中文翻译

Proceedings of the Combustion Institute

V olume 32, Issue 1, 2009, Pages 229-237

doi:10.1016/j.proci.2008.05.005 | How to Cite or Link Using DOI

Copyright ? 2009 Elsevier Inc. All rights reserved.

Permissions & Reprints

A chemical kinetic study of n-butanol oxidation at elevated pressure in a jet stirred reactor

P. Dagauta, S.M. Sarathyb and M.J. Thomsonb, ,

aCNRS, 1C, Avenue de la recherche scientifique, 45071 Orléans Cedex 2, France bDepartment of Mechanical and Industrial Engineering, University of Toronto, 5 King’s College Road, Toronto, Ont., Canada M5S 3G8

Available online 20 September 2008.

Abstract

Biofuels are attractive alternatives to petroleum derived transportation fuels. n-Butanol,or biobutanol, is one alternative biofuel that can replace gasoline and diesel in transportation applications. Similar to ethanol, n-butanol can be produced via the fermentation of sugars, starches, and lignocelluloses obtained from agricultural feedstocks. n-Butanol has several advantages over ethanol, but the detailed combustion characteristics are not well understood. This paper studies the oxidation of n-butanol in a jet stirred reactor at 10 atm and a range of equivalence ratios. The profiles for CO, CO2, H2O, H2, C1–C4 hydrocarbons, and C1–C4 oxygenated compounds are presented herein. High levels of carbon monoxide, carbon dioxide, water, hydrogen, methane, formaldehyde, ethylene, and propene are detected. The experimental data are used to validate a novel detailed chemical kinetic mechanism for n-butanol oxidation. The proposed mechanism well predicts the concentration of major product species at all temperatures and equivalence ratios studied. Insights into the prediction of other species are presented herein. The proposed mechanism indicates that n-butanol consumption is dominated by H-atom abstraction from the α, β, and γ carbon atoms. A sensitivity analysis is also presented to show the effects of reaction kinetics on the concentration of several poorly predicted species. Keywords: n-Butanol; 1-Butanol; Jet stirred reactor; Kinetic modeling; Reaction mechanism Article Outline

1.

Introduction

2.

Experimental methods

3.

Computational methods

4.

Results and discussion

5.

Conclusions

Acknowledgements

Appendix A.

Supplementary data

References

1. Introduction

A potential biofuel for use in both gasoline and diesel engines is n-butanol.Historically, industrial scale production of n-butanol from biomass feedstocks was the second largest fermentation process, exceeded only by ethanol. However, its demise was brought about in the early 1960s when petroleum derived n-butanol became more economically feasible [1]. Recent advances in n-butanol production in the laboratory have spurred interest in commercial scale production of the n-butanol[2] and [3]. Recently, BP and Dupont announced that they would commercially produce n-butanol,which they call biobutanol, as a gasoline blending component for automotive fuels [4] and [5]. n-Butanol is produced via a fermentation process similar to that of ethanol, and therefore its feedstocks could include sugar beet, sugar cane, corn, wheat and also cellulosic biomass. n-Butanol has several advantages over ethanol including enhanced tolerance to water contamination allowing the use of existing distribution pipelines, the ability to blend at higher concentrations without retrofitting vehicles, and better fuel economy.

Relatively few engine studies of n-butanol have been published. Yacoub et al. used gasoline blended with a range of C1–C5 alcohols (including n-butanol)to fuel a single-cylinder spark ignition (SI) engine [6]. They found that the n-butanol blends had less knock resistance than neat gasoline. The n-butanol blends also had reduced CO and hydrocarbon emissions but increased NOx emissions. This may be due to the n-butanol blends having a higher flame temperature and earlier spark timing. Of particular interest to the present study is that the primary oxygenated hydrocarbon emissions were n-butanol,formaldehyde and to a lesser extent, acetaldehyde. A study by Miller et al. successfully operated unmodified gasoline and diesel engines on blends containing 0–20% n-butanol in gasoline and 0–40% n-butanol in diesel fuel [7]. Another study successfully ran a compression ignition (CI) engine fueled with n-butanol and diesel fuel microemulsions [8].

Predictive models provide a better understanding of the combustion performance and emissions characteristics of biofuel compositions and why they differ from petroleum derived materials. The development of an n-butanol model requires understanding of its fundamental pyrolysis and oxidation kinetics. However, few studies have examined the combustion chemistry of n-butanol, while none have developed a detailed chemical kinetic mechanism of the fuel. A 1959 study by Barnard examined the pyrolysis of n-butanol[9]. The experiments were carried out in a static reactor at temperatures between 579 and 629 °C. Barnard suggested that, in the absence of oxygen, n-butanol primarily reacts by the fission of the molecule at the C3H7–CH2OH bond. This produces formaldehyde, ethylene and a methyl radical, following the decomposition of the n-propyl radical. Barnard also conducted a similar study of t-butanol[10]. A study by Roberts measured the burning velocities of n-butanol using schlieren photographs of the flames [11], and found that the maximum burning velocity of n-butanol is similar to that of isopropyl alcohol and isopentyl alcohol. A recent study by McEnally and Pfefferle [12] measured the temperature and species in an atmospheric-pressure coflowing laminar nonpremixed flames. The fuels consisted of methane doped with one of the four isomers of butanol.They claimed that unimolecular dissociation was dominant, not H-atom abstraction. For n-butanol,this consisted of C–C fission

followed by β scission of the resulting radicals. In the case of n-butanol,complex fission involving four-center elimination of water was estimated to account for only 1% of n-butanol decomposition. The most important measured species included ethylene (C2H4) and propene (C3H6). More recently, Yang and co-workers [13] studied laminar premixed flames fuelled by one of four isomers of butanol(including n-butanol). Their results identify combustion intermediates in the butanol flames, but do not provide concentration profiles. The qualitative data provided lends support to the aforementioned dissociation mechanism proposed by McEnally and Pfefferle [12].

In this paper, we report new experimental data obtained in a jet stirred reactor (JSR) for the oxidation of n-butanol at a pressure of 10 atm and a range of equivalence ratios (0.5–2.0) and temperatures (800–1150 K). In addition, a chemical kinetic model of n-butanol is developed using the JSR experiments as validation data. Both experimental and kinetic insights are offered below.

2. Experimental methods

The JSR experimental setup used in this study has been described earlier [14] and [15]. The JSR consists of a small sphere of 4 cm diameter (39 cm3) made of fused silica (to minimize wall catalytic reactions), equipped with four nozzles of 1 mm i.d. for the admission of the gases which achieve stirring. The reactants were diluted by high-purity nitrogen (<50 ppm O2, <1000 ppm Ar, <5 ppm H2) and mixed at the entrance of the injectors. A high degree of dilution (0.1% volume of fuel) was used, reducing temperature gradients and heat release in the JSR. High-purity oxygen (99.995% pure) was used in these experiments. All the gases were preheated before injection to minimize temperature gradients inside the reactor. A regulated heating wire of ca. 1.5 kW maintained the temperature of the reactor at the desired working temperature. The n-butanol was sonically degassed before use. A Shimadzu LC10 AD VP pump with an on-line degasser (Shimadzu DGU-20 A3) was used to deliver the fuel to an atomizer–vaporizer assembly maintained at 200 °C. Good thermal homogeneity along the vertical axis of the reactor (gradients of ca. 1 K/cm) was observed for each experiment by thermo-couple (0.1 mm Pt–Pt/Rh (10%) located inside a thin-wall silica tube) measurements. The reacting mixtures were probe sampled by means of a fused silica low pressure sonic probe. The samples were analyzed online by FT-IR and off-line after collection and storage in 1 L Pyrex bulbs. Off-line analysis was done using gas chromatographs equipped with capillary columns (DB-624 and Carboplot-P7), a TCD (thermal conductivity detector), and an FID (flame ionization detector).

The experiments were performed at steady state, at a constant mean residence time of 0.7 s and a constant pressure of 10 atm. The reactants were continually flowing in the reactor while the temperature of the gases inside the JSR was increased stepwise. A good repeatability was observed in the experiments and reasonable good carbon balance of 100 ± 15% was achieved.

3. Computational methods

The kinetic modeling was performed using the PSR computer code [16] that computes species concentrations from the net rate of production of each species by chemical reactions and the difference between the input and output flow rates of the species. These rates are computed from the kinetic reaction mechanism and the rate constants of the elementary reactions calculated at the experimental temperature.

The reaction mechanism used here is based on a previously proposed oxidation mechanism [17], [18] and [19] for C1–C4 chemistry. Additional reactions have been added to represent the

butanol mechanism and are listed in Table 1. The oxidation of n-butanol proceeds via unimolecular initiation and hydrogen abstraction reactions. The fuel radical species formed are consumed via unimolecular decomposition (β-scission) and biomolecular reactions. Isomerization of radical species is also included in the proposed model. Table 2 presents the structure of species produced during the oxidation of n-butanol.The rate expression for new reactions derives from tabulations for alkanes and alcohols [18] and [19]. This mechanism, including references and thermochemical data, is available as Supplementary material to this article. The rate constants for reverse reactions are computed from the corresponding forward rate constants and the appropriate equilibrium constants, calculated from thermochemistry [20] and [21].

Table 1. Reactions representing the oxidation of n-butanol

Full-size table

Note: X denotes a radical species (OH, H, CH3, O, HCO, HO2, CH2OH, CH3O, C2H5, C2H4, C4H7, aC3H5).

View Within Article

aC4H8OH

bC4H8OH

cC4H8OH

dC4H8OH

cC3H6OH

aC3H6OH

Full-size table

View Within Article

4. Results and discussion

Molecular species concentration profiles were measured by sonic probe sampling and GC and FT-IR analyses from the oxidation of n-butanol in a JSR: hydrogen (H2), water (H2O), carbon monoxide (CO), carbon dioxide (CO2), methane (CH4), acetylene (C2H2), ethylene (C2H4), ethane (C2H6), propene (C3H6), 1-butene (C4H8), acetaldehyde (CH3HCO), formaldehyde (CH2O), butyraldehyde (C3H7CHO), and n-butanol(C4H9OH). Figure 1 presents the experimental measurements and modeling results of n-butanol obtained at = 1.0. The

experimental results (solid symbols) show that with increasing temperature, the n-butanol levels drop significantly between 800 and 900 K. This corresponds to a large increase in the concentrations of butyraldehyde, 1-butene, and propene, all of which are products of H abstraction pathways. The concentration of these compounds then quickly decreases as the temperature increases. Ethylene, ethane, acetaldehyde, and formaldehyde concentrations are also shown to increase between 800 and 900 K. However, as the temperature increases further, the concentrations of these species tends to diminish at a slower rate than the aforementioned species.

Full-size image (63K)

Fig. 1. Comparison of the experimental concentration profiles obtained from the oxidation of n-butanol in a JSR at = 1, P = 10 atm, τ = 0.7 s.

View Within Article

The following oxygenated products were detected: butanal, ethyloxirane, propanal, 2-propenal, methyloxirane, oxirane, and acetaldehyde. The oxiranes, 2-propenal, and propanal are formed at low ppm levels, and therefore no concentration profiles are reported. Enols were not detected. A comparison with results obtained for ethanol in similar conditions and keeping the initial carbon content shows butanol oxidation produces less aldehydes overall. The maximum amount of

acetaldehyde production is reduced by ca. 70% when changing the fuel from ethanol to butanol. The model predictions (open symbols with line) for = 1.0 are also shown in Fig. 1. Reasonably

good agreement is obtained for all measured species. The major product species (i.e., CO, CO2, and H2O) are well predicted by the model. Methane, ethylene, hydrogen, and formaldehyde are also reasonably well predicted across the entire temperature range. The reactivity of n-butanol is well predicted between 800 and 950 K, but at greater temperatures the reactivity is overpredicted. Species concentrations of butyraldehyde, 1-butene, and acetaldehyde are well predicted until approximately 1000 K, above which they become underpredicted. The propene concentration is underpredicted across the entire temperature range, while ethane and acetylene concentrations are overpredicted across the entire temperature range.

Figure 2 presents the experimental measurements and modeling results of n-butanol obtained at = 0.5. For the most part, the experimental results show a similar trend to that observed at

= 1.0. The concentration of n-butanol is lower at = 0.5 than at = 1.0 due to the fact that a

greater oxygen concentration exists in the oxygen–fuel mixture. The model better predicts the concentration of most species at = 0.5 than it does at = 1.0. 1-Butene, propene, butyraldehyde,

carbon monoxide, carbon dioxide, methane, ethylene, acetaldehyde, formaldehyde, water, and hydrogen are well predicted across the entire temperature range. Similar to the case of = 1.0, the

reactivity of n-butanol is overpredicted above 900 K. Again, the concentrations of acetylene and ethane are overpredicted across the entire temperature range.

Full-size image (64K)

Fig. 2. Comparison of the experimental concentration profiles obtained from the oxidation of n-butanol in a JSR at = 0.5, P = 10 atm, τ = 0.7 s.

View Within Article

Figure 3 presents the experimental measurements and modeling results of n-butanol obtained at = 2.0. Similar trends as those observed for other equivalence ratios are observed for the

experimental data at = 2.0. At = 2.0, the reactivity of n-butanol is well predicted across the

entire temperature range, something which was not observed at other equivalence ratios In addition, there is good prediction of carbon monoxide, carbon dioxide, methane, ethylene, acetaldehyde, ethane, formaldehyde, water, and hydrogen. Qualitatively, the prediction of acetylene concentration is satisfactory. The butyraldehyde concentration is well predicted below 1000 K, while above 1000 K the model underpredicts the experimental data. The concentration of 1-butene is overpredicted above 900 K, while the concentration of propene is under underpredicted across the entire temperature range.

Full-size image (63K)

Fig. 3. Comparison of the experimental concentration profiles obtained from the oxidation of n-butanol in a JSR at = 2, P = 10 atm, τ = 0.7 s.

View Within Article

Some general trends are observed via analysis of the data across the three equivalence ratios. The model’s prediction of carbon monoxide, carbon dioxide, methane, ethylene, for maldehyde, water, and hydrogen concentrations is reasonably accurate across all equivalence ratios. The prediction of n-butanol,acetaldehyde, and acetylene concentrations tends to improve with increasing equivalence ratio. On the other hand, an increase in equivalence ratios results in poorer prediction of 1-butene, propene, butyraldehyde, and ethane concentrations.

A reaction pathway analysis was performed at = 1.0 at T = 1000 K to determine the most

dominant pathways for n-butanol consumption. Figure 4 presents the results of the analysis in diagram format, wherein heavier weight arrows represent more dominant reaction pathways. According to the proposed model, n-butanol is consumed primarily via H-atom abstraction from the α, β, and γ carbon atoms, with each pathway accounting for approximately 22% of the total n-butanol consumption. The next most dominant pathway is H-atom abstraction from the hydroxyl group, which accounts for nearly 20% of n-butanol consumption. H-atom from the δ carbon atom accounts for nearly 14% while all the unimolecular decomposition pathways combined account for less than 0.5% of n-butanol consumption. Similarly, a reaction pathway analysis at T = 1200 K showed that unimolecular decomposition accounted for less than 4% of n-butanol consumption. Therefore, it is reasonable to conclude that n-butanol consumption in the JSR is dominated by H-atom abstraction.

Full-size image (16K)

Fig. 4. Reaction pathway diagram for n-butanol oxidation in the JSR at = 1, P = 10 atm,

τ = 0.7 s, T = 1000 K.

View Within Article

The pathways diagram in Fig. 4 indicates that the aC4H8OH radical primarily undergoes β-scission to form acetaldehyde and an ethyl radical (C2H5). The consumption of the bC4H8OH radical is also consumed primarily by β-scission to form a hydroxyl radical and 1-butene. The cC4H8OH radical primarily undergoes β-scission to form propene and a hydroxymethyl radical

(CH2OH). The hydroxymethyl radical, which is also an intermediate in several n-butanol unimolecular decomposition pathways, undergoes β-scission to create formaldehyde. The C4H9O radical, which is formed primarily via H-atom abstraction from the n-butanol hydroxyl group, undergoes β-scission to form butyraldehyde. The least prominent n-butanol H-atom abstraction pathway leads to the formation of the dC4H8OH radical, which isomerizes to form the aC4H8OH radical. The n-butanol unimolecular dissociation reactions proceed to form radical species, which then undergo β-scission to form stable species such as acetylene, ethylene, and formaldehyde, and a number of radical species.

Sensitivity analyses were conducted for n-butanol,propene, and acetylene as these compounds were not always well predicted by the model. n-Butanol was underpredicted above 900 K at both = 1.0 and = 0.5. Figure 5a displays the normalized sensitivity coefficients for the top 12

reactions to which the n-butanol concentration is sensitive at T = 1050 K and all equivalence ratios. A positive sensitivity coefficient implies that an incre ase in the reaction’s forward rate will increase the n-butanol concentration at the specified temperature and equivalence ratio. At all equivalence ratios, the n-butanol concentration is very sensitive to the reaction producing OH radicals via the oxidation of H radicals. At = 0.5, the n-butanol concentration is mainly

sensitive to elementary reactions between hydrogen and oxygen containing species. However, at = 1.0 and = 2.0, the n-butanol concentration is more sensitive to reactions involving

hydrocarbon radical species. This is because the pool of hydrocarbon radicals becomes more predominant as the fuel concentration in the oxygen–fuel mixture increases. Of all the n-butanol consumption reactions, the n-butanol concentration is most sensitive to those involving H-abstraction by OH radicals from the α and γ carbons.

Full-size image (44K)

Fig. 5. Sensitivity of n-butanol and propene to select reactions in the JSR at P = 10 atm, τ = 0.7 s.

View Within Article

Propene concentrations were not well predicted at = 1.0 and = 2.0. Figure 5b displays the

normalized sensitivity coefficients for the top 11 reactions to which the propene concentration is sensitive at T = 1000 K and all equivalence ratios. The propene concentration is sensitive to elementary reactions between hydrogen and oxygen containing species, as well as reactions involving small molecular weight hydrocarbon species. In addition, the propene concentration is sensitive to n-butanol consumption reactions involving H-abstraction from the α, β, and γ carbons.

A sensitivity analysis on acetylene (not in figure) indicated the acetylene concentration is sensitive to reactions involving the C2H3 radical, and to elementary reactions between hydrogen and oxygen containing species. Adjusting the reaction rates of n-butanol consumption reactions had

little effect on the concentration of acetylene.

5. Conclusions

New experimental data for n-butanol oxidation in a JSR at 10 atm and equivalence ratios between 0.5 and 2.0 are compared to a novel chemical kinetic model for n-butanol oxidation. The most abundant measured product species were carbon monoxide, carbon dioxide, water, hydrogen, methane, formaldehyde, ethylene, and propene. Measured in lesser amounts were butyraldehyde, 1-butene, acetaldehyde, ethane, and acetylene. The model proposed herein provides good overall agreement with the experimental data obtained across various temperatures and equivalence ratios. It is shown that H-abstraction is the major pathway of n-butanol consumption in the JSR, while unimolecular decomposition is relatively insignificant. Further model validations are still needed; they are awaiting the availability of ongoing flame measurements.

Acknowledgment

This research acknowledges funding from NSERC.

References

[1] D.T. Jones and D.R. Woods, Microbiol. Rev. 50 (4) (1986), pp. 484–524. View Record in Scopus | Cited By in Scopus (326)

[2] T.C. Ezeji, N. Qureshi and H.P. Blaschek, Curr. Opin. Biotechnol. 18 (2007), pp. 220–227.

Article | PDF (327 K) | View Record in Scopus | Cited By in Scopus (79)

[3] D. Ramey, S. Yang, Production of Butyric Acid and Butanol from Biomass, Report No. DE-F-G02-00ER86106. US Department of Energy, Morgantown, Washington, 2004.

[4] G. Hess, Chem. Eng. News 84 (26) (2006), p. 9. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (7)

[5] G. Hess, Chem. Eng. News 85 (27) (2007), p. 8. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (1)

[6] Y. Yacoub, R. Bata and M. Gautam, Proc. Inst. Mech. Eng. 212 (1998), pp. 363–379. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (15)

[7] G.L. Miller, J.L. Smith and J.P. Workman, Trans. ASAE 24 (1981), pp. 538–540.

[8] T.K. Bhattacharya, S. Chatterjee and T.N. Mishra, Appl. Eng. Agric. 20 (3) (2004), pp. 253–257. View Record in Scopus | Cited By in Scopus (10)

[9] J.A. Barnard, Trans. Faraday Soc. 53 (11) (1957), pp. 1423–1430. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (11)

[10] J.A. Barnard, Trans. Faraday Soc. 55 (6) (1959), pp. 947–951. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (4)

[11] A.F. Roberts, J. Imp. Coll. Chem. Eng. Soc. 12 (1959), pp. 58–73.

[12] C.S. McEnally and L.D. Pfefferle, Proc. Combust. Inst. 30 (2005), pp. 1363–1370. Article |

PDF (389 K) | View Record in Scopus | Cited By in Scopus (41)

[13] B. Yang, P. Oswald and Y. Li et al., Combust. Flame 148 (2007), pp. 198–209. Article | PDF (1138 K) | View Record in Scopus | Cited By in Scopus (54)

[14] P. Dagaut and M. Cathonnet, Prog. Energy Combust. Sci. 32 (2006), pp. 48–92. Article |

PDF (1414 K) | View Record in Scopus | Cited By in Scopus (77)

[15] P. Dagaut and S. Gail, J. Phys. Chem. A 111 (19) (2007), pp. 3992–4000. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (27)

[16] P. Glarborg, R.J. Kee, J.F. Grcar, J.A. Miller, PSR: A FORTRAN Program for Modeling Well-Stirred Reactors, Report No. SAND86-8209, Sandia National Laboratories, Livermore, CA, 1986.

[17] T. Le Cong, P. Dagaut, Kinetics of natural gas, natural gas/syngas mixtures oxidation and effect of burnt gas recirculation: experimental and detailed modeling, in: Proceedings of the ASME Turbo Expo 2007: Power for Land, Sea and Air, Montréal, Canada, May 14–17, 2007, GT2007-27146, pp. 1–9.

[18] P. Dagaut, Phys. Chem. Chem. Phys. 4 (2002), pp. 2079–2094. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (100)

[19] F. Battin-Leclerc, R. Bounaceur, G.M. C?me, et al., EXGAS-ALKANES, a software for the automatic generation of mechanisms for the oxidation of alkanes, 2004, CNRS-DCPR.

[20] Y. Tan, P. Dagaut, M. Cathonnet and J.C. Boettner, Combust. Sci. Technol. 102 (1994), pp. 21–55. Full Text via CrossRef

[21] C. Muller, V. Michel, G. Scacchi and G.M. C?me, J. Chim. Phys. Phys.-Chim. Biol. 92 (5) (1995), pp. 1154–1178.

Appendix A. Supplementary data

Download this File (85 K)

Supplementary data. The proposed n-butanol chemical kinetic mechanism in CHEMKIN format View Within Article

Download this File (160 K)

Supplementary data. The proposed n-butanol thermochemical data in CHEMKIN format

View Within Article

Download this File (515 K)

Supplementary data. An MS Excel data file with all the experimental and model data

View Within Article

Corresponding author. Fax: +1 416 978 7753.

Proceedings of the Combustion Institute

V olume 32, Issue 1, 2009, Pages 229-237