Hydrogen production from ethanol decomposition by TIAGO ... - ICPIG

4 downloads 274 Views 268KB Size Report
31st ICPIG, July 14-19, 2013, Granada, Spain ... 1 Laboratorio de Innovación en Plasmas, Edificio Einstein (C2), Campus de Rabanales, Universidad de.
31st ICPIG, July 14-19, 2013, Granada, Spain

16

Hydrogen production from ethanol decomposition by TIAGO torch discharge R. Rincón1 , J. Muñoz1 , M. Jiménez 1 , A. Marinas2, M. Sáez 1, M.D. Calzada1 P

P

UP

,

P

P

1

Laboratorio de Innovación en Plasmas, Edificio Einstein (C2), Campus de Rabanales, Universidad de Córdoba, 14071Córdoba, Spain 2 Dpto. Química Orgánica, Edificio Marie Curie (C3), Campus de Rabanales, Universidad de Córdoba. 14071 Córdoba, Spain

Argon TIAGO torch plasma in air ambience has been studied in order to produce the decomposition of ethanol molecules introduced into it and to obtain molecular hydrogen at the plasma exit. Dependence of both hydrogen yield and by-products obtained on Ar and ethanol flows has been studied. Ethanol molecules were completely decomposed and an important conversion to hydrogen was obtained. The main gaseous products generated were H2, H2O, CO, C2H2 and HCN when larger flows of Ar and EtOH were considered, whereas H2, H2O, and CO2 were formed during the ethanol reforming process with lower flows of either argon or ethanol. The hydrogen yield increased in the cases in which larger amounts of ethanol were added to Ar flow used as plasma gas.

1. Introduction Hydrogen is a promising option for the energy supply of fuel cells. However, only two major technologies: catalytic route through the water gas shift reaction and the electrolytic decomposition of water [1,2] to produce this fuel at large scale can be found. Nevertheless, the use of catalyst involves some shortcomings due to some catalyst lifetime, catalytic activity and stability improvements are needed. Therefore, a more economic and a environmental friendly alternative, such as plasma technology, has been investigated to avoid some conventional hydrogen production drawbacks. In the last decades, the number of researches involved in this concern has grown albeit microwave at atmospheric pressure plasmas has been pointed out as one of the most promising technologies due to some operational, stability or reproducibility advantages. In the literature, many researches devoted to produce hydrogen from reforming of alcohols or hydrocarbons [3-7] can be found. Nonetheless, in spite of several profits, microwave torches are relatively unexplored plasma sources with this purpose. In this work the microwave open to air TIAGO torch discharges [8,9] have been utilized to decompose ethanol which can be obtained by fermentation of surplus of agricultural residues and thus being a renewable hydrogen energy source. This research leads with the study of the experimental conditions influence on hydrogen generation and decomposition process. Furthermore, it has been taken advantage of helpful mass

spectrometry techniques to identify the produced gases and to quantify the produced molecular hydrogen in the reforming process. On the other hand, emission spectroscopy techniques have been used to understand the processes which take place within the discharge. 2. Experimental set up Part of the experimental set-up used to perform this research is detailed described in [9]. The discharge was generated inside a glass reactor equipped with 3 outflows. The coaxial and the lower lateral outflows were kept closed. At the exit of the upper lateral outflow a quadrupule Mass Spectrometer (QMS; Prisma, Pfeiffer Vacuum Technology) and previously calibrated in hydrogen was placed. It was used to analyze the composition of gas exiting the discharge after a filter which prevents solid carbon by-products from entering in Mass Spectrometer. In addition, the reactor was provided with a quartz window in order to be able to register discharge-emitted radiation within an interval from 200 to 750 nm, avoiding the absorption of UV-region radiation. The radiation was taken by an optical fibre and directed to the entrance of a 1-m Jobin-Yvon-Horiba monochromator (1000 M, Czerny-Turner type) a Symphony CCD (CCD-1024x256-OPEN-SITE) was used as a radiaton detector. To eliminate the influence of external fields, the reactor was isolated in a cylindrical Faraday cage. High purity (99.999%) Ar gas was used to initiate and feed the discharge with gas mass flow

31st ICPIG, July 14-19, 2013, Granada, Spain

(a)

0.06 NOγ

0.0010

Ar I (4p levels)

+

0.004

0.05

N2

CN

0.0005

0.002

Ar I

0.0000

0.04

0.000

I (a.u)

240

260

280

385

390

395



0.03

0.0005

CN 0.0000

0.02

OH NH

Ar I 655.5 (5p levels)

0.01

656.0

656.5

657.0

0.00 150

200

250

300

350

400

450

500

550

600

650

700

750

λ (nm)

(b)

1.0 0.12

CN

CN

0.09 0.06

0.8

NH

Fe I

0.02 C2 (Swan System)

OH 0.03

0.01

0.00 320

I (a.u)

3. Decomposition of ethanol in the plasma The species derived from the ethanol decomposition into the plasma can explain the formation of molecular substances at the plasma exit. The formation of these species is observed from the radiation emitted by the plasma, which was recorded by an optical fiber and leaded to the entrance of the monochromator, such as it appears described in Section 2. In Figure 1 emission spectra from Ar-EtOH discharges sustained with 300 W and Ar flow of 0.5 L/min and 0, 0.22 and 0.80 g/h as ethanol flows, are presented. It can be observed, when ethanol is introduced in Ar discharge the visible spectra emitted by the plasma undergoes several changes and, consequently, its internal kinetics. The emission lines of ArI system corresponding to 4p and 5p levels can be observed in Figure 1 (a). When the lowest ethanol flow is introduced (0.22 g/h, Figure 1 (b)) these ArI level are not observed which suggest that a part of plasma electrons collide with EtOH molecules instead of Ar atoms inducing dissociation of ethanol molecules and providing excited-stated of the formed radicals. As for molecular emission, CN (violet system) dominates the spectra in this condition. When greater amount of ethanol is introduced (0.80 g/h, Figure 1(c)) both, CN (violet system) and C2 (Swam system) emissions, rise. The arisen emission of C2 band is the most noticeable difference between both cases, which can explain the dissimilarities found in the species analyzed at the exit plasma (see Section 4). As has been previously addressed in CO2 microwave torches sustained at atmospheric pressure in controlled atmospheres [10], the formation of CN excimers, in those cases where the atmosphere contains nitrogen, is favoured respect to the formation of C2. In this sense, when a little quantity of ethanol is decomposed, nitrogen influences the discharge kinetics meanwhile with higher EtOH amounts, nitrogen from the

surrounding atmosphere seems to be less influential.

340

360

0.6

0.00 440

460

480

0.008

0.4

500

520



0.004

0.2

0.000 655.5

CI

656.0

656.5

657.0

0.0 150

200

250

300

350

400

450

500

550

600

650

700

750

λ (nm)

(c)

CN

1.0

C2 (Swan System)

NH

0.010 CH

0.8

0.005 0.010

0.000

I (a.u)

controllers (HI-TEC, Bronkhorst). Different Ar flows of 0.25, 0.50, 0.75, 1.00, 1.25 and 1.50 L/min (litres per minute) were used in this research. Ethanol gas flow was controlled using a gas phase liquid delivery system (CEM, Bronkorst) with 0.2010 ethanol g/h (grams per hour) upper and lower limits. Once vaporized in the gas phase liquid system, the ethanol is led to the plasma within a steel tube heated at 110 ºC to prevent condensation which could produce plasma extition. Ethanol flows were 0.22, 0.40, 0.60, 0.80 and 1.00 g/h and power feed to the discharge was kept equal to 300 W.

310

0.6

320

330



340 0.005

0.000 655.5

0.4

0.2

656.0

656.5

657.0

CI

0.0 150

200

250

300

350

400

450

500

550

600

650

700

750

λ (nm)

Figure 1. Emission spectra of discharges generated at 300 W, 0.50 L/min Ar flow and (a) 0 g/h, (b) 0.22 g/h and (c) 0.80 g/h as ethanol flow.

4. Species at the plasma exit The composition of gas exiting the discharge was analyzed by a mass spectrometer.

31st ICPIG, July 14-19, 2013, Granada, Spain

In Figure 2, intensities normalized to Ar atom (m/z=40 a.m.u.) mass spectra in the 0-80 a.m.u. range of the gas exhaust stream of Ar and Ar-EtOH discharges generated with Ar flows of 0.25 (a) and 0.50 (b) L/min are shown. In all cases, the total decomposition process of EtOH leads to H2 (m/z=2 a.m.u.) formation, however, different by-products were obtained depending on the Ar flow rates.

(a)

1

H2

Plasma Ar ON Plasma Ar-EtOH ON P=300W Fl(Ar)=0.25 L/min Fl(EtOH)=0.80 g/h

H2O

0.1

I (a.u)

0.01

CO2

1E-3

1E-4

1E-5

1E-6 0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

m/z (a.m.u.)

(b)

1

H2

0.1

H2O C H 2 2

0.01

I (A)

plasma Ar ON plasma Ar-EtOH ON P=300 W Fl(Ar) = 0.50 L/min Fl(EtOH) = 0.80 g/h

CO HCN

1E-3

1E-4

C4H2, C4H4 1E-5

appearance due to the oxygen presence into the discharge. On the other hand, when Ar flow is increased the entrance of oxygen is not as possible as nitrogen one because of less concentration of O2, in relation to N2, in the air. Hence, flow HCN is found in the exhaust stream gas of discharges generated with larger Ar. As far as the toxic HCN formation is concerned, it may be formed via CN reacting with H [11] or hydrocarbons. One could think about the contribution of N2O molecule to the peak at 44 a.m.u. in the case of discharges sustained with small flow of Ar due to the entrance of N2. Nevertheless, the 45/46/22 relative intensity ratios are ca. 100/1.3/1.2, which is typical of CO2. Furthermore, peak at m/z=28 a.m.u. could be ascribed to the existence of ethylene in the exhaust gas because its typical mass spectra includes less intense peaks at m/z=27, 26 and 25 a.m.u. However, relative intensities of 28/29, 26/25/24 and 27/26 are close to 100/1.1, 100/20/6 and 100/16.7 which can be certainly related to CO, C2H2 and HCN molecules, respectively. Following the above discussion about species identification, different amounts of ethanol led to different species at the plasma exit, with exception of H2 and H2O, which were obtained in all conditions. As the added ethanol flow became larger, the main gases changed from CO2 to C2H2, CO and HCN, as well as, C4H2, C4H4 and C6H6 as trace levels. It can be observed in the mass spectrum shown in Figure 3.

C6H6

0 g/h

0.22 g/h

0.40 g/h

0.60 g/h

0.80 g/h

1.00 g/h

1E-6

H2

-7

0

5

10

15 20 25

30 35

40 45 50

55 60 65

1x10

70 75 80

CO

m/z (a.m.u.)

C2H2

-8

1x10

When the discharge was generated with an Ar flow lower than 0.50 L/min, H2O (m/z=18 a.m.u.) and CO2 (m/=44 a.m.u) were produced (Figure 2 (a)). Nevertheless, in the case of Ar flows were 0.50 L/min or larger, unalike products were found: H2O, C2H2 (m/z=26 a.m.u.), CO (m/z=28 a.m.u.), HCN (m/z=27 a.m.u.) and C4H2, C4H4 and C6H6 (m/z=50, 52, 78 a.m.u, respectively) as trace levels (Figure 2(b)). An important dependence of the processes in the plasma on the Ar flow used to sustain the discharge was verified in [9]. When Ar flow decreases, air components such as O2 and N2 can penetrate into the discharge. In this case, the oxidation reaction of ethanol leads to CO2

HCN H2O

I (A)

Figure 2. Ar and Ar-EtOH plasmas at 300W mass spectra generated with Ar flow of (a) 0.25 and (b) 0.50 L/min keeping constant EtOH flow of 0.80 g/h.

-9

1x10

-10

CO2

-11

EtOH

1x10

10

0

20

40

60

80

100

120

140

160

180

t (min)

Figure 3. Change of the mass spectrum versus time for the six ethanol flows studied.

Figure 4 shows both experimental and theoretical hydrogen production for each studied experimental condition. In this figure, square symbols show the dependence of hydrogen production on ethanol flows.

31st ICPIG, July 14-19, 2013, Granada, Spain

0.16 0.14

experimental max. value

0.12

H2 (g/h)

0.10 0.08 0.06 0.04 0.02 0.00 0.0

0.2

0.4

0.6

0.8

1.0

1.2

Fl(EtOH) (g/h)

Figure 4. Hydrogen yield for different EtOH flows.

5. Conclusions In this work, the dependence of hydrogen yield and other by-products from the ethanol reforming by TIAGO torch has been studied. The air ambience that surrounds the discharge seems to have an important influence on the type of gases generated at the plasma exit. Besides, this research reveals the ethanol is decomposed almost in its totality into the plasma which leads to larger amounts of hydrogen; these amounts are increased with the ethanol flows . Acknowledgements This work was partially subsidised by the Spanish Ministry of Science and Technology within the framework of Project no. ENE2008-01015 (FEDER funds) and by the Andalusia Regional Council (Consejería de Economía e Innovación) under Project no. FQM-7489.

6. References [1] M.S. Yaghmaee, B. Shorik, N.H. Khiabani, A.Sarani, Plasma Processes and Polymers 6 (2009) 631. [2] M. Mlotek, j. Sentek, K. Krawczyk, K. Schmidt-Szalowski, Applied Catalysis a-general 366 (2009) 232 [4] M. Jiménez, C. Yubero, M.D. Calzada. J. Phys. D Appl. Phys. 41 (2008) 175201. [5] J. Henriques, N. Bundaleska, E. Tatarova, F.M. Dias, C.M. Ferreira. Int. J. Hydrogen Energ. 36 (2011) 345. [6] M. Mora, M.C. García, C. JiménezSanchidrián, F.J. Romero-Salguero. Plasma Process. Polym. 8 (2011) 709. [7] Y.F. Wang, Y.S You, C.H. Tsai, L.C. Wang. Int. J. Hydrogen Energ. 35 (2010) 9637. [8] M. Moisan, Z. Zakrewski, J.C. Rostaing, Plasma Sources Sci. Technol. 10 (2001) 387. [9] R. Rincón, J. Muñoz, M. Sáez, M.D. Calzada. Spectrochim. Acta Part B 81 (2013) 26. [10] E.A.H. Timmermans, J. Jonkers, I.A.J. Thomas, A. Rodero, M.C. Quintero, A. Sola, A. Gamero, J.A.M. van der Mullen. Spectrochim Acta B 53 (1998) 1553 [11] W. Tsang. J. Phys. Chem. Ref. Data 21 (1992) 753