaction conversion will essentially always be 100 %. This is due to the fact that glycerol decomposition is in the presence of steam, but this cannot be adequately expressed directly while specifying the equations in ASPEN Plus. In practice, there is a wide range of parameter allowance for which the reaction will always proceed to 100 % completion. Coupled with the fact that metal catalysts are generally used in the industry for the process conversion is not our interest as maximum conversion can be easily achieved. The eqs (4–7) were used to calculate the performance of the system.
The Optimisation goal is to maximise the hydrogen gas yield from the system, and to minimise methanation. The key factors affecting the performance parameters/system responses are Reaction Temperature, Pressure and steam to glycerol ratio (STGR).
2.2 Simulation methodology
In ASPEN Plus V8.8 the components are classified into several major classes. In this work, only conventional components are involved. In the global settings of the simulation, the stream class is set to CONVEN. The components added to the simulation were Glycerol, carbon monoxide, carbon dioxide, methane, hydrogen, and water. The property method selected for the simulation was UNIFAC. The UNIQUAC Functional Group Activity Coefficient model (UNIFAC) is an improvement on the UNIQUAC model as it incorporates the functional group of the chemical species as one of the basis for determining activity coefficients. UNIFAC is quite accurate for low pressure systems (1–5 bars). Also, UNIFAC in ASPEN plus uses Redlich-Kwong equation of state which is also accurate for calculations on chemical systems such as those involving synthesis gas. The reformer was modelled in the simulation by an equilibrium reactor (REQUIL). The equations of reaction stated earlier were specified in the reactor block. Reduction of product vapour temperature to induce condensation of water vapour is modelled by a heater set to ambient conditions.
2.3 Simulation environment
The design of the simulation was done specifically with the aim of studying the interaction of factors for the glycerol steam reforming process. Hence, a simple approach/design will be utilised focusing mainly on the REQUIL block. The flow diagram of the process is given in Figure 1.
Glycerol at ambient conditions and steam at 150 0C was sent into the reformer at a specific molar ratio. The steam reforming reactions take place at set temperature and pressure of the equilibrium reactor. The Steam to glycerol ratio (STGR), reactor temperature, and reactor pressure can be easily varied from the simulation. The stream S1 does not have any flowrate as the reaction occurs entirely in the vapour phase. However, it is placed there as ASPEN Plus always require the specification of a liquid stream for an equilibrium reactor. The products vapour is condensed back to ambient conditions to remove any excess water from the gas stream and then to consequently obtain synthesis gas.
2.4 Optimisation studies
Central Composite Design (CCD) was used to design the experiments for the study of factor interactions and effects and the determination of optimum parameters for the steam reforming of glycerol. Design-Expert 10.0 was utilised for this purpose. In order for more convenient report, notations in Table 1 were given to the factors/independent variables.
The lower and upper limits of the independent variables in the experimental plan were fixed as presented in Table 2.
3 Results and discussion
3.1 Model results and validation
The result for Hydrogen gas obtained from the system with respect to temperature and pressure variations are plotted below and validated with experimental work  to demonstrate the relevance and adequacy of the model.
Figure 2 shows a remarkable consistency between simulation model results and experimental results by . For the results shown above, ,utilised a steam to gas ratio of 9:1 while varying temperature from 600 K to 1000 K. The operation was done at two pressures; 1 and 5 atm. The unit on the y-axis of the plot is the number of moles of Hydrogen gas produced from the process per mole of glycerine (as presented by ). From the simulation, the corresponding results were easily computed from molar flowrates of the glycerine in the feed stream in relation to the Hydrogen gas in the product stream. This value is a ratio hence it is independent of the magnitudes of the feed stream. The simulation results for the exact same conditions are placed alongside. The model can be said to be valid in representing the steam reforming of glycerol to form synthesis gas. At 1 atm, simulation model result has a correlation coefficient of 1.00 with experimental work while at 5 atm, correlation is at 0.995. The simulation result is still consistent with those obtained by .
The simulation results were furthermore examined by the products distribution in the synthesis gas. The results of ,was used to validate the model. The results above by ,were obtained at steam/carbon ratio of 1.35, a glycerol feed-rate of 0.0065 ml/min and 1 atm pressure. The results from the simulation are at similar S/C ratio, temperature and pressure conditions albeit at a feed-rate of 100 mol/hr. Considering that mole fractions are being compared the feed-rates due not bear any significance on the results. The simulation model shows considerable similarity with the experimental result. There is a fairly good correlation of the yields of all products with experiments except at temperatures below 700 0C. We can say with a good measure of confidence that the model is in good agreement with experimental results for the steam reforming of glycerol.
The response of the Central Composite Design (CCD) used to design the experiments for the study of factor interactions and effects and the determination of optimum parameters for the steam reforming of glycerol is given in Table 3. Design-Expert 10.0 was utilised for the analysis of factors and their interactions. Some of the values of factors 1, 2 and 3 in Table 3 are outside the range of the upper and lower limits set in Table 2. This is because Central Composite Design is a statistical tool for optimisation. Each experimental run is either a center point or an axial point in the design matrix. Typically, axial points will sometimes fall outside the factor limits. There are no negative temperatures (factor 1) in the experimental runs. The negative pressure in factor 2 (run 3) was taken as positive on the simulation as it is considered that the magnitude of the pressure only denotes effective pressure. Also, ASPEN Plus does not facilitate the input of negative pressure in the simulation. For Factor 3 (row 17), we have a negative steam-to-gas-ratio (STGR). This is obviously not physically possible as the value is simply a ratio of the two reacting feedstock, hence the positive value will be utilised and the model is validated in Figure 3.
3.3 Selectivity to pressure
Little variability was given for pressure because previous research [3, 12] has reliably informed that atmospheric pressure is the optimum pressure level for glycerol steam reforming. The simulation was run at different pressures (Figure 4) with an STGR of 10:1 to reinforce this. It can be easily deduced that optimum pressure is somewhere near atmospheric pressure. In his study, , presented a similar behaviour for the H2 selectivity with pressure.
Optimisation goals were to maximise H2 selectivity and minimise methanation. Apart from the atmospheric pressure that was set to 1 atm, the other factors were ranges as similar to the limits in the initial design of experiments. The results of optimisation showed that maximum yield of H2 and minimal methanation can be obtained at a temperature of 900 0C, an STGR of 15.75 mol/mol and at atmospheric pressure. The desirability of the result was 1.000. Design expert also gave the results from the optimum factors as H2 = 67.77 %, CO = 9.38 %, CO2 = 22.8 % and CH4 = 0 %. However, for higher accuracy the optimum parameters were taken to the ASPEN Plus simulation to the obtain optimum results as H2 = 66.72 %, CO = 11.76 %, CO2 = 21.52 % and CH4 = 0 %.
3.4 Sensitivity analysis
The simulation temperature sensitivity (Figure 5) is in good agreement with experimental results [3, 4, 12, 24]. CO and CO2 are considered impurities as they do not compete with Hydrogen gas production. However, CO is more desired in the gas stream because it is combustible. However, they are good indicators of the extent of steam reforming reaction and water-gas shift reaction respectively. Hydrogen production is favoured at higher temperatures because of the endothermic nature of the reaction. Methanation is favoured at lower temperatures. Temperature rise will therefore lead to a drop in methanation and an increase in temperature until the optimal threshold is exceeded.
3.5 Factor effects and interactions for H2 production
The Different factors have combinatory effects on the Hydrogen gas selectivity of the process. Taking a look at the Figure 6(a–c) a lot can be understood about the process.
In the domain of all pressures, Hydrogen gas selectivity generally rises with temperature until the optimum temperature is exceeded before it starts to fall. Higher STGR favours Hydrogen gas production and this effect is more pronounced at lower temperature than at higher temperatures. Irrespective of pressure, STGR increase favours Hydrogen selectivity.
3.6 Factor effects and interactions for CO production
As presented in the system of equations utilised in developing the model, CO is produced in glycerol decomposition and used up in the water gas shift and methanation reactions. Hence, the CO in the system gives an indication of the extent to which the combination of factors favour glycerol decomposition over the other reactions.
Temperature increase favours CO production and this seems fairly independent of slight pressure changes. CO production falls with increasing STGR and this trend is similar at all temperatures. CO production also falls with increasing STGR for slight pressure change. Generally, we notice a fall in CO production due to a rise in STGR. This is due to the fact that more steam in the system will favour the water gas shift reaction which is a CO consuming process.
3.7 Factor effects and interactions CO2 production
CO2 is produced as a by-product of the water gas shift reaction hence the CO2 production is a very good indicator of how these factors interact to affect the reaction.
From Figure 8(a–b), we see that CO2 production falls with increasing temperature and this is quite independent of pressure. This temperature behaviour is due to the water gas shift reaction being exothermic. Heat releasing reactions will always proceed faster at lower temperatures and vice-versa. A retardation of the water gas shift reaction (a CO consuming and CO2 producing reaction) will favour a higher CO content (Figure 7) and lower CO2 content in the product. There is a steep climb in CO2 with increasing STGR and this is more pronounced at lower temperatures. Also, CO2 production is also favoured at higher STGR for all slight pressure changes. This gives a general indication of the importance of excess steam in favouring an increase in CO2 (and consequently hydrogen). Excess steam will always favour the water gas shift reaction. In economically optimising the process, it will be more favourable to reduce steam input (STGR) to reduce the CO2 formation albeit keeping an eye on the Hydrogen gas production (as steam also favours Hydrogen gas production too)
3.8 Factor effects and interactions CH4 production
Methane is formed from methanation reaction therefore it’s production will be an easy indication of the reaction extent (Figure 9).
There is a steep rise in methane yield at lower temperatures and higher pressures. Methanation reaction is an exothermic reaction hence will be favoured at lower temperatures. Methane production also climbs at lower STGR and temperature. Higher pressures will always favour methane production and this is fairly independent of STGR. Generally, lower temperatures, lower STGR and higher pressures favour methane production in the process.
The simulation model was prepared using ASPEN Plus v8.8 and was summarily validated with experimental results. Results were obtained according to the optimisation plan developed using central composite design (CCD). The variables (and range) were temperature ( 7000C – 1100 0C), Pressure (0.1 atm – 1.9 atm) and steam to glycerol ratio (1 mol/mol – 12 mol/mol). The results of optimisation showed that maximum yield of H2 and minimal methanation can be obtained at a temperature of 900 0C, an STGR of 15.75 mol/mol and at atmospheric pressure. The desirability of the result was 1.000.
The optimum parameters were taken to the ASPEN Plus simulation to the obtain optimum results as H2 = 66.72 %, CO = 11.76 %, CO2 = 21.52 % and CH4 = 0 %. Sensitivity analysis was carried out to show that Hydrogen production is favoured at higher temperatures and methanation is favoured at lower temperatures. A critical investigation of the factor effects and interactions for each product in the synthesis gas (dry basis) was also carried out.
Dou B, Song Y, Wang C, Chen H, Xu Y. Hydrogen production from catalytic steam reforming of biodiesel byproduct glycerol: issues and challenges. Renewable and Sustainable Energy Reviews. 2014;30:950–60. Web of ScienceCrossrefGoogle Scholar
Adhikari S, Fernando S, Haryanto A. A comparative thermodynamic and experimental analysis on hydrogen production by steam reforming of glycerin. Energy & Fuels. 2007a;21:2306–10. Web of ScienceCrossrefGoogle Scholar
Adhikari S, Fernando SD, To SF, Bricka RM, Steele PH, Haryanto A. Conversion of glycerol to hydrogen via a steam reforming process over nickel catalysts. Energy & Fuels. 2008;22:1220–26. Web of ScienceCrossrefGoogle Scholar
Chattanathan SA, Adhikari S, Abdoulmoumine N. A review on current status of hydrogen production from bio-oil. Renewable and Sustainable Energy Reviews. 2012;16:2366–72. CrossrefWeb of ScienceGoogle Scholar
Fu P, Yi W, Li Z, Bai X, Zhang A, Li Y, et al. Investigation on hydrogen production by catalytic steam reforming of maize stalk fast pyrolysis bio-oil. Int J Hydrogen Energy. 2014;39:13962–71. CrossrefWeb of ScienceGoogle Scholar
Wang D, Czernik S, Montane D, Mann M, Chornet E. Biomass to hydrogen via fast pyrolysis and catalytic steam reforming of the pyrolysis oil or its fractions. Ind Eng Chem Res. 1997;36:1507–18. CrossrefGoogle Scholar
Hajjaji N, Chahbani A, Khila Z, Pons M-N. A comprehensive energy–exergy-based assessment and parametric study of a hydrogen production process using steam glycerol reforming. Energy. 2014;64:473–83. CrossrefWeb of ScienceGoogle Scholar
Adhikari S, Fernando S, Gwaltney SR, To SF, Bricka RM, Steele PH, et al. A thermodynamic analysis of hydrogen production by steam reforming of glycerol. Int J Hydrogen Energy. 2007;32:2875–80. Web of ScienceCrossrefGoogle Scholar
Chen H, Zhang T, Dou B, Dupont V, Williams P, Ghadiri M, et al. Thermodynamic analyses of adsorption-enhanced steam reforming of glycerol for hydrogen production. Int J Hydrogen Energy. 2009;34:7208–22. Web of ScienceCrossrefGoogle Scholar
Da Silva AL, Müller IL. Hydrogen production by sorption enhanced steam reforming of oxygenated hydrocarbons (ethanol, glycerol, n-butanol and methanol): thermodynamic modelling. Int J Hydrogen Energy. 2011;36:2057–75. CrossrefWeb of ScienceGoogle Scholar
Wang X, Wang N, Li M, Li S, Wang S, Ma X. Hydrogen production by glycerol steam reforming with in situ hydrogen separation: a thermodynamic investigation. Int J Hydrogen Energy. 2010;35:10252–56. CrossrefWeb of ScienceGoogle Scholar
Yang G, Yu H, Peng F, Wang H, Yang J, Xie D. Thermodynamic analysis of hydrogen generation via oxidative steam reforming of glycerol. Renewable Energy. 2011;36:2120–27. Web of ScienceCrossrefGoogle Scholar
Dou B, Dupont V, Rickett G, Blakeman N, Williams PT, Chen H, et al. Hydrogen production by sorption-enhanced steam reforming of glycerol. Bioresour Technol. 2009;100:3540–47. CrossrefPubMedWeb of ScienceGoogle Scholar
Adhikari S, Fernando SD, Haryanto A. Kinetics and reactor modeling of hydrogen production from glycerol via steam reforming process over Ni/CeO2 catalysts. Chem Eng Technol. 2009;32:541–47. CrossrefWeb of ScienceGoogle Scholar
Chiodo V, Freni S, Galvagno A, Mondello N, Frusteri F. Catalytic features of Rh and Ni supported catalysts in the steam reforming of glycerol to produce hydrogen. Appl Catalysis A: Gen. 2010;381:1–7. CrossrefWeb of ScienceGoogle Scholar
Nichele V, Signoretto M, Menegazzo F, Gallo A, Dal Santo V, Cruciani G, et al. Glycerol steam reforming for hydrogen production: design of Ni supported catalysts. Appl Catalysis B: Environ. 2012;111:225–32. Web of ScienceGoogle Scholar
Profeti LP, Ticianelli EA, Assaf EM. Production of hydrogen via steam reforming of biofuels on Ni/CeO2–al2O3 catalysts promoted by noble metals. Int J Hydrogen Energy. 2009;34:5049–60. Web of ScienceCrossrefGoogle Scholar
Zhang B, Tang X, Li Y, Xu Y, Shen W. Hydrogen production from steam reforming of ethanol and glycerol over ceria-supported metal catalysts. Int J Hydrogen Energy. 2007;32:2367–73. CrossrefWeb of ScienceGoogle Scholar
Dou B, Dupont V, Williams PT. Computational fluid dynamics simulation of gas− solid flow during steam reforming of glycerol in a fluidized bed reactor. Energy & Fuels. 2008;22:4102–08. CrossrefWeb of ScienceGoogle Scholar
Dou B, Song Y. A CFD approach on simulation of hydrogen production from steam reforming of glycerol in a fluidized bed reactor. Int J Hydrogen Energy. 2010;35:10271–84. Web of ScienceCrossrefGoogle Scholar
Licker DM. Dictionary of engineering. 2nd ed. Chicago: Mc-Graw Hill Publishers; 2003. Google Scholar
About the article
Published Online: 2018-09-01