Energy for Cleaner Transportation

K. ZaghibHydro-QuebecVarennes, Quebec, Canada

J. PrakashIllinois Institute of TechnologyNaperville, Illinois, USA

R. D. McConnellNational Renewable Energy LaboratoryLakewood, Colorado, USA

F. R. McLarnonLawrence Berkley National LaboratoryBerkeley, California, USA

Editors:

Engery Technology

Sponsoring Divisions:

Vol. 1 No. 14ISBN 1-56677-495-0

Published by

The Electrochemical Society65 South Main Street, Building DPennington, NJ 08534-2839, USA

tel 609 737 1902fax 609 737 2743

www.electrochem.org

TM

Battery

Copyright 2006 by The Electrochemical Society, Inc.All rights reserved.

This book has been registered with Copyright Clearance Center, Inc. For further information, please contact the Copyright Clearance Center,

Salem, Massachusetts.

Published by:

The Electrochemical Society, Inc.65 South Main Street

Pennington, New Jersey 08534-2839, USA

Telephone 609.737.1902Fax 609.737.2743

e-mail: ecs@electrochem.orgWeb: www.electrochem.org

ISBN 1-56677-495-0

Printed in the United States of America

iii 

Preface 

Energy for Cleaner Transportation 

This symposium covered a wide range of topics associated with power sources for hybrid electric cars. Major emphasis was on the research, development, and applications of lithium–ion batteries lithium polymer batteries, polymer electrolyte fuel cells and capacitors. The battery  papers were presented in sessions on a. performance testing and evaluation, b. diagnostics of failure modes, and c. low cost components. The fuel cell related papers emphasize modeling, testing, and performance. 

Appreciation is extended to the authors for their efforts in preparation and presentation of their manuscripts. Also, we take this opportunity to thank the chairmen of sessions, the speakers, and attendees, for making this an interesting and stimulating event. 

K. Zaghib, J. Prakash, R. McConnell, and F. McLarnon

v 

ECS Transactions, Volume 1, Issue 14 Energy for Cleaner Transportation 

Table of Contents 

Preface  iii 

Electronic and Magnetic Properties of Carbon­Coated Lithium Iron Phosphates C. M. Julien, K. Zaghib, A. Ait Salah, A. Mauger, and F. Gendron 

1 

Synthesis of Birnessite­Type Layered Manganese Oxide and their Composites for Supercapacitors 

F. Favier, Y. Zhou, C. Fournier, G. Dorval­Douville, S. Shanley, D. Jones, B. Mula, J. Roziere, T. Brousse, and D. Belanger 

9 

The Influence of Binders on the Performances of Ni(OH)2 Positive Electrode in Supercapacitor W. Yu, X. Yang, P. Wang, and L. Meng 

19 

Rotating Rate Dependency of Methanol Oxidation on a Smooth Polycrystalline Platinum Surface G. Hou and J. Prakash 

27 

The Effect on In­Cell Measurements from Water and Methanol Transport in Direct Methanol Proton Exchange Membrane Fuel Cells 

M. Lefebvre and D. Olmeijer 

35 

Hydrogen Production from Mill­Scale via Metal­Steam Reforming S. Kesavan and A. Azad 

43 

Hybrid Supercapacitors with Ionic Liquid Electrolytes M. Mastragostino, C. Arbizzani, A. Balducci, S. Passerini, P. Simon, and F. Soavi 

55 

Electrochemical Properties Carbon Derived from Sol­Gel Catechol­Formaldehyde Gels R. Nagireddy and R. Reddy 

61 

Ionic Liquid Based Electrolytes for High Energy Electrochemical Storage Devices S. Passerini, F. Alessandrini, G. B. Appetecchi, and M. Conte 

67 

Author Index  73

HYDROGEN PRODUCTION FROM MILL-SCALE VIA METAL-STEAM

REFORMING

Sathees Kumar Kesavan* and Abdul-Majeed Azad

Department of Chemical and Environmental Engineering The University of Toledo, Toledo, OH 43606, USA.

ABSTRACT

Mill-scale wastes from steel industry have been successfully reduced to pure metallic iron via hydrogen and carbothermic reduction. Use of high temperature and H2 gas has been eliminated by a novel solution-based room temperature reduction technique whereby nanoscale iron powder is produced. This new scheme totally obviates the issue of sintering of the iron/iron oxide and hence the deactivation during the cyclic operation of metal-steam reforming. Some of the preliminary results of this investigation are discussed.

INTRODUCTION

Hydrogen, though not a fundamental energy source, has been touted to be the ideal energy-carrying medium for fuel cells, particularly the PEMFCs. However, it is not as abundant a commodity in nature as oxygen and nitrogen and must be derived from other natural resources such as fossil fuels or water. The question of its storage and supply arises only after it is produced in large enough quantity so as to cater to the need of its use. In the light of the serious economic constraints and safety concerns associated with production, storage and supply of gaseous hydrogen, one needs to find ways and means of generating, storing, transporting and supplying hydrogen to the end use port - in more reversible, much simpler and far safer ways.

An economically viable and environment friendly method of generating hydrogen is by the reaction of certain metals with steam, also appropriately called ‘steam-metal reforming’. For example, the reaction,

3 Fe + 4 H2O → Fe

3O

4 + 4 H

2 [1]

which occurs around 600ºC, has long been known and has recently been revived as one of the promising ways of generating hydrogen (1-3); the reverse reaction could be viewed as a hydrogen storage scheme. The theoretical amount of hydrogen being stored is 4.8 wt% which corresponds to ca. 4211 L H2/L Fe at STP. However, the kinetic of metal oxidation (forward) and oxide reduction (reverse) in a cyclic fashion has to be significantly improved, in order to mitigate sintering and coarsening of iron and iron oxide particles (4-5) during repeated redox cycles.

We have developed an effective technique of producing H2 by using the so-called ‘mill-scale’ waste from steel industry, as an iron source, in a way consistent with the criteria listed above: environment, availability and price. The mill-scale is a hard brittle coating of several distinct layers of iron oxides formed during the processing of steel. It is

ECS Transactions, 1 (14) 43-53 (2006)10.1149/1.2214609, copyright The Electrochemical Society

43

44

predominantly Fe3O

4 and magnetic in nature with iron content ranging between 66 to

88% depending upon the source and conditions employed for its stripping from the steel.

A perceived design for the generation of hydrogen via metal-steam reforming consists of the following. Iron derived from mill-scale is packed into cartridges and loaded on the vehicles. Addition of water to the pre-heated cartridges produces pure humidified hydrogen that could be supplied directly to PEMFC on the vehicles. The cartridges with corresponding oxides after water decomposition are exchanged for the new ones packed with fresh metal; the oxide is again reduced to ready-to-use active metal later. Thus, the process of hydrogen generation from steel industry’s waste in an efficient and cyclic fashion consists of: (a) conversion of magnetite to elemental iron, (b) iron-steam reaction and, (c) reduction of the spent iron oxide into active elemental iron for the next cycle.

Along these lines, we have recently carried out a systematic and thorough investigation on steel waste material to establish the feasibility of hydrogen production and recycling of the oxide on a laboratory scale. It was found that the mill-scale samples from different vendors could be quantitatively (~100%) reduced to elemental iron, both via H2 and carbothermic reduction. With an objective to eliminate the use of high temperature and H2 gas and to mitigate the CO/CO2 emission, a solution-based technique has been devised resulting in the formation of nanoscale elemental iron particles from mill-scale powder at room temperature, which could be used directly in metal-steam reforming process to produce H2. Metal-steam reforming experiments carried out using active iron powder derived from mill-scale magnetite waste showed conversion of water into hydrogen as per Eq. 1 was near stoichiometric (≥ 95%). This paper discusses the methodology and the results of the H2 generation via M-steam reaction using steel waste.

EXPERIMENTAL

Sample processing Five mill-scale samples were procured from three different vendors: North American Steel (KY), Midrex (NC) and Nucor-Yamato (AR). The samples from North American Steel (NAS) were labeled as pickled, sludge rinse and entry loop by the supplier. Samples from Midrex and Nucor-Yamato are named MR and NY, respectively, for brevity. All the five mill-scale samples were attrition-milled in 2-propanol and sieved through 325 mesh (average particle size≤ 45 µm) prior to any processing or characterization. Oxide reduction Hydrogen reduction was conducted in high purity dry H2 at 1173K for 8h. The total pressure of the gas flowing through the fixed bed of mill scale samples was maintained at ~101 kPa. For the carbothermic reduction, the processed mill-scale powders were mixed with activated carbon in the molar ratio of 1:3, ball-milled as above, dried and pelletized after adding small amount of aqueous polyvinyl alcohol (PVA) solution as a binder to aid compaction. The carbothermic reaction (6) carried out at 1373K for 4h, using nitrogen as a blanket gas can be represented as:

Fe3O4 + 3C → 3Fe + 2CO + CO2 [2] Metal-steam reforming reaction Metal-steam reforming was done in a stainless steel tubular reactor (ID ~ 0.68 in.). A thermocouple placed in sample’s proximity monitored the temperature. Helium was used

ECS Transactions, 1 (14) 43-53 (2006)

45

as an inert background during the temperature ramp to 873 K. After the sample attained the final temperature, a 50:50 v/v mixture of steam (water preheated to 673 K) and He was introduced while maintaining the total pressure ~101 KPa. The reaction was allowed to run until no hydrogen formation was detected in the mass spectrometer attached to the reactor at the outlet. Characterization Chemical analysis of the as-received mill scale samples was done by X-ray fluorescence (XRF) technique. Magnetic measurements were carried out by Mössbauer spectroscopy while X-ray powder diffraction patterns were collected at room temperature on a Philips diffractometer (PW 3050/60 X’pert Pro), using monochromatic CuKα1 radiations (λ = 1.54056 Ǻ) and Ni filter. Morphology of the samples before and after reduction and steam reforming were examined by a Philips scanning electron microscope (XL30 FEG). The morphology and crystal structure of the nano-iron particles were studied by transmission electron microscopy (JEOL 3010) operated at 300 kV together with selected area electron diffraction (SAED). Hydrogen formation was quantified by quadrupole mass spectrometer (QMS 200, Pfeiffer Vacuum Omnistar).

RESULTS AND DISCUSSION

Table I shows the XRF analyses of the as-received mill-scale samples from various vendors. Two of the three NAS samples were predominantly rich in Cr (52 to 67% by weight), while the third sample contained about 18% Cr and 8% Ni by weight. On the other hand, MR and NY samples contained 92-94 wt% Fe and no significant Cr or Ni.

Table I. XRF analyses of the mill scale samples used in this study. Wt %

M Pickled Sludge

rinse Entry loop

Midrex Nucor-Yamato Fe 29.13 42.3 70.42 93.92 92.91 Cr 66.57 51.59 18.11 0.16 0.168

Mn 2.75 2.49 1.4 1.19 1.3

Ni 0.178 0.872 8.24 0.232 0.189

Ca - 0.128 - 1.33 3.14 Si 0.332 0.267 0.357 1.49 1.05 Mo 0.282 1.3 0.53 - - Cu - 0.337 0.366 0.608 0.515

V 0.412 0.259 - - -

The presence of impurities (especially Cr) in significant amounts has serious implications on the environmental aspects of the process. If the NAS mill-scales were to be used for hydrogen production, one has to isolate Fe which means disposing nickel and hexavalent chromium, the latter being a known health hazard in water system. Preliminary reduction experiments failed to generate elemental iron from NAS samples (since Ni and Cr-ferrites are difficult to breakdown), as revealed by the XRD (Fig. 1) and gravimetry. Hence, they were not used in subsequent experiments. On the other hand,

ECS Transactions, 1 (14) 43-53 (2006)

46

hydrogen and carbothermic reduction of MR and NY samples resulted in complete conversion of the scales into elemental iron. Figure 2 shows the XRD signatures of as-received and reduced samples from Midrex and Nucor-Yamato, which corroborates the above statement. Gravimetric analyses of the samples before and after reduction gave an average weight loss of 27.5 wt% (hydrogen reduction) and 27 wt% (carbothermic reduction) for the Midrex and Nucor-Yamato samples, respectively; this agrees very well with the theoretical value of 27.64 wt% for the Fe3O4 to Fe reduction.

FIGURE 1. XRD spectra of the as-received and reduced mill-scale samples from NAS.

FIGURE 2. XRD patterns of the as-received and reduced MR and NY samples.

ECS Transactions, 1 (14) 43-53 (2006)

47

The morphology of the mill-scales before and after hydrogen reduction is compared in Fig. 3.

a1 a2

b1 b2

c1 c2

ECS Transactions, 1 (14) 43-53 (2006)

48

FIGURE 3. SEM features of the mill-scale samples: (a) Entry loop (b) Pickled (c) Sludge rinse (d)

MR (e) NY. 1- before and 2- after H2 reduction.

Figure 4 shows the micro structural features of Midrex and Nucor-Yamato samples after carbothermic reduction.

FIGURE 4. Morphological features of carbothermically reduced samples: (a) MR (b) NY. As mentioned earlier, regeneration of elemental iron from the spent oxide via hydrogen or carbothermic reduction is an energy-intensive process and hence undesirable

d1 d2

e1 e2

a b

ECS Transactions, 1 (14) 43-53 (2006)

49

in a commercial setting. To obviate these commercial predicaments the Midrex and Nucor-Yamato samples were digested in mineral acid and treated with aqueous sodium borohydride solution in the presence of sodium hydroxide as pH stabilizer (pH ~6.8) (7). This resulted in immediate precipitation of nanoscale iron particles – a clear advantage over the agglomeration seen during hydrogen and carbothermic reduction. Figure 5 and 6 shows the morphology of the so-formed iron particles from Midrex and Nucor-Yamato samples, respectively.

FIGURE 5. SEM micrographs of solution-derived iron nanoparticles from Midrex sample.

.

FIGURE 6. SEM features of solution-derived iron nanoparticles from Nucor-Yamato.

Figure 7 shows the TEM images and Figure 8 shows the EDS spectrum of the nanoscale iron particles from Midrex; Cu and Mo traces in the EDS originate from the Cu-Mo grid used for the TEM imaging. The SAED pattern of the nanoscale iron is showed in Figure 9. The diffraction rings belong to the (110), (200) and (211) planes of bcc iron.

Figure 10 compares the results of metal-steam reforming (MSR) carried out with carbothermically and H2 reduced NY sample and the nano-iron obtained via aqueous reduction. It can be seen that for a 90% conversion value, higher activity with regard to H2 generation is observed for nano-Fe compared to that derived from the other two techniques. It is interesting, however, to note that irrespective of the source from which

ECS Transactions, 1 (14) 43-53 (2006)

50

iron for MSR was acquired, the magnetite formed after the steam reformation can be dissolved in mineral acids from which highly active nano-iron is again recovered.

FIGURE 7. TEM images of iron particles obtained from steel industry waste by room

temperature processing. Scale bar: 20 nm (top and right) and100 nm (left). Energy dispersive X-Ray

0

40

80

120

160

0 500 1000 1500 2000 2500 3000 3500

eV

intensity

Mo

Fe

Cu

Mo

O

FIGURE 8. EDS pattern of the solution-derived nano iron.

ECS Transactions, 1 (14) 43-53 (2006)

51

FIGURE 9. Selective area electron diffraction pattern of the solution-derived nano iron.

FIGURE 10. Amount of hydrogen formation vs. time from Nucor-Yamato samples.

ECS Transactions, 1 (14) 43-53 (2006)

52

Fig 11 shows the SEM pictures of the NY sample before and after metal-steam reforming reaction using Fe from the three reduction techniques.

FIGURE 11. Morphological artifacts of the NY samples before (1) and after (2) MSR using Fe from: (a) H2 (b) carbothermic and (c) aqueous reduction schemes.

a1 a2

b1 b2

c1 c2

ECS Transactions, 1 (14) 43-53 (2006)

53

CONCLUSIONS

Metal-steam reforming using iron is a safe and economically viable method of hydrogen generation from inexpensive raw materials for PEMFC utilization. An attractive and inexpensive source of iron is the waste from the steel industry in the form of magnetite, provided it can be reduced to metallic iron. In this work we have shown that, two of the five mill-scale samples from commercial vendors were quantitatively reduced to iron via hydrogen reduction, and carbothermic processes at high temperatures. A novel solution-based technique using sodium borohydride as the reductant yielded nanoscale iron with an average particle size of ~40 nm. Optimization of the latter process also obviated the formation of sodium borate as a by-product and hence the need for its separation. The hydrogen production via metal-steam reforming reaction represented by Eq. (1) shows that the nano iron possesses a higher propensity of reaction with steam than that derived via either of the high temperature reduction processes adopted here, under identical experimental conditions. Furthermore, the spent oxide can be quantitatively recycled to yield highly active nanoscale metallic iron again by an acid dissolution route without the loss of its reactivity, thereby maintaining the hydrogen generation capacity of the starting material over several cycles. Work is in progress for the search of an alternative reductant in lieu of sodium borohydride.

ACKNOWLEDGEMENTS

The financial support provided by the URAF and the Department of Chemical and Environmental Engineering at the University of Toledo for this work is gratefully acknowledged. We also thank Dr. Martin Abraham and his group for allowing us to run the Metal-steam reforming reactions in his laboratory at the University of Toledo. Authors also wish to thank Dr. John Mansfield and his colleagues at the EMAL center of University of Michigan, Ann Arbor for allowing us to use the scanning and tunneling electron microscope.

REFERENCES

1. K. Otsuka, A. Mito, S. Takenaka, I. Yamanaka, Int. J. Hydrogen Energy, 26 (2001) 191.

2. K. Otsuka, C. Yamada, T. Kaburagi, S. Takenaka, in: R.D. Venter, T.K. Bose, N. Veziroglu (Eds.), Proceedings of the 14th World Hydrogen Energy Conference, Montreal, Canada, Session Al.9–Storage and Material Systems I, CD-ROM, 2002..

3. K. Otsuka, T. Kaburagi, C. Yamada and S. Takenaka, J. Power Sources, 122 (2003) 111- 121

4. K. Otsuka, C. Yamada, T. Kaburagi, S. Takenaka, Int. J. Hydrogen Energy 28 (2002) 335.

5. S. Takenaka, T. Kaburagi, C. Yamada, K. Nomura and K. Otsuka, J. Catal. 228 (2004) 66- 74.

6. D.M.D. Santos and M.D. Maurao, Scand. J. Metall. 33 (2004) 229-235. 7. Glyn D. Forster, Luis Fernandez Barquin b, Quentin A. Pankhurst, Ivan P. Parkin,

J. Non-Crystalline solids, 244 (1999) 44-54.

ECS Transactions, 1 (14) 43-53 (2006)