Syngas mixtures

Renewable Energy 130 (2019) 495e509

Contents lists available at ScienceDirect

Renewable Energy journal homepage: www.elsevier.com/locate/renene

DFT and canonical ensemble investigations on the thermodynamic properties of Syngas and natural gas/Syngas mixtures Abel F.G. Neto a, b, *, Francisco C. Marques c, Adriana T. Amador a, d, Amanda D.S. Ferreira d, Antonio M.J.C. Neto a, b, d
, C. P. 479, 66075-110, Bel
Laboratory of Preparation and Computation of Nanomaterials (LPCN), Federal University of Para em, PA, Brazil
, 2626, 66.050-540, Bel
Post-graduation Program of Natural Resource Engineering of the Amazon e PRODERNA; ITEC, State University of Para em, PA, Brazil c ~o Paulo, Brazil Gleb Wataghin Institute of Physics, State University of Campinas, C. P. 13083-859, Sa d
, C. P. 479, 66075-110, Bel
Faculty of Physics e ICEN, Federal University of Para em, PA, Brazil a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 September 2017 Received in revised form 5 March 2018 Accepted 21 June 2018 Available online 22 June 2018

Density Functional Theory and canonical ensemble were used to investigate thermodynamic properties of Syngas and its mixture with natural gas. The following thermodynamic potentials were obtained: internal energy, enthalpy, entropy and Gibbs free energy for temperatures ranging from 0.5 K to 1500 K. It was observed that CO and H2 were the most stable Syngas components, possessing the ability to render Syngas less favorable to the temperature increase. Also, we verified that Syngas presents properties similar to an antiknock agent for natural gas, raising its resistance to temperature increases. Were determined the Poisson coefficients and Bulk modulus for Natural gas/Syngas mixtures and Shomate equation coefficients for some Syngas types, providing a more complete thermodynamic description for these gases. Additionally, thermodynamic potentials of combustion for Natural gas/Syngas mixtures were predicted, showing that this biofuel can reduce the calorific power of natural gas and makes its combustion less favorable due its antiknock behavior. However, a mixture with 30% of Syngas may be useful for natural gas combustion, since it present a calorific power between 73.41% and 79.49% of that of natural gas, which is a substantial fraction of energy released during combustion, showing good future prospects to the Natural gas/Syngas mixture to the renewable energy generation. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Syngas Natural gas DFT Thermodynamics Temperature effect Natural gas combustion

1. Introduction The combining of fossil fuels with biofuels is becoming common practice in the most diverse sectors of industry around the world, and principally in the automotive sector [1]. This is due to the nonrenewable character of fossil fuels, as well as their environmental drawbacks, like their emission of polluting gases into the atmosphere [1e4]. Some examples of these mixtures are ethanolgasoline [5e7], kerosene-biokerosene [8,9] and diesel-biodiesel blends [10,11]. However, another proposal for fuel which incorporates this kind of blend points to natural gas (NG) combined with a synthesis gas (Syngas), which is a biogas produced from biomass gasification [12e14], a technique that converts a solid biomass into a

* Corresponding author. Laboratory of Preparation and Computation of Nano
m, PA, Brazil. materials (LPCN), Federal University of Par
a, C. P. 479, 66075-110, Bele E-mail address: [email protected] (A.F.G. Neto). https://doi.org/10.1016/j.renene.2018.06.091 0960-1481/© 2018 Elsevier Ltd. All rights reserved.

combustible gas. Thus, since the composition of Syngas contains lighter molecules (H2, CO, CH4, …) [15e19], this biofuel has characteristics similar to NG. Consequently, recent studies have been developed which have aimed to use Syngas as a complement to the composition of other fuels, such as NG, in internal combustion engines [20e23]. Hagos et al. [20]compared the performance of Syngas and NG during the direct injection stage in spark ignition engines, noting that in this stage Syngas is capable of supporting higher cylinder pressures than NG, being more stable and presenting a greater rate of heat release during combustion. Johansson et al. [21] performed some tests with Syngas and NG in a system for chemical-looping combustion (CLC), in order to compare the conversion of these fuels due to combustion when interacting with oxygen-carriers. This system was operated successfully, presenting a good gas conversion rate. They observed a conversion around 99% for the Syngas and for the NG. With regard to Syngas, the final amounts of hydrogen and carbon monoxide

496

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

were 1.8% and 2%, while for NG the methane fraction was between 0.5 and 1.5%. Donohoe et al. [22] presented a study on the influence of steam dilution on the autoignition of some gases; as in, for example, the Syngas and natural gas mixtures under gas turbine-relevant conditions. Throughout the experiments, machine rapid compression was carried out for fuel/air mixtures. However, it was observed that significant changes in the thermal properties of the mixtures had an effect on their reactivity. Le Cong et al. [23] performed an experimental study on the oxidation of methane-based fuels in the presence of Syngas. From this investigation, a detailed chemical kinetic model was produced, yielding a good agreement with the empirical data, such as in the burning velocities. The greater stability of Syngas indicates that it has antiknock characteristics, which can be highly important for better engine performance [24]. In a review by Boehman and Le Corre [25] this importance was demonstrated in their investigation on the capacity of Syngas to raise the combustion temperature of some fuels, increasing their efficiency and optimizing their combustion [26]. Despite its great relevance on issues related to the use of biomasses for the generation of energy, until recently few studies have reported on a thermodynamic analysis of an NG/Syngas mixture. These results are to a large extent useful in contributing to a better understanding of the antiknock properties of Syngas, and of the other effects associated with the addition of Syngas to NG. Therefore, the present work aims to show a recommended DFT method, for predicting some of the thermodynamic properties of Syngas and the NG/Syngas blend as a function of any useful temperature and for their combustion properties. Additionally, these amounts correspond to properties calculated for several quantities of Syngas fractions, providing greater knowledge about the effect of Syngas when it is combined with NG, as for example, its ability to act as an antiknock agent for NG, which could be observed through this study. For this investigation, two models were adopted, the Density Functional Theory (DFT) [27,28], which is a sophisticated model based on quantum mechanics, and the canonical ensemble model of statistical thermodynamics [29]. 2. Methodology The simulations performed in this research describe the thermodynamic properties of nine kinds of Syngas from different biomasses and/or different gasification conditions [15e18]. The pressure and temperature conditions, under which the calculations were performed, are similar to those of the many mechanical processes involving fuels, such as the fuel injection stage in internal combustion engines. For instance, for Syngas and NG, the fuel injection process is generally performed at a temperature and an absolute pressure of around 600 K [30,31] and 1atm [31], respectively, so that the fuels, throughout this step, are at chemical equilibrium (i.e., there are no combustion reactions). Thus, calculations based on quantum mechanics and statistical thermodynamics were performed in order to obtain the following quantities for Syngas and the NG/Syngas mixtures: internal energy (U), enthalpy (H), Gibbs free energy (G) and entropy (S). These quantities were calculated for temperatures from 0.5 K to 1500 K. To obtain these thermodynamic quantities for the fuels, we first modeled each one of their major components. This modeling was, initially, given by a conformational search [32] of the molecules, where the first energy minimization for the molecular structures was performed using the classical method of Molecular Mechanics (MMþ) [33,34] through Hyperchem 7.5 software [35]. During these calculations, torsion and bond angles, along with lengths of atomic bonds, were randomly disturbed in order to verify, based on the

classical model, the molecular conformation with lower energy for each major component of each fuel. After the conformational searches, a second energy minimization process was performed for each molecule, now utilizing Gaussian 09 W software [36] for simulations based on the DFT model [27,28]. In these simulations we used the B3LYP functional [37,38], along with the sophisticated basis set: 6e311þþg(d,p) [39], which is recommended for NG thermodynamic DFT analysis [40]. Geometry optimization provided the molecular conformations with lower energy for each component of both Syngas and NG fuels. The third stage of simulations consisted of frequency calculations (IR and Raman) [41,42] of the chemical species, where the same functional and basis set considered previously in the geometry optimization was used. From this calculation the thermodynamic properties presented in this study were obtained. No imaginary frequencies were observed for the molecular structures, which mean that the geometries were well optimized. Thus, for these calculations, each molecular system was simulated through the canonical ensemble [29], where partition functions were obtained for the following motions: translational, rotational, vibrational and electronic, and the monomolecular polyatomic gas models ware considered. These functions are given by:

qt ¼

3
2pmkB T 2 Vðtranslational motionÞ; h2

(1)

where m, kB, h end V are, respectively, the mass particle, Boltzmann and Planck constants, and the molecular volume.

qr ¼

1




sr

 8p2 IkB T ðrotational motionÞ; h2

(2)

I being the moment of inertia of the molecular structures, and sr corresponding to their symmetry number. Q

qy ¼

Y e 2Ty;k k

Qy;k

1  e

ðvibrational motionÞ;

(3)

T

with Qv,k ¼ hvk/kB, with n being the vibrational frequency associated with the k-th normal mode of vibration. ε

k 0T

qe ¼ u0 e

B

ε

k 1T

þ u1 e

B

e

k 2T

þ u2 e

B

þ :::ðelectronic contributionÞ; (4)

where u is the degeneracy of energy levels and 3n corresponds to the energy of the n-th level. Thus, once the energy of the first excited state is much greater than kBT, all excited states are inaccessible, regardless of temperature. However, the internal energy and entropy of the system are obtained, respectively, from Eq. (5) and (6):

U ¼ NkB T 2

  vlnðqt qr qy qe Þ ; vT V 

S ¼ R lnðqt qr qy qe Þ þ RT

vlnðqt qr qy qe Þ vT

(5)  ;

(6)

V

where N is the number of molecules and R corresponds to the universal gas constant. In order to obtain the thermodynamic properties of Syngas and NG from the properties of their components, we calculated the quantities of each of the chemical species in the fuel compositions, such that the properties of mixtures were obtained from weighted averages, where we take into account that both fuels (Syngas and

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

NG) are approximately ideal gases [43]. The weighted averages were calculated according to Table 1, which shows the composition, as well as the percentage fraction of the major components of 9 kinds of Syngas, each referring to different biomasses and/or gasification conditions from industrial, agriculture and forestry wastes [15e18]. Thus, according to Table 1, these gases are divided into four types: Type A, composed of the synthesis gases from Bagasse (sugarcane press residue), Pine Sawdust, Poplar Sawdust, Cotton Stalks, Almond Shells and Olive Wastes, which were presented by Tomasi et al. [15]; Type B, composed of gas originating from Poplar Wood, in a gasification process described in the work of Michiel and Tijmensen [16]; Type C, which corresponds to Syngas produced by the Eucalyptus Sawdust, presented by Corujo et al. [17]; and lastly, Type D, referring to Syngas from Wood Shavings, presented by Patil et al. [18]. The composition of each Syngas investigated is shown in Table 1, as well as some of the parameters and conditions of gasification used in the production of these gases. Additionally, Table 1 presents the composition of Liquefied and Compressed Natural Gas (L-CNG), which was considered in a work by Karavalakis and his coworkers [44]. Considering that during the injection of fuel into combustion chambers the temperature change suffered by gases is in the range of 298.15 Ke600 K [29,31,45,46], changes in their thermodynamic properties were estimated for this increase of temperature. Estimated variations were calculated for the following quantities: in600K ternal energy ðDU600K 298:15K Þ, enthalpy ðDH298:15K Þ; Gibbs free energy 600K ðDG600K 298:15K Þ and entropy. ðDS298:15K Þ: It is important to note that these changes are calculated ac600K ¼ Xð600KÞ cording to the following equation: DX298:15K Xð298:15KÞ (7). These results may be relevant to a better understanding about the influence of the chemical composition of fuels on their thermodynamic behavior, as well as on their heating facility. These same quantities were calculated for the NG þ Syngas mixtures in various proportions, in order to investigate the effect of incorporating Syngas into NG. All these methodology steps were based on Neto et al. [47]. In addition, curves of specific heat at a constant pressure (CP), and H and S as temperature functions, were interpolated for each Syngas investigated, where we used Shomate equations [48] to

497

perform this step. Thus, the coefficients of the equations were obtained for each Syngas, enabling a more complete description of their thermodynamic properties, since these equations can describe the quantities with desirable accuracy for any temperature between 200 K and 1500 K. Then the following parameters were estimated for each Syngas: the Poisson constant (g), given by g ¼ CP/CV (8), and the Bulk modulus (b), given by Eq. (9):

b ¼ V

 vP : vV




(9)

Once the NG and Syngas can be described with good approximation by the equation of state for the ideal gas, we may write the Bulk modulus according to Eq. (10):

b ¼ 101:325kPa 

CP : CV

(10)

The Bulk modulus is an important quantity, which can give information about the degrees of freedom for gases, as well as about their “hardness” e this being a highly relevant physical quantity in the modeling of fuels [49]. In order to predict the thermodynamic potentials of combustion for various NG/Syngas mixtures, some calculations were performed under standard conditions for temperature and pressure, in order to obtain the following combustion properties for Syngas: internal energy ðDc U q Þ, enthalpy ðDc H q Þ; entropy ðDc Sq Þ and Gibbs free

energy ðDc Gq Þ: In order to estimate these quantities, the composite method was used: CBS-QB3 [40], which presents a good performance in the frequency (IR and Raman) calculation for the major components of NG, as shown by Neto et al. [40]. Thus, we used the same method to obtain the thermodynamic potentials of the two main components of Syngas (H2, CO and CH4), where the following combustion equations were considered:

1 H2ðgÞ þ O2ðgÞ /H2 OðlÞ ; 2

(11.a)

Table 1 Major components of the fuels investigated in this work. Nine types of Syngas were considered, obtained from different sources and gasification processes, and one type of natural gas, which corresponds to a liquefied natural gas composition [15e18,44]. Characteristics

Type Bagasse Pine Sawdust

Poplar Sawdust

Cotton Stalks

Almond Shells

Olive Wastes

Poplar Wood

Type A [15] Moisture (%) Ash (%) Gasification agent Conversion temperature (K) Conversion pressure (bar)

Eucalyptus Sawdust

Wood Shavings

Natural Gas (L-CNG) [44]

Type B [16] Type C [17]

Type D [18]

7.1 0.9 Steam 1173

9.4 0.9 Steam 1173

10.0 3.9 Steam 781

7.9 4.5 Steam 774

11.50 2.92 Steam 830

13.03 03.57 Steam 806

15 n.a. Steam 1241

n.a. 4.6 Steam 1173

5.30 0.43 Air 1139

n.a n.a e e

1

n.a

n.a

n.a

n.a

n.a

20.3

n.a.

n.a.

n.a

49.49 2.99 43.33 2.88 0.13 1.18 e e

50.19 2.88 43.00 2.69 0.06 1.18 e e

49.46 3,55 42.02 3.50 0.13 1.34 e e

49.35 4.33 39.77 4.31 0.66 1.58 e e

49.44 1.93 45.48 1.74 0.56 0.85 e e

50.85 2.57 42.38 2.28 0.80 1.12 e e

31.7 0 15.85 35.9 0.8 11.6 e e

46.2 0 33.2 16.1 0 4.4 e e

10.9 0 22.2 11.5 50.9 4.5 e e

e e e e e 98.40 1.20 0.30

Components H2 H2O CO CO2 N2 CH4 C2H6 C3H8 Note: n.a. ¼ not available.

498

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

3. Results and discussions

1 COðgÞ þ O2ðgÞ /CO2ðgÞ : 2

CH4ðgÞ þ 2O2ðgÞ /CO2ðgÞ þ 2H2 OðlÞ :

(11.b)

(11.c)

Note that in Eqs. (11.a) and (11.c) the combustion reaction has water at a liquid-phase as a product, which H2O's condensation thermodynamic properties were obtained from Armarego and Chai [50]. Once the combustion properties of H2, CO and CH4 were calculated, we used the percentage fractions of these components (according to the composition of each Syngas shown in Table 1) to predict the combustion potentials of synthesis gases. Finally, the potentials of each Syngas were combined with those of NG [40] in order to estimate the properties of the NG/Syngas mixtures for various proportions of Syngas, where we considered the weighted average of the potential of each fuel.

3.1. Thermodynamic properties of syngas components In order to provide a better understanding on the properties (to be discussed in the subsequent sections) of the different types of Syngas considered in this study, Fig. 1a-1 d show some thermodynamic properties as a function of temperature for the major components of the different Syngas types, as shown in Table 1. At this stage, the following quantities were calculated: Enthalpy, CP, Entropy and Gibbs free energy, so that from these calculations, it can be verified that the components' enthalpy values are arranged according to the following relation: CH4 > H2O > CO2 > H2 > N2 z CO. This relationship indicates that CH4 is the highest energy component; however, this feature does not make methane particularly relevant to the heating of Syngas, as it usually appears in low amounts in this biofuel mixture. However, when analyzing afresh the enthalpy of H2 and CO molecules, which are generally characterized as the main components of Syngas, it can be observed that

Fig. 1. Thermodynamic potentials e (a) internal energy, (b) enthalpy, (c) Entropy and (d) Gibbs free energy for the major components of the each kind of Syngas considered in this study. The properties are corresponding to chemical equilibrium for various temperatures based on the canonical ensemble.

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

hydrogen gas is a little more energetic than carbon monoxide, although both still have significantly less energy than do other components, such as CH4, for example. Another important point regarding the components of Syngas is the rate of enthalpy increase for these components due to the rise in temperature, which corresponds to CP. The CP values shown are arranged according to the following relationship: CH4 > CO2 > H2O > CO z N2 > H2, indicating that during the heating of Syngas, carbon monoxide presents greater energy absorption than hydrogen gas. However, in addition to the fact that CO absorbs more energy than H2 during the heating process, when analyzing the components' entropy values, these are found to be e due also to the temperature rise e in accordance with the following relation: CO2 z CH4 > CO z H2O > N2 > H2. In this sequence, it may be observed that H2 has the smallest entropy values among all the considered components, including CO. It is also important to note that with regard to CO2 and CH4, which have slightly more complex molecular structures, it is verified that a rise in the temperature of these molecules promotes an increase in their structural rotation and vibrational modes. Conversely, the molecules of N2 and H2 are those with the lowest values for entropy, which can be respectively justified by the three bonds that make up N2, and by the low molecular mass of H2, which lead to a reduction in their degrees of freedom. An additional relevant point is that with the increase in these components' temperature, two changes occur in the sequence with which the curves are arranged in Fig. 1 c. First, for the temperature of 600 K, the H2O molecule, which presented slightly less entropy than CO, shows higher values. Similarly, for the 1100 K the CH4 structure will possess slightly more entropy than the CO2 molecule, reversing the sequence with which its curves are arranged. Finally, Fig. 1 d brings Gibbs free energy values to the components of Syngas. It can be observed that CH4 is the component which has the highest slope, presenting as the most favorable component to heating. Inversely, H2 is the component the curve of which has the smallest slope, and therefore, is the least favorable to an increase in temperature. The CO, N2 and CO2, in turn, are the components that had the lowest Gibbs free energy values, among the molecules analyzed; the DG value for these components, however, are approximate and relatively inexpressive, indicating that their molecular structures are less favorable to a temperature rise. Thus, the main components of Syngas (H2 and CO) show that H2 is more resistant than CO throughout the heating of this gas and can act as an important component in the thermal control of the same. 3.2. Evaluations of a number of DFT methods for NG thermodynamic prediction As previously mentioned in the methodology, calculations, based on DFT and canonical ensemble, and were performed for the thermodynamic prediction of Syngas and its mixtures with NG. In this way, initially, it was tried to verify the method of the DFT more suitable for such investigation of NG and Syngas, taking into account the study realized by Neto et al. [40]. The CP values of NG and, later, for Syngas at different temperatures in the range of 0.5 Ke1500 K were calculated for six different methods. The methods correspond to: B3lyp/6e311þþg(d,p), B3lyp/6-31 g(d) and the composed methods CBS-QB3, G3 and G4. Besides these methods, the values of CP corresponding to the mean G3/G4 were also recorded, which was recommended by Neto et al. [40] to predict the thermodynamic properties of hydrocarbons. Fig. 2 a presents the CP values for the different methods already mentioned for the NG and Syngas (of the Poplar Wood type), and at the same time compares them with the values obtained from the

499

data provided in NIST. It can be observed that all six methods present curves similar to NIST (both for NG and Syngas), however the B3lyp/6e311þþg(d,p) and CBS-QB3 methods are the ones that presented the best results and better computational cost. In order to analyze the accuracy of the obtained values, the absolute errors of each method were calculated in relation to the results generated from the NIST. These errors are presented in Fig. 2b to 2 c, where for the NG it can be seen that all methods had relatively low errors, except G3 and, consequently, the G3/G4 mean. On the hot hand for the Syngas the G3 was a best method. However, taking in account both (accuracy and computational cost) it is possible to suggest that the results obtained by the B3lyp/6e311þþg(d,p) and CBS-QB3 methods are more advisable. It is important to note that although these methods are almost equivalent in predicting CP of NG, for the temperatures between 298.15 Ke600 K the method B3lyp/ 6e311þþg(d,p) still presents slightly better precision in comparison to CBS-QB3, which only becomes the most suitable method for temperatures higher than 700 K. Thus, the methods B3lyp/ 6e311þþg(d,p) may be suggested as the most recommended, among the methods considered, for the thermodynamic prediction of NG. Since the properties of NG mixtures with Syngas will also be presented, this method was also used to perform Syngas predictions. In addition to the analysis of the performance of each method, it can also be investigated, from Fig. 2 a, that NG has CP significantly higher than that of each of the types of Syngas considered in this study. Only Syngas from Poplar Wood has slightly higher specific heat. 3.3. Thermodynamic properties of syngas The internal energy, enthalpy, Gibbs free energy and entropy of the gases produced from the biomasses and/or gasification processes, previously shown in Table 1, were calculated as functions of temperature (Figs. 3e6a). In Figs. 3e6 b the thermodynamic changes for the gases are presented in a range of 298.15e600 K, which were extracted from the results for the panels (a), corresponding to the region of temperature highlighted in green. The same panels also show the changes in thermodynamic potentials for NG considered in this study, in order to compare its properties to Syngas. In Fig. 3 a, we observed an increase in the gases internal energy proportionally to temperature. This behavior is mainly a consequence of an increase in the translational and rotational energies of gas molecules, which for NG is around six times higher than for Syngas, as can be seen in their curves. Vibrational movement of the molecules also contributed to the increase in U, but was not as significant as the other two contributions. Fig. 3 a indicates that for all simulation temperatures (0.5 Ke1500 K) the values obtained for the U of each gas are arranged as follows: Almond Shells < Olive Wastes < Pine Sawdust z Bagasse < Poplar Sawdust < Cotton Stalks < Wood Shavings < Eucalyptus < Poplar Wood < NG. This sequence also applies to theDU600K 298:15K values (Fig. 3b), however it may be verified that Poplar Wood has a larger internal energy variation since it can absorb energy faster than others molecules. This may be related to its richer concentration of CH4 and CO2 which are molecular components with higher degrees of freedom. Despite this, in the other gases, it can be seen that their

DU600K 298:15K values (due to the temperature increase under constant pressure)

are

proportional

to

their

own

internal

energy.

ðDU600K 298:15K aUÞ: As previously mentioned, an important analysis to be made with respect to Syngas is to relate its composition with its U andDU600K 298:15K values. Thus, taking the composition of Syngas into

500

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

Fig. 2. a) Specific heat at constant pressure for the natural gas (L-CNG) and Syngas (Poplar Wood) predicted by different DFT methods. These results are compared to the values generated from the NIST data. In addition, the accuracy of each DFT method in this prediction was verified from the absolute error calculation for b) Natural Gas and for c) Syngas.

Fig. 3. (a) Internal energy for each Syngas and natural gas investigated in this study. This quantity corresponds to the gas at chemical equilibrium for various temperatures based on the canonical ensemble; (b) internal energy variation of the gases due to the temperature change from 298.15 K up to 600 K under a constant pressure of 1atm.

account, it is possible to observe two main points. Firstly, for Syngas of the Poplar Wood type, it was seen that it presented more internal energy for the whole considered temperature range when compared to the other types of Syngas, but it was lower than for NG. However, this same Syngas is the variety with the highest energy absorption due to the same temperature variation

(298.15e600 K). This characteristic may be related to the high concentration of CO2 (35.9%) and CH4 (11.6%) in it composition, indicating that Syngas produced by gasification methods which generate many methane and combustion products, i.e. CO2, increases its specific heat at a constant volume (described by the following equation: CV ¼ ðvU=vTÞV (12)). It is also important to note

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

501

Fig. 4. (a) Enthalpy for each Syngas and natural gas investigated in this study; (b) enthalpy variation of gases due to the temperature change from 298.15 K up to 600 K under a constant pressure of 1atm.

that from an idealized point view, the Poplar Wood gas arises from a poor instance of gasification, since there is a large amount of CH4 and CO2 in its composition. In addition, it can be observed that for the six gases produced through the same gasification conditions, in the gas obtained from Cotton Stalks, which presents the lowest CO (39.77%) fraction and a slightly higher amount of H2O (4.33%) and CO2 (4.31%), the U values were the largest for all simulated temperatures, as well as their

Fig. 5. Entropy for each Syngas and for the natural gas at chemical equilibrium for various temperatures based on the canonical ensemble. The inset shows the entropy variation of gases due to the temperature change from 298.15 K up to 600 K under a constant pressure of 1atm.

variation, being between 298.15 K and 600 K ðDU600K 298:15K ¼ 6:60kJ=molÞ. These results indicate that CO (which CV, calculated by B3LYP at 298.15 K, is 20.83 J/mol.K) even at a lower quantity, is a component capable of increasing the CV of Syngas, as well as the H2O (which CV, calculated by B3LYP for 298.15 K, is 25.16 J/mol.K) and CO2. In addition, in analyzing the Syngas obtained from Almond Shells, which has the richest composition of CO (45.48%), and simultaneously, contains slightly less H2O (1.93%) and CO2 (1.74%), we observed that its U values were the lowest at all simulated temperatures, as well as having the lowest energy variation between 298.15 K and 600 K ðDU600K 298:15K ¼ 6:43kJ=molÞ. Therefore, with these results we conclude that H2O and CO2 tend to increase

Fig. 6. Gibbs free energy for each Syngas and the natural gas at chemical equilibrium for various temperatures based on the canonical ensemble. The inset shows details for the Gibbs free energy variation of gases due to the temperature change from 298.15 K up to 600 K. All values were calculated under a constant pressure of 1atm.

502

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

the CV of Syngas, as does CO, albeit to a lesser degree. Fig. 4 a shows the enthalpy of each Syngas and the NG as a function of temperature. This quantity corresponds to the gas at chemical equilibrium for various temperatures based on the canonical ensemble. The enthalpies of the gases follow the same sequence previously shown for their internal energy (Fig. 3a): Almond Shells < Olive Wastes < Pine Sawdust z Bagasse < Poplar Sawdust < Cotton Stalks < Wood Shavings < Eucalyptus < Poplar Wood < NG. Thus, the Syngas that comes from Poplar Wood had the highest enthalpy values for all the temperatures, but substantially lower than the enthalpy change for NG, while the gas obtained from Almond Shells presented the lowest enthalpy values. Fig. 4 b shows the enthalpy change ðDH600K 298:15K Þ for each Syngas due to heating from 298.15 K to 600 K. One can observe that the variation in enthalpy of the gases is proportional to their own enthalpy. Such result is similar to that which was observed for U andDU600K 298:15K ; and also for the gas of the Poplar Wood type presenting greater variation than the other types of Syngas, such that the enthalpy values versus temperature of this first Syngas form a sharper curve compared to the other kinds of Syngas. This result is also associated with the high concentration of CO2 and CH4 in their compositions. Thus, Syngas from Poplar Wood requires a higher enthalpy variation to present a temperature change. However, for the six types of Syngas of Type A (see Table 1), Cotton Stalks needed the highest enthalpy variation ðDH600K 298:15K ¼ 9:01kJ=molÞ while Syngas obtained from Almond Shells presented the lowest enthalpy variation ðDH600K 298:15K ¼ 8:84kJ=molÞ. Since the thermodynamic property calculations for the gases are performed at a constant pressure, DH600K 298:15K represents the heat absorbed by Syngas due to its temperature increase. Thus, it is noted that the Syngas of the Poplar Wood type is that which requires the most amount of heat to increase its temperature from 298.15 K to 600 K. For the six gases of Type A, Syngas produced from Cotton Stalks needs a greater amount of heat for its temperature to increase, while the Syngas from Almond Shells requires the least amount of heat to suffer the same change in temperature. These results imply that the heat required to increase the temperature of Syngas increases together with the CO, CO2 and CH4 fractions in its mixture and reduces in line with the H2 fraction, as shown in Fig. 2 b, where the enthalpy change of H2 is lower than that of the remaining components. The entropy of the gases, in Table 1, was also calculated for several temperatures (Fig. 5) in order to supplement the analysis previously presented about the internal energy and enthalpy of the gases. This analysis allows for a more complete view regarding the effect of Syngas and NG composition on the ease of a rise in temperature up to combustion. Therefore, for the gases of Type A, it is notable that there were no significant differences between their entropy values for the simulated temperatures. Nonetheless the gases from Poplar Wood, Wood Shavings and Eucalyptus, and even NG, presented the highest entropy for the entire temperature range, which may be related to the rising CO and CO2 fraction of their compositions. It is important to note that for the gas from Wood Shavings, entropy values are higher, as well as its N2 fraction, this molecule presenting a stable structure notwithstanding. It occurs because for temperatures around 600 K, the rotational contribution for entropy increases considerably. However, though theDS600K 298:15K of each Syngas is proportional 600K toDU600K 298:15K and DH298:15K ; it is possible to observe, correlating Fig. 4 b and the entropy changes in Fig. 5, that the entropic energy variations are always larger than the heat absorbed; that is, 600K DH600K 298:15K < DðSTÞ298:15K . Therefore, taking into account that Gibbs

free energy (G) is given by the equation: G ¼ HeST (13), it can be seen that an increase in temperature is a favorable process for Syngas e that is, DG600K 298:15K < 0. Fig. 6 presents the G values as functions of temperature for each Syngas and for the NG investigated. Similarly, added to what had already been done with regard to the other thermodynamic potentials (U, H and S), we also estimated the values ofDG600K 298:15K . This quantity allows the ease of gas heating to be estimated, considering the heat absorbed and the entropic energy variation. From theDG600K 298:15K value it is possible to verify that Syngas presents most “resistance” to an increase in temperature in comparison with NG. It is also a good parameter to probe the effect of Syngas composition. When analyzing the results of the G for the gases in Table 1, it may be observed that for temperatures at around 800 K, every kind of Syngas, with the exception of that from Wood Shavings, presents approximate G andDG600K 298:15K values, both being equally favorable to heating. However, for other temperatures, the Poplar Wood gas presents slightly greatest variations, showing in their results more accentuated curves as functions of temperature. This result is associated with the larger CH4 concentration in their compositions, since methane causes Syngas to be more favorable to heating, which is most notable when these curves are compared to the NG curve. It is important to note that for the six gases of Type A no significant differences were observed, which can be seen in Fig. 6. However, Fig. 6 b highlights the small differences between their values ofDG600K 298:15K for each Syngas. It is notable that the Syngas from Cotton Stalks presents the highest change of G ðDG600K 298:15K ¼  52:52kJ=molÞ, indicating that heating with an increasing temperature is the most favorable; while Syngas produced from the Almond Shells presents the lowest change of G ðDG600K 298:15K ¼  52:08kJ=molÞ, and consequently, a less favorable temperature increase. Therefore, these results indicate that the Syngas from Almond Shells, the composition of which is rich in CO and H2, contains the least amount of H2O and CO2 (products of combustion), and has a lower ease of increase in temperature. Thus, the amount of CO and H2, especially hydrogen gas, tends to make Syngas less favorable to temperature increases in comparison with the products of their burning, like H2O and CO2, which have opposite effects. Thus, in the kinds of Syngas produced by gasification techniques where heating is not fully controlled, their composition presents CO2 and H2O as major components, which not only contributes to the calorific power of the mixture, but also renders this gas less favorable to a temperature increase. Therefore, CO and especially H2, confer an improvement to the thermal resistance of Syngas, making it more stable during the heating process. 3.4. CP/CV ratio, bulk modulus and shomate equations From the following thermodynamic equation: CP ¼ (vH/vT)P (14), we utilized a numerical differentiation to estimate the CP of Syngas. Conversely, the CV values for Syngas components were calculated during the frequency calculations, thus the CV for each Syngas was estimated through a weighted average. Therefore, the specific heat ratio (g ¼ CP/CV), which supplies information about the changes to degrees of freedom of Syngas molecular components, was estimated at various temperatures (Fig. 7). The Bulk modulus, which definition was previously shown in Eq. 10, is also present in the same Figure. This quantity supplies information about material hardness, which can be very important for the understanding of the burning, storage, injection and conduction of

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

. CP ½J=mol:K ¼ A þ B*t þ C*t 2 þ D*t 3 þ E t 2 ;

503

(15.a)

. . . Ho ½kJ=mol ¼ A*t þ B*t 2 2 þ C*t 3 3 þ D*t 4 4  E=t þ F  H; (15.b) . . So ½J=mol:K ¼ A*lnðtÞ þ B*t þ C*t 2 2 þ D*t 3 3  .  E 2*t 2 þ G ;

Fig. 7. Specific heat ratio (Poisson coefficient) and Bulk modulus for each Syngas investigated at chemical equilibrium for various temperatures based to the canonical ensemble.

combustible gases [49]. Therefore, it is seen that the g values decrease substantially with an increase in temperature, although the kinds of Syngas of Type A do not show considerable changes between 298.15 K and 600 K, indicating that for this temperature range, the degrees of freedom for these gases are more stable. In addition to the change of the specific heat ratio with temperature, Fig. 7 also shows that the composition of the gases also influences their degrees of freedom; that is, although all the gases of Type A behave similarly, the Syngas from Cotton Stalks presents a slightly lower curve, indicating that its composition presents a smaller degree of freedom when compared to the other gases. Thus, taking into consideration that these kinds of Syngas have considerable fractions of CO and H2, and at the same time lower quantities of H2O and CO2, it may be suggested that the amount of these molecules in the Syngas composition can increase its stability, rendering the fuel less prone to increases in temperature. On the other hand, Poplar Wood gas presents the lowest values for the gamma and Bulk modulus, so that the curves show reductions with the increase in temperature. These results indicate that the compositions of this gas contain less ‘hardness’ when compared to the other types of Syngas studied, therefore presenting components with higher degrees of freedom. It is interesting to note that up to 300 K their curves overlap, presenting the same values for gamma and Bulk moduli. This behavior, of the Poplar Wood gas, is due to its composition being poor in CO and H2, and rich in CO2 and CH4, which makes them more stable during a temperature rise, such as observed for the gases of Type A. It can also be seen, in Fig. 7, that the curves of the Eucalyptus and Wood Shavings gases show themselves to be rather close to each other, so that the former has lower values for temperatures between 300 K and 600 K, but has practically the same values for higher temperatures (>900 K). For Eucalyptus gas, the gamma curve presents a large reduction when temperature increases because its composition has a wide CO2 (16.1%) fraction, such that, even with its great quantities of H2 (46.2%) and CO (33.2%), which improve Syngas stability, the greater degree of freedom of carbon dioxide makes this Syngas less stable. In addition, for Wood Shavings, the CO2 (11.5%) fraction reduces the stability of this Syngas, but its effect is counterbalanced by the effects of the N2 (50.9%) molecule not enabling an even greater reduction of the gamma values. For the calculation of specific heat at a constant pressure (CP), enthalpy (H) and entropy (S) of the gases investigated in this work (Table 1), we adopted numerical interpolation calculations based on the Eqs. 15a-15c, also known as Shomate Equations, which are temperature functions, given by:

(15.c)

The coefficients A, B, C, D, E, F, G and H were obtained by the Levenberg-Marquardt algorithm [51] for each Syngas, which is valid for temperatures within the range of 200 Ke1500 K. The coefficients are shown in Table 2. All equations presented good agreement, such that the coefficient of determination (R2) was 0.99 < R2 < 1. These results indicate that the equations present a favorable fit with the interpolated curves, and thus, they can accurately describe the thermodynamic properties of gases (CP, H e S, as well as other quantities derived from these) obtained from the canonical ensemble model. 3.5. Natural gas/syngas mixture 3.5.1. Thermodynamic properties in equilibrium We also investigated the effect of Syngas acting as an additive in NG during the temperature increase at a constant pressure. The thermodynamic descriptions of the NG/Syngas mixtures were made from the changes of their thermodynamic potentials due to a temperature increase from 298.15 K up to 600 K. These variations were calculated as functions of the Syngas fraction in the mixture 600K 600K forDU600K 298:15K (Fig. 8a), DH298:15K (Fig. 8b) andDS298:15K (Fig. 8c). The effect of incorporating Syngas does not differ significantly between the gases from Olive Wastes, Pine Sawdust, Almond Shells, Bagasse, Poplar Sawdust and Cotton Stalks which were obtained from different biomasses but from the same gasification conditions. On the other hand, the gases from Eucalyptus, Poplar Wood and Wood Shavings presented quite different results. However, it can be seen for all nine types of Syngas investigated that these thermodynamic properties decrease with the increase of the Syngas fraction in the mixture, indicating that it requires less energy to raise its temperature as its composition becomes richer in Syngas.

It can be observed thatDU600K 298:15K decreases with the Syngas fraction according to the following equation:

DU600K 298:15K ¼ 10:40  0:04*X (16.a) for the six gases produced in the same gasification conditions (Type A), which presented similar results, such that for a mixture of 50% Syngas, DU600K 298:15K ¼ 8:45kJ=mol, corresponding approximately to 81.21% of the necessary energy to promote the same temperature changes of NG without Syngas. These gases had the most accentuated reductions in comparison to the other types of Syngas investigated. This result may be related to the lager H2 and CO fractions in their compositions, which makes the NG/Syngas mixture softer and more predisposed to energy absorption, reducing the specific heat at a constant volume in the NG, since this mixture needs less energy to raise its temperature when blended with Syngas. For the other gases straight lines were obtained, respectively, in sequence: Poplar Wood (DU600K 298:15K ¼ 10:40  0:021*X, (16.b)), Eucalyptus (DU600K 298:15K ¼ 10:3  0:03*X, (16.c)) and Wood Shavings (DU600K 298:15K ¼ 10:40  0:034*X, (16.d)). Note that Poplar Wood gas

504

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

Table 2 Shomate Coefficients (equations 15a-15c) obtained through numerical regression for the nine synthetic gases investigated in this study. These equations are valid in the 200 Ke1500 K temperatures range. Shomate Coefficients Syngas

A

B

C

D

E

F

G

H

Bagasse Pine Sawdust Poplar Sawdust Cotton Stalks Almond Shells Olive Wastes Poplar Wood Eucalyptus Wood Shavings t ¼ T(K)/1000

28.89602 28.91618 28.85260 28.79655 28.98348 28.96039 26.47806 27.41977 27.53469

0.60822 0.74769 0.10675 0.55546 1.56855 1.09608 28.33447 9.70912 6.36803

6.41589 6.46805 6.12187 5.74984 6.97682 6.64862 10.27256 0.07759 4.88198

2.42969 2.43178 2.36028 2.27484 2.56179 2.46431 1.11680 0.94888 2.80396

0.00339 0.00316 0.00362 0.00397 0.00281 0.00271 0.02872 0.00884 0.01777

5.31134 4.74601 4.75430 4.75270 4.73381 4.74266 5.36052 4.70363 4.65588

199.85257 199.42114 199.78392 199.70954 200.05740 199.01659 205.19324 199.67635 220.73030

5.54493 3.95301 3.95145 3.96270 3.95303 3.95538 4.58601 3.97181 3.94236

Fig. 8. Variation of the internal energy (a), enthalpy (b) and entropy (c) for the Natural gas/Syngas mixtures due to the temperature increase from 298.15 K up to 600 K under 1atm.

presents the line with the highest values ofDU600K 298:15K , which should again be associated with a large fraction of CO2 and H2O in its composition. However, it can be verified that synthesis gases with larger quantities of these three components increase the CV of NG when added to it. Similarly, for the enthalpy change, which represents heat absorbed by the fuel during its temperature increase, we can see that with the increase of the Syngas fraction in the mixture, the amount of heat absorbed by the gas decreases, such that their CP values suffer a reduction when Syngas is added, as observed for the CV. Thus, for the gases of Type A theirDH600K 298:15K values reduce ac600K cording to the following equation: DH298:15K ¼ 12:05  0:03*X 600K ¼ 10:49kJ=mol; (17.a). Thus, for a mixture of 50% Syngas, DH298:15K this is equivalent to approximately 87% of the heat absorbed by NG,

which is not a significant difference. However, it is important to note that these gases have the lowest values ofDH600K 298:15K . Conversely, for the other gases, the results diverged considerably, such that their graphs were arranged in descending order in the sequence: Poplar Wood (DH600K 298:15K ¼ 12:05  0:0138*X, (17.b)), Eucalyptus (DH600K 298:15K ¼ 12:05  0:02*X, (17.c)) and Wood Shavings (DH600K 298:15K ¼ 12:05  0:027*X, (17.d)). It is important to note that Poplar Wood gas possesses the highest values ofDH600K 298:15K , indicating that compared to other gases, their compositions are the ones that least change the CP values of NG, which is due to their larger fractions of CO2, H2O and also CH4, which is the main component of NG. It is also observed, for the gases of Type A, that theDS600K 298:15K decreases together with the fraction of Syngas,

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

following the equation: DS600K 298:15K ¼ 32:25  0:1149*X (18.a), such that the entropy variation of the gas decreases when the amount of Syngas in the mixture increases. For instance, in a fuel having 50% NG and 50% of these types of Syngas, theDS600K 298:15K ¼ 26:51kJ=mol, corresponding to 82.19% of the entropy change for NG without Syngas. Concerning the other gases, it can be verified that their values were arranged according to that which was observed previously for the internal energy and enthalpy variations. That is, the graphs relating to such NG/Syngas mixtures which also presented divergences proportionally to the Syngas fraction in their compositions, were in agreement with the following sequence in descending order: Poplar Wood (DS600K 298:15K ¼ 32:2  0:1816*X, (18.b)), Eucalyptus (DS600K 298:15K ¼ 32:2  0:1019*X, (18.c)) and Wood Shavings (DS600K 298:15K ¼ 32:2  0:10528*X, (18.d)). Therefore, for the Poplar Wood gas, it may be suggested that the high fractions of CH4 and CO2 in their compositions are responsible for increasing their levels of entropy, since they have higher respective degrees of freedom. 600K From the results obtained forDH600K 298:15K and forDS298:15K ; we were finally able to describe how the temperature change of NG develops when it is combined with Syngas. This analysis is based on

the values of DG600K 298:15K (Fig. 9a) as a function of the Syngas fraction in the mixture. Firstly, we observe that theDG600K 298:15K values for all mixtures simulated increase together with the Syngas fraction

505

added. This behavior, verified for the six types of Syngas produced by the same gasification conditions (Type A), may be described by the following equation: DG600K 298:15K ¼ 68:12  0:157*X (19.a), also obtained by numerical interpolation. On the other hand, for the other NG/Syngas mixtures, the following equations were obtained: Poplar Wood (DG600K 298:15K ¼  68:12 þ 7:8157*X, (19.b)), Eucalyptus (DG600K 298:15K ¼  68:12 þ 0:1417*X, (19.c)) and Wood Shavings (DG600K 298:15K ¼  68:12 þ 0:0825*X, (19.d)). Therefore, it is possible to verify that the ease of increase in NG temperatures decreases with the amount of Syngas (for all Syngas gases simulated) added into its composition. However, for the Poplar Wood gas, it can be observed that its line is the least sloped due to their concentrations of Syngas, showing that this type of Syngas does not modify the behavior of NG to any great degree when the temperature increases, which may be explained by the higher concentration of CH4 in its composition, rendering it similar to NG. It is important to note that NG mixtures with this type of Syngas also appeared to be the most favorable to energy absorption, due to their lower concentrations of CO and H2, components which make heating less favorable. This observed characteristic for CO and H2 also justifies the results of the six gases of Type A, whose graphs ofDG600K 298:15K were the most slope. In general, this characteristic of Syngas increases the value ofDG600K 298:15K for NG, indicating that it tends to act as an additive capable of making NG less sloped to

Fig. 9. (a) Gibbs free energy variation for the Natural gas/Syngas mixtures due to the temperature increase from 298.15 K up to 600 K under 1atm; (b) specific heat ratio (Poisson coefficient) and Bulk modulus for several Natural gas/Syngas mixtures.

506

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

suffer a rise in temperature. Since Syngas is composed of molecules with a smaller degree of freedom than NG molecules, when Syngas is added into NG, this reduces the entropy of the resulting gas, increasing its “hardness” and leaving it less slope to a temperature increase. Fig. 9b and 9 c shows the Bulk (b) modulus as a function of temperature for various concentrations of Syngas from Almond Shells and Poplar Wood, respectively. We can see that the b values increase significantly for higher Syngas concentrations. 3.5.2. Thermodynamic properties of combustion In order to investigate the effects on the combustion of NG when it is mixed with Syngas, some thermodynamic potentials referring to these mixtures were estimated, all for a standard temperature and pressure. This information is important for understanding, for example, what the influence of the Syngas fraction may be on the combustion power and the ease of burning of NG when these gases are combined. Therefore, what follows are the thermodynamic combustion potentials of the NG/Syngas mixtures, for several fractions of Syngas. Thus, analyzing the internal energy and enthalpy of combustion (Fig. 10a and 10 b), the results showed that NG, when combined with Syngas, presents a reduction of the released energy in its burning. Such behavior can be seen from the values ofDc U q andDc H q ; respectively, which decrease with an increasing

amount of Syngas in the blend. Thus, since Dc Hq corresponds to the energy released during combustion under normal conditions of pressure and temperature, it may be noted that Syngas releases less energy compared to NG during combustion. For the six gases of Type A which had very close results, the

internal energy and enthalpy changes as a function of the Syngas fraction are on average described by the following equations

Dc Uq ¼ 879:832 þ 780:929*X (20.a) and q Dc H ¼ 879:806 þ 779:690*X (20.b), which were interpolated from data obtained in the simulations. These same equations can describe the variations concerning the mixture of NG with Syngas of the Wood Shavings type which despite being produced by other gasification conditions, presented similar results to the gases of Type A. Conversely, for the other three synthesis gases, the following linear equation was obtained for the Poplar Wood gas:

Dc Uq ¼ 879:832 þ 602:254*X (21.a) and Dc Hq ¼ 879:806 þ 601:240*X (20.b). It can be observed that this gas modifies the combustion power of NG less, in comparison to other types of simulated Syngas. This result is due to the fact that the composition of this synthesis gas presents a higher amount of H2O and CO2, which are of a non-combustible nature. Moreover, this synthesis gas also presents large amounts of CH4 in its composition, which make its properties more similar to those of NG, especially when blended with it. Additionally, based on Fig. 10 b, it was found that the ratio between theDc H q values for these six types of Syngas mixed with the NG is approximately equal to 0.1138. This result indicates that the burning of Syngas releases around 11.38% of the energy released in NG burning. Differently, while for the Syngas from Poplar Wood the energy released was approximately 31.7% of the energy released by NG. However, despite the lower calorific power of Syngas, it can be verified and then suggested, from Fig. 10 b, that just as it occurs nowadays for a gasoline and ethanol blend, a mixture of 70% NG

Fig. 10. Standard thermodynamic potentials of combustion for the Natural gas/Syngas mixtures at several proportions. (a) Internal energy, (b) enthalpy, (c) entropy and (d) Gibbs free energy.

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

and 30% Syngas (or other closer proportions) would be a desirable proportion, since for this mixture the resulting fuel presents approximately 73.41% (for the Syngas obtained from Cotton Stalks) or 79.49% (for the Syngas obtained from Poplar Wood) of the calorific power of NG, which may still be a desirable percentage for combustion power. Simultaneously, it can be seen in Fig. 10 c that the addition of Syngas reduces the values ofDc Sq . It was observed that all 9 types of investigated Syngas exhibited this same behavior when combined with NG. Thus, it is concluded that the addition of Syngas reduces the combustion entropy of NG when both are mixed together. However, it was found that depending on their compositions, and consequently, on their gasification conditions, the different types of Syngas have more or fewer variations. Thus, for the gases of Type A, similar results were observed forDc Sq ; thus these gases, when mixed with NG, were the ones that most diminished in their combustion entropy, this being reduced approximately by the same proportion, which is described by the closer graphs in Fig. 10 c, given by the equation: Dc Sq ¼ 263:156 þ 122:018*X (23.a). These results are due to the fact that these types of Syngas present similar compositions (with H2 and CO as their major components) and effects, which are components of low entropy (or equivalent, with few degrees of freedom) so the addition of these types of Syngas in NG reduces the combustion entropy of NG. That being said, for Poplar Wood gas, it was verified that its effects on the entropy of NG combustion are significantly lower, presenting values that can be described by the equation

Dc Sq ¼ 263:156 þ 91:278*X (23.b). This result is justified by the greater amount of CH4 and CO2 in the composition of this Syngas type. Thus, since these components have higher degrees of freedom, the entropy associated with them becomes larger, and therefore closer than those obtained for NG, consequently, there are no large reductions in the entropy of NG when mixed with this Syngas type. Finally, given the values to enthalpy and entropy of combustion for the NG/Syngas mixtures, how favorable is the combustion of each mixture can be analyzed. This analysis may be carried out, for each mixture, from the calculation of standard Gibbs free energy of combustion, given byDc Gq ¼ Dc H q  T Dc Sq (24), being T ¼ 298.15 K.

For the gases of Type A the values ofDc Gq (see Fig. 10d) can be described on average by the equation Dc Gq ¼ 801:346 þ 743:311* X (25.a). It is noted that NG mixtures with each of the six gases of Type A present the least expressive values (least negative) ofDc Gq ; indicating that their compositions more effectively render NG combustion less favorable. This result is directly related to the high fractions of H2 and CO in the composition of these types of Syngas. On the other hand, the results of the mixture of NG with the gas

from Poplar Wood (which equation is given byDc Gq ¼  801:346 þ 574:026*X, (25.b)), demonstrated that it is the type that produced the most expressive values (most negative) ofDc Gq ; which is justified by its composition, which has a lower amount of H2 and CO, when compared to the six types of Syngas previously cited. Furthermore, this synthesis gas also presents the largest fraction of CH4, rendering its properties closer to those of NG, and therefore, not promoting major changes when mixed. Nonetheless, based on what has been discussed on the enthalpy and entropy of combustion, it is possible to verify that theDc Sq term reduces in a smaller proportion compared to the reduction suffered byDc H q for all types of NG/Syngas mixtures. Consequently, the equation for theDc Gq indicates that the termDc H q becomes more expressive in comparison to the T Dc Sq (called entropic energy), and that this difference increases proportionally to the amount of

507

Syngas combined with NG. However, it can be concluded that the combustion of the NG/Syngas mixture becomes less favorable (the

Dc Gq values become less negative) as the Syngas fraction increases, showing again that this biofuel has the property of raising NG's resistance to combustion. 4. Conclusions In this work we used density functional theory along with the canonical ensemble model, in order to investigate the thermodynamic properties of nine kinds of Syngas, as well as the influence of these gases when combined with natural gas. Firstly, we determined a satisfactory DFT method to predict thermodynamic properties for natural gas and for Syngas. This investigation showed that B3lyp functional, with 6e311þþg(d,p) as a basis set, is a positive method for performing this kind of analysis. Additionally, the effect on the composition of Syngas was investigated in its thermodynamic properties from a temperature increase at constant pressure, which permitted us to gain important information on the heating of this fuel. These results showed that H2 and CO, which are initially the major components of Syngas, reduce the ease of temperature increases in Syngas, while synthesis gases with compositions rich in CH4 and CO2 are more favorable to increases in temperature. This characteristic is related to these components' degrees of freedom. Thus, in order to obtain a better explanation about this property, it may be suggested that H2 and CO can act as important components in the thermal control of Syngas stability. We also estimated the Poisson coefficients and the Bulk modulus for each Syngas investigated, where we verified that the degrees of freedom or “hardness” of the gas in question increases proportionally with the amount of H2 and CO in its composition. Therefore, we concluded that the H2 and CO fraction in the composition of Syngas is proportional to the thermal stability of this biofuel. However, in order to provide a more complete description about the effect of temperature on Syngas, the Shomate equations were interpolated for the following parameters: specific heat at a constant pressure, and enthalpy and entropy of the gases. From these equations, which presented good agreement with the results obtained in the frequency calculations, we calculated the properties of Syngas for any temperature in the range of 200 K up to 1500 K. The results may be very useful for works related to fuel and biofuel modeling. The change in the propensity for natural gas to increase its temperature was also verified when Syngas was added to its composition. The results indicated that Syngas can act as an additive capable of making natural gas less liable to suffer a temperature increase. We observed that Syngas presents a composition with a smaller degree of freedom in comparison to natural gas's components, making it less energetic. Thus, when this biofuel is added to natural gas, it reduces the entropy of the fuel, resulting in the increase of its “hardness”, as well as leaving it less prone to temperature increases. Finally, the prediction about the standard thermodynamic potentials of combustion for the NG/Syngas mixtures showed that the Syngas fraction can reduce the calorific power of natural gas, besides reducing this fuel's propensity for burning. Therefore, from the interpolation of these quantities, it is possible to suggest that an addition of around 30% of a Syngas rich in H2 and CO to the composition of natural gas can be useful for its combustion, since this mixture releases approximately 11.38% of the energy released by the combustion of natural gas. Similarly, for a Syngas rich in CH4, the energy released by a similar mixture has been in the 27.63e31.7% range of the energy released by natural gas. In other words, for a mixture of 70% natural gas þ30% Syngas, the resulting

508

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

fuel presents a calorific power between 73.41% and 79.49% of that of natural gas, which occurs with gasoline, for instance, when mixed with ethanol, in several different countries. It shows a very important relationship of the use of Natural gas/Syngas mixture as very helpful information to next generation of these gaseous fuels. Besides this, a synthesis gas having a higher fraction of H2O, CO2 and CH4 tends to render it less favorable to both the heating and combustion of natural gas. However, CO is significantly more effective in this behavior, since the Gibbs free energy variation of the synthesis gases rich in CO present less expressive values for both cases (heating and combustion). These results are related to the fact that these Syngas types present a rich composition of H2 and CO, which are components that each has a smaller degree of freedom. Acknowledgments Abel F.G. Neto, Amanda D.S. Ferreira, Adriana T. Amador, Antonio M.J.C. Neto are grateful for the support conferred by CAPES, CNPq and PROPESP/UFPA.

[17]

[18]

[19]

[20]

[21]

[22]

[23]

References [1] G.C. Barker, Biofuels, in: B. Thomas, B.G. Murray, D.J. Murphy (Eds.), Encyclopedia of Applied Plant Sciences, second ed., 2017, pp. 153e158, https:// doi.org/10.1016/B978-0-12-394807-6.00243-4. [2] F. van der Ploeg, A. Rezai, Cumulative emissions, unburnable fossil fuel, and the optimal carbon tax, Technol. Forecast. Soc. Change 116 (2016) 216e222, https://doi.org/10.1016/j.techfore.2016.10.016. [3] B.H. Tsai, C.J. Chang, C.H. Chang, Elucidating the consumption and CO2 emissions of fossil fuels and low-carbon energy in the United States using LotkaeVolterra models, Energy 100 (2016) 416e424, https://doi.org/10.1016/ j.energy.2015.12.045. [4] Z.X. Wang, D.J. Ye, Forecasting Chinese carbon emissions from fossil energy consumption using non-linear grey multivariable models, J. Clean. Prod. 142 (2016) 600e612, https://doi.org/10.1016/j.jclepro.2016.08.067. [5] F. Yüksel, B. Yüksel, The use of ethanolegasoline blend as a fuel in an SI engine, Renew. Energy 29 (2004) 1181e1191, https://doi.org/10.1016/ j.renene.2003.11.012. [6] A.K. Das, P. Hong, Soluteesolvent friction kernels and solution properties of methyl oxazolineephenyl oxazoline (MeOxePhOx) copolymers in binary ethanolewater mixtures, Phys. Chem. Chem. Phys. 13 (2011) 11892e11904, https://doi.org/10.1039/C0CP02982C. [7] L.G.T. Carpio, F.S. de Souza, Optimal allocation of sugarcane bagasse for producing bioelectricity and second generation ethanol in Brazil: scenarios of cost reductions, Renew. Energy 111 (2017) 771e780, https://doi.org/10.1016/ j.renene.2017.05.015. [8] D. Chiaramonti, M. Prussi, M. Buffi, D. Tacconi, Sustainable bio kerosene: process routes and industrial demonstration activities in aviation biofuels, Appl. Energy 136 (2014) 767e774, https://doi.org/10.1016/j.apenergy.2014. 08.065. [9] K.F. Mustafa, S. Abdullah, M.Z. Abdullah, K. Sopian, A.K. Ismail, Experimental investigation of the performance of a liquid fuel-fired porous burner operating on kerosene-vegetable cooking oil (VCO) blends for micro-cogeneration of thermoelectric power, Renew. Energy 74 (2015) 505e516, https://doi.org/ 10.1016/j.renene.2014.08.061. [10] S.P. Singh, D. Singh, Biodiesel production through the use of different sources and characterization of oils and their esters as the substitute of diesel: a review, Renew. Sust. Energy Rev. 14 (2010) 200e216, https://doi.org/10.1016/ j.rser.2009.07.017. [11] P. Dubey, R. Gupta, Influences of dual bio-fuel (Jatropha biodiesel and Turpentine oil) on single cylinder variable compression ratio diesel engine, Renew. Energy 115 (2017) 1294e1302, https://doi.org/10.1016/ j.renene.2017.09.055. [12] P. Lv, Z. Yuan, C. Wu, L. Ma, Y. Chen, N. Tsubaki, Bio-syngas production from biomass catalytic gasification, Energy Convers. Manag. 48 (2007) 1132e1139, https://doi.org/10.1016/j.enconman.2006.10.014. [13] A. Kumar, D.D. Jones, M.A. Hanna, Thermochemical biomass gasification: a review of the current status of the technology, Energies 2 (2009) 556e581, https://doi.org/10.3390/en20300556. [14] K. Li, R. Zhang, J. Bi, Experimental study on syngas production by cogasification of coal and biomass in a fluidized bed, Int. J. Hydrogen Energy 35 (2010) 2722e2726, https://doi.org/10.1016/j.ijhydene.2009.04.046. [15] C. Karavalakis, M. Baratieri, B. Bosio, E. Arato, P. Baggio, Process analysis of a molten carbonate fuel cell power plant fed with a biomass syngas, J. Power Sources 157 (2006) 765e774, https://doi.org/10.1016/ j.jpowsour.2005.12.038. [16] M.J.A. Tijmensen, A.P.C. Faaij, C.N. Hamelinck, M.R.M. van Hardeveld,

[24]

[25]

[26]

[27] [28]

[29] [30]

[31]

[32] [33]

[34] [35] [36]

[37]

[38]

[39]

Exploration of the possibilities for production of Fischer Tropsch liquids and power via biomass gasification, Biomass Bioenergy 23 (2002) 129e152, https://doi.org/10.1016/S0961-9534(02)00037-5. A. Corujo, L. Yerm
an, B. Arizaga, M. Brusoni, J. Castiglioni, Improved yield parameters in catalytic steam gasification of forestry residue; optimizing biomass feed rate and catalyst type, Biomass Bioenergy 34 (2010) 1695e1702, https://doi.org/10.1016/j.biombioe.2010.06.010. K. Patil, P. Bhoi, R. Huhnke, D. Bellmer, Biomass downdraft gasifier with internal cyclonic combustion chamber: design, construction, and experimental results, Bioresour. Technol. 102 (2011) 6286e6290, https://doi.org/10.1016/ j.biortech.2011.03.033. L.P.R. Pala, Q. Wang, G. Kolb, V. Hessel, Steam gasification of biomass with subsequent syngas adjustment using shift reaction for syngas production: an Aspen Plus model, Renew. Energy 101 (2017) 484e492, https://doi.org/ 10.1016/j.renene.2016.08.069. F.Y. Hagos, A.R.A. Aziz, S.A. Sulaiman, Syngas (H2/CO) in a spark-ignition direct-injection engine. Part 1: combustion, performance and emissions comparison with CNG, Int. J. Hydrogen Energy 39 (2014) 17884e17895, https://doi.org/10.1016/j.ijhydene.2014.08.141. E. Johansson, T. Mattisson, A. Lyngfelt, H. Thunman, Combustion of syngas and natural gas in a 300 W chemical-looping combustor, Chem. Eng. Res. Des. 84 (2006) 819e827, https://doi.org/10.1205/cherd05024. N. Donohoe, K.A. Heufer, C.J. Aul, E.L. Petersen, G. Bourque, R. Gordon, H.J. Curran, Influence of steam dilution on the ignition of hydrogen, syngas and natural gas blends at elevated pressures, Combust. Flame 162 (2015) 1126e1135, https://doi.org/10.1016/j.combustflame.2014.10.005. T. Le Cong, P. Dagaut, G. Dayma, Oxidation of natural gas, natural gas/syngas mixtures, and effect of burnt gas recirculation: experimental and detailed kinetic modeling, J. Eng. Gas Turbines Power 130 (2008) 1e10, https://doi.org/ 10.1115/1.2901181. T. Shudo, T. Takahashi, Influence of reformed gas composition on HCCI combustion engine system fueled with DME and H2-CO-CO2 which are onboardreformed from methanol utilizing engine exhaust heat, Trans. Japan Soc. Mech. Eng. Pt. B. 70 (2004) 2663e2669, https://doi.org/10.4271/2004-011908. A.L. Boehman, O. Le Corre, Combustion of syngas in internal combustion engines, Combust. Sci. Technol. 180 (2008) 1193e1206, https://doi.org/10.1080/ 00102200801963417. H. Li, G.A. Karim, Exhaust emissions from an SI engine operating on gaseous fuel mixtures containing hydrogen, Int. J. Hydrogen Energy 30 (2005) 1491e1499, https://doi.org/10.1016/j.ijhydene.2005.05.007. P. Hohenberg, W. Kohn, Inhomogeneous electron gas, Phys. Rev 136 (1964) B864, https://doi.org/10.1103/PhysRev.136.B864. W. Kohn, L.J. Sham, Self-consistent equations including exchange and correlation effects, Phys. Rev 140 (1965) A1133, https://doi.org/10.1103/ PhysRev.140.A1133. D.A. McQuarrie, J.D. Simon, Molecular Thermodynamics, University Science Books, 1999. D. Ogunkoya, T. Fang, Engine performance, combustion, and emissions study of biomass to liquid fuel in a compression-ignition engine, Energy Convers. Manag. 95 (2015) 342e351, https://doi.org/10.1016/j.enconman.2015.02.041. R. Chen, K. Nishida, Spray evaporation of ethanolegasoline like blend and combustion of ethanolegasoline blend injected by hole-type nozzle for directinjection spark ignition engines, Fuel 134 (2014) 263e273, https://doi.org/ 10.1016/j.fuel.2014.05.082. E. Juaristi, Introduction to Stereochemistry and Conformational Analysis, Wiley, Minnesota, 1991. A.L. Norman, Conformational analysis. 130. MM2. A hydrocarbon interatomic potential utilizing V1 and V2 torsional terms, J. Am. Chem. Soc. 99 (1977) 8127e8134, https://doi.org/10.1021/ja00467a001. A. Hocquet, Langgård, an evaluation of the MMþ force field M, J. Mol. Med. 4 (1998) 94, https://doi.org/10.1007/s008940050128. HYPERCHEM (TM), Hypercube, Inc., 1115 NW 4th Street, Gainesville, Florida 32601; www.hyper.com. M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, € Farkas, J.B. Foresman, J.V. Ortiz, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. J. Cioslowski, D.J. Fox, Gaussian 09W, Revision A.1, Gaussian Inc, Wallingford, 2009. A.D. Becke, Density-functional exchange-energy approximation with correct asymptotic behavior, Phys. Rev. 38 (1988) 3098e3100, https://doi.org/ 10.1103/PhysRevA.38.3098. A.D. Becke, Density-functional thermochemistry. III. The role of exact exchange, J. Chem. Phys. 98 (1993) 5648e5652, https://doi.org/10.1063/ 1.464913. X. Huang, D. Cheng, F. Chen, X. Zhan, Reaction pathways of hemicellulose and

A.F.G. Neto et al. / Renewable Energy 130 (2019) 495e509

[40]

[41]

[42]

[43] [44]

[45]

mechanism of biomass pyrolysis in hydrogen plasma: a density functional theory study, Renew. Energy 96 (2016) 490e497, https://doi.org/10.1016/ j.renene.2016.04.080. A.F.G. Neto, M.N. Huda, F.C. Marques, R.S. Borges, A.M.J.C. Neto, Thermodynamic DFT analysis of natural gas, J. Mol. Model. 23 (2017) 224, https:// doi.org/10.1007/s00894-017-3401-1. M.N. Makadsi, M.F.A. Alias, A.A. Essa, H.R. AI-Azawi, FT-IR, XPS analysis of aSi1-xGex:H thin films, Renew. Energy 28 (2003) 975e984, https://doi.org/ 10.1016/S0960-1481(02)00061-7. D. Gracin, K. Juraic, A. Gajovic, P. Dubcek, C. Devilee, H.J. Muffler, W.J. Soppe, S. Bernstorff, The structural ordering of thin silicon films at the amorphous to nano-crystalline phase transition by GISAXS and Raman spectroscopy, Renew. Energy 33 (2008) 326e330, https://doi.org/10.1016/j.renene.2007.05.015. S. Turns, An Introduction to Combustion: Concepts and Applications, third ed., McGraw-Hill, New York, 2011. G. Karavalakis, M. Hajbabaei, T.D. Durbin, K.C. Johnson, Z. Zheng, W.J. Miller, The effect of natural gas composition on the regulated emissions, gaseous toxic pollutants, and ultrafine particle number emissions from a refuse hauler vehicle, Energy 50 (2013) 280e291, https://doi.org/10.1016/j.energy.2012. 10.044. T.N.C. Anand, A.M. Mohan, R.V. Ravikrishna, Spray characterization of

[46]

[47]

[48]

[49]

[50] [51]

509

gasoline-ethanol blends from a multi-hole port fuel injector, Fuel 102 (2012) 613e623, https://doi.org/10.1016/j.fuel.2012.06.107. Y. Huang, S. Huang, R. Huang, G. Hong, Spray and evaporation characteristics of ethanol and gasoline direct injection in non-evaporating, transition and flash-boiling conditions, Energy Convers. Manag. 108 (2016) 68e77, https:// doi.org/10.1016/j.enconman.2015.10.081. A.F.G. Neto, F.S. Lopes, E.V. Carvalho, M.N. Huda, A.M.J.C. Neto, N.T. Machado, Thermodynamic analysis of fuels in gas phase: ethanol, gasoline and ethanol gasoline predicted by DFT method, J. Mol. Model. 21 (2015) 267e275, https:// doi.org/10.1007/s00894-015-2815-x. P.J. Linstrom, W.G. Mallard (Eds.), NIST Chemistry WebBook, NIST Standard Reference Database Number 69,, National Institute of Standards and Technology, Gaithersburg MD, July 2001, p. 20899. http://webbook.nist.gov. R. Ebrahimi, Effect of specific heat ratio on heat release analysis in a spark ignition engine, Sci. Iran. 18 (2011) 1231e1236, https://doi.org/10.1016/ j.scient.2011.11.002. W.L.F. Armarego, C.L.L. Chai, Purification of Laboratory Chemicals, sixth ed., Butterworth-Heinemann, Oxford, 2013. K. Levenberg, A method for the solution of certain non-linear problems in least squares, Q. Appl. Math. 2 (1944) 164e168, https://doi.org/10.1090/qam/ 10666.