PRO

CEE

DIN

GS

2010

EM

ISSI

ON

S C

ON

FERE

NC

E &

EXPO

Volu

me

55

Proceedingsemissions 2010June 15-16, 2010

Volume 55

© 2010 Global Automotive Management Council ISSN 1543-9658

Proceedings of the 2010 Emissions Conference & Exposition

Editors

John Hoard

University of Michigan

M. Nasim Uddin

Global Automotive Council

Assistant Editors

Tom Tang

Rafiq Uddin

Powertrain International Journal

June 15-16, 2010

Emissions 2010 5305 Plymouth Rd.

Ann Arbor, Michigan

P 734.418.2365

F 734.418.2356

www.gamcinc.org

Volume 55

© 2010 Global Automotive Management Council

Programs Leaders and Chairs

Levi Thompson, PhD

University of Michigan

Future Mobility - Energy, Environment and Decarbonization

Dr. Levi Thompson earned his B.ChE. from the University of Delaware, and M.S.E. degrees in Chemical

Engineering and Nuclear Engineering, and a PhD in Chemical Engineering from the University of

Michigan. Dr. Thompson has established a strong record of service at all levels. At the national level, he

is Consulting Editor for the AIChE Journal, a member of the Advisory Committee for the NSF Engineering

Directorate, serves on the External Advisory Committee of the NSF STC for Advanced Materials for

Water Purification at the University of Illinois.

John Hoard

University of Michigan

Fuel Efficiency and Powertrain Optimiziation

John Hoard holds a BS in Engineering from UCLA (1970). He spent 37 years in the Ford Motor Company

Research Laboratory, working on projects including exhaust aftertreatment, engine-out emission control,

powertrain controls systems, and fuel efficiency. Recent work included diesel engine emissions control,

diesel catalyst systems (LNT, SCR, DOC, DPF), and EGR cooler fouling.

Harold Sun, PhD

Ford

Emissions - Improvements, Standards, & Controls

Dr. Harold Sun is currently a Technical Expert at Research and Innovation Center, Ford Motor Company,

responsible for boost system integration and development for diesel engine applications. He was a

Technical Specialist at Cummins, responsible for ISC/ISL diesel combustion, performance and emission

development, prior joining Ford in 1999. Dr. Sun holds a Ph.D. from Wayne State University (1993), MS

from Tianjin University (1987) and BS from Hefei University of Technology (1982). He holds 8 patents.

Tom McKarns

ECO PHYSICS

Meeting New EPA NOx Limits, Standards, and Procedures

Miami University, Oxford, Ohio 1976 with a degree in Chemistry and Zoology. Employed in the scientific

products industry over the past 30 plus years, including 16 years with Mettler-Toledo in the U.S. and

Switzerland (1981- 1995) and finally settling in Ann Arbor where I started my business of marketing the

products of Switzerland’s ECO PHYSICS AG (air emission measurements and medical research).

Volume 55

© 2010 Global Automotive Management Council

Table of Contents

Mitigation of Greenhouse Gases via Utilization

Abdul-Majeed Azad, Sirhan Al-Batty □ University of Toledo

Optimization of a Turbocharger for High EGR Applications

Harold Sun, Dave Hanna □ Ford, Eric Krivitzky, Nick Baines, Louis Larosiliere, Ron Forni □ ConceptsNRE,

Liangjun Hu, Jizhong Zhang, Ming-Chia Lai □ Wayne State University

Enhancement of Turbocharged Inter-cooling Diesel Engine Performance by Hydrous Ethanol Port

Injection

Chunde Yao, Xuping Chen, Qi Xia, Xianglan Yang, Yuanli Xu, Junheng Liu □ State Key Laboratory of

Engines, Tianjin University

Development and Performance Assessment of a Multi-Functional Reactor Under Conventional and

Advanced Combustion Diesel Engine Exhaust Conditions

Dimitrios Zarvalis, Nickolas Vlachos, Souzana Lorentzou □ Aerosol & Particle Technology Laboratory,

Athanasios Konstandopoulos □ Department of Chemical Engineering, Aristotle University, Jacques Javy,

Stephane Zinola □ Energy Applications Techniques Division, Engines Laboratory Department, Institut

Francais du Petrole (IFP)

Integrated Design of Powertrain Emission Systems: A Modeling Study

Santhoji Katare, Eric Kurtz, Jeong Kim, Paul Laing □ Ford

Improvement on Emissions of Diesel Engine with Methanol Fumigation

Chunde Yao, Qi Xia, Xianglan Yang, Xuping Chen, Lijiang Wei, Yuanli Xu □ State Key Laboratory of

Engines, Tianjin University

Nonroad Small SI Exhaust and Evaporative Emissions Standards

Julia Giuliano, Sara Zaremski □ EPA

Refueling Emission Control

Sam Reddy □ Evaporative Emissions Consulting, Inc.

Experimental and Numerical Investigation Of Vaned Diffuser’s Effect on Turbocharger Compressor

Performance

Liangjun Hu, Ce Yang □ Beijing Institute of Technology, Eric Krivitzky, Ronald Forni □ ConceptsNREC,

Jizhong Zhang □ National Key Laboratory of Diesel Engine Turbocharging Technology, Harold Sun □

Ford, Mingchia Lai □ Wayne State University

Improving Results of Engine Emissions Tests: Accuracy Counts!

Stephen Miller □ Air Liquide

The Future is Driven by Structurally Optimized Transmissions

Jörg Müller, Mirko Leesch, Rico Resch, Tom Tibbles □ IAV

1

Mitigation of Greenhouse Gases via Utilization

Abdul-Majeed Azad1 and Sirhan Al-Batty

Chemical Engineering Department, The University of Toledo 2801 W. Bancroft St., Toledo, OH 43606-3390, USA.

Abstract

The ultimate chemical fate of the conventional fossil fuel combustion is always CO2 and H2O, two well-known greenhouse gases responsible for contributing considerably to the global warming. We propose a radically new concept where CO2 and H2O are converted into carbon monoxide and hydrogen (individually) at one hand and into syngas (if present together – a more likely scenario) at the other, under mild experimental conditions. The products then become practical feeds for generating clean electric power via solid oxide fuel cells. Syngas is also an ideal precursor for the manufacturing of scores of valuable organic compounds including synthetic fuels by Fischer-Tropsch process. Thus, the technique promotes beneficiation rather than sequestration and storage - currently the most widely accepted option for addressing the issue of mitigating the CO2-related greenhouse gas emission. The process employs a ferrous metal or its oxide for the target-specific conversion. In addition to being inexpensive and readily available commodities, the oxide in particular happens to be an industrial waste as well; most of it is landfill destined. Furthermore, while the two greenhouse gases are reduced into ready-to-use fuel precursors, the solid reactants are transformed into a component from which ceramic magnets (ferrites) could be manufactured. Thus, the concept is innovative since it not only creates valuable clean energy precursors, it does so while reducing the impact of yet another industrial and ecological pollutant. The spent material is also regenerable by reduction via carbothermic route or with gasified low quality biomass. Keywords: Global warming; greenhouse gas effect, carbon dioxide, water vapor, Fischer-Tropsch synthesis; iron; magnetite; maghemite; fuel cells.

1 Corresponding author; +1 419 530 8103; abdul-majeed.azad@utoledo.edu

2

Introduction

Global warming, defined as the increase in the average temperature of earth’s

atmosphere, is believed to have already changed earth’s climate irreversibly as evidenced

by the rising sea levels, glacier retreat, arctic shrinkage, rapidly altering agricultural

patterns, unseasonably cold and warm weathers, expansion of tropical diseases and

explosion of forest fires. This poses a great threat to the quality of life on this planet.

Irrespective of the origin and modality of their use, the ultimate chemical fate of the

conventional fossil fuel combustion is always CO2 and H2O. Both these species are well

known greenhouse gases and are responsible for contributing considerably to the global

warming - a subject of contentious political, social and technical debate. As a result of

constant increase in their concentration and retention, scientific projections estimate that

the average global temperature will rise between 1.4°C and 5.8°C by the year 2100. In

2007, the global level of CO2 was 30 billion metric ton and is projected to be 43 billion

metric ton by 2030. The US contributes the largest - 22.2% - towards global CO2

emission [1-2].

It is less known that water vapor is a naturally occurring greenhouse gas and accounts for

the largest percentage of the greenhouse effect - between 36 and 66% in terms of

radiation absorbance. Water vapor concentrations fluctuate regionally, though unlike in

the case of carbon dioxide, human activity does not directly affect water vapor

concentrations except at local scales (for example, near irrigated fields). In terms of mass,

water vapor is ~80% of all greenhouse gases by mass (~90% by volume). However,

radiative importance of water vapor is less than that of CO2, CH4 and N2O. Moreover, the

3

lifetime of water vapor in the environment is about 9 days compared to hundreds of

thousands of years for CO2 [3].

The efforts to diminish the influence of carbon footprints on the way energy is produced

and used today, will be the most visible legacy of the technological developments of the

21st century. Both catalytic (using bi-or tri-metal supported alumina matrices) and non-

catalytic schemes of direct thermal cracking of natural gas (methane) leading to CO2-free

H2 generation at high temperatures have been demonstrated on a laboratory scale [4-5];

several non-fossil approaches are also being actively pursued. One of those alternative

pathways is biofuels, but even biofuels including those derived from waste biomass

involve the logistic burden of and energy for harvesting, processing, separation,

purification and delivery. However, their use too ultimately leads to the generation and

accumulation of carbon dioxide and water in the atmosphere.

In the mean time solutions must evolve without the problems associated with greenhouse

gases, using nation’s available energy resources. The two, however are not mutually

exclusive for no fossil fuel usage could avoid greenhouse gas generation. Hence, it is

relevant and prudent to have a technology that separates CO2 in the flue gas stream

without expending additional energy. The resulting CO2 is ready for compression and

transport to storage facility.

There is a frantic race to control, combat and counter the menace of global warming,

albeit with very limited success. For example, the National Energy Technological

4

Laboratory under the auspices of the Department of Energy has been working to develop

a range of carbon capture and storage technologies [6]; in July 2008, it awarded 15 new

cooperative agreements to the R&D projects focused on development, testing and

demonstration of oxy-combustion, chemical looping, membranes, solvents, and sorbents

for post-combustion CO2 capture [7].

In this connection, the on-going efforts of the Lyngfelt group at Chalmers University of

Technology in Sweden are significant. This group has developed a novel approach to

separate CO2 by using the benign redox capability and thermochemical aspects of a host

of solid oxygen carriers in a fluidized bed mode [8-10]. They have coined the acronym

CLOU - chemical looping with oxygen uncoupling - to describe the process. CLOU is

essentially an extension of the CLC (chemical-looping combustion) concept which is

based on a 55-year old patent by Lewis and Gilliland [11] describing the production of

high purity carbon dioxide, by circulating a solid oxygen carrier between two fluidized

bed reactors: an air reactor and a fuel reactor. The fuel is introduced in a fuel reactor

containing a solid oxygen carrier ( )yxOM wherein it is oxidized to CO2 and H2O. The

reduced oxygen carrier ( )1−yxOM is transported to the air reactor where it is reoxidized.

After water removal by simple condensation, CO2 is separated without any direct loss in

efficiency, since the fuel does not encounter air.

Recently, it has been shown that the methodology is applicable to solid fuel, such as coal,

as well [12-13]. This broadens the scope of capturing CO2 by CLOU process

tremendously, since there is an abundance of coal and similar solid fuels (including

5

biomass) all across the globe. The United States herself has coal reserves enough to

supply energy for the next 300 years and Australia exports about one-third of the total

global cal supply annually.

A number of unsupported- as well as supported- oxides of the first d-group transition

metals (notably Mn, Fe, Co, Ni and Cu) that are amenable to facile redox transformation

have been successfully employed; recently, some compositions containing two metals

(Fe/Mn or Fe/Ni) and perovskites endowed with tailored stoichiometric defects

(CaMn0.875Ti0.125O3) were also used [10]. In some cases, inert zirconia [14] and silica [15]

supports have also been used. During the 238th annual American Chemical Meeting in

Washington, D.C., several approaches were presented in the symposium on ‘Advances in

CO2 Conversion and Utilization’ [16]. The methodologies presented in 15 papers ranged

from catalytic to electrochemical, photochemical and thermochemical using a number of

stand-alone or supported systems, leading to variety of end-products such as carbon

monoxide (CO), formaldehyde (HCHO), formic acid (HCOOH), methanol (CH3OH), and

methane (CH4).

With such extensive efforts dedicated towards mitigating greenhouse gas effect in general

and lowering the carbon footprints in particular, an energy revolution could be initiated

by developing a methodology and demonstrating its practical implications, wherein

carbon dioxide and water could be catalytically and/or photocatalytically converted into

syngas (CO+H2). This value-added calorific commodity could then be used in a number

of high-end applications, including (a) the synthesis of value-added liquid hydrocarbons

6

by Fischer-Tropsch process [17] or (b) generating electricity by fuel cells. A visual of the

greenhouse gas beneficiation concept is shown in Fig. 1.

Fig. 1. Schematic pathways of the proposed CO2 mitigation technology.

Innovation

In contrast to the chemical looping with oxygen uncoupling (CLOU) process, our

technique essentially causes targeted and quantitative conversion of CO2 and H2O both

into CO and H2, respectively, through a simple heterogeneous gas/solid redox process

[18]. This is based on the simple fact that any reducing reaction must be accompanied by

a concurrent oxidation process and vice versa. Thus, the reduction of CO2 and/or H2O

must be accompanied by a concomitant oxidation of another species elsewhere in the

7

system. In that sense, this methodology is in fact CLOU - operating in reverse – as is

evident from Fig. 2, albeit with more attractive features such as less energy intensiveness,

less expensive and added beneficiation aspect rather than sequestration.

Fig. 2. Overall summary of the chemical looping combustion process (left) for CO2 capture and the greenhouse gas beneficiation technique (right) proposed here.

While the CLOU process obviates the separation step by using a solid oxidizer in lieu of

air, the GHG beneficiation paradigm offers several self-evident advantages over it. For

example, it converts the two green house gases into syngas at much lower temperatures,

for generating electrical power via fuel cells and/or liquid fuel(s) via F-T process.

Moreover, the spent reactant is regenerated easily by using inferior quality syngas gas,

such as from biomass. Finally, the exits streams of the fuel cell as well as the regenerator

are CO2 and H2O, which could again be fed into the cycle without any processing or

additional cleanup.

CO2, H2O from flue

streamCO2, H2O CO2, H2O

Gasified biomass/ Carbothermic

M or MOx 853 K

CO, H2

MOx or

MOx+1873K

Liquid fuel

SOFC 923 K

F-T reactor 483 K

Electric power

Air Fuel

Air reactor 1123 K

MO

M

Fuel reactor 1223 K

CO2, H2O N2

CO2

- H2O

Capture & store

8

We describe here a strategy of converting carbon dioxide into carbon monoxide, of water

vapors into hydrogen and a mixture of carbon dioxide and water into syngas (CO+H2),

via innovative heterogeneous surface catalysis over a simple metal or metal oxide under

very mild experimental conditions (temperature: 580°C; pressure: 1atm.). These streams

when fed into an IT-SOFC at 650°C, created a open circuit voltage, quite comparable to

that of the same SOFC run with H2 (ideal fuel) and/or CO. The metal and metal oxide

that mediate the conversion of carbon dioxide into carbon monoxide, of water into

hydrogen or the CO2+H2O mixture into syngas (CO+H2), are in turn converted into an

oxide with enhanced magnetic characteristics. The spent oxide could either be

regenerated or used for manufacturing a variety of ceramic magnets (spinel ferrites of the

general formula AIIBIII2O4) [19] which have an annual revenue of about $5.4 billion. The

regeneration of the spent oxide could be carried out either carbothermically or with

gasified biomass (a low quality substitute for conventional reductant) under mild

experimental conditions.

Theoretical Rationale

We consider the following gas-solid redox reactions:

)()()()( 12 gCOsMOsMOgCO xx +→+ + (1)

)()()()(2 gCOsMOsMgCO +→+ (2)

9

In this work, the greenhouse gas was made to interact with iron oxide (magnetite, Fe3O4)

in one case and with elemental iron (Fe) in the other. The reduction of CO2 to CO results

in formation of Fe2O3 due to the oxidation of Fe3O4 and Fe as per Eq. (3) and (4):

)()(3)(2)( 32432 gCOsOFesOFegCO +→+ (3)

)(3)()(2)(3 322 gCOsOFesFegCO +→+ (4)

The numerical value of the overall equilibrium constant (K) for a stoichiometric reaction,

is a measure of the occurrence propensity of the said reaction. It is also a strong function

of temperature, and at a given temperature K is unique for a particular reaction. In

stoichiometry calculations, one can assume the reaction to run to completion. However,

by definition, the equilibrium constant (K) is the ratio of the concentration of the

reactants and the products at equilibrium. Therefore, at a given temperature, a large K

value (>1) implies that the fraction of product(s) is larger than the reactant(s) and that the

equilibrium lies to the right, whereas, a small K value shifts the reaction to the left.

Finally, the equilibrium constant is related to the Gibbs’ energy of the reaction, which

dictates its thermodynamic feasibility.

The overall equilibrium constant is given by [20]: 21 KKKK O ⋅⋅= (5)

Where,

10

⎟⎟⎠

⎞⎜⎜⎝

⎛ ∆+

∆−=

⎥⎦

⎤⎢⎣

⎡⎟⎠⎞

⎜⎝⎛ −

∆=

⎟⎟⎠

⎞⎜⎜⎝

⎛ ∆−=

∫ ∫T

T

T

T

oP

oP

O

O

oO

O

oO

o

O OTdT

RCdT

RC

TK

TT

RTH

K

RTG

K

1exp

1exp

exp

2

1

(6)

OK is the equilibrium constant at the reference temperature OT (298K), 1K is a

multiplier that contributes towards the temperature effect when the heat of reaction is

independent of temperature and 2K accounts for the change is heat of reaction with

temperature; in most of the instances, 2K is small and is ~1. Using reliable

thermodynamic data for various components [21], the net equilibrium constant for

reaction (3) at 580ºC (853 K) was computed to be:

( ) ( ) ( ) 561021 10387.2110144.41076.5 ×=⋅×⋅×=⋅⋅= −KKKK o

Clearly the equilibrium constant of the reaction is far >> 1, and accounts for the fact that

the equilibrium lies to the right. Accordingly, the reduction of CO2 into CO with

magnetite is feasible.

In addition, the thermodynamic equilibrium oxygen partial pressure (2Op ) in the biphasic

mixture of Fe3O4 and Fe2O3 could be computed to be 1.82 x 10-15 atm. at 853 K is

computed atm. At 2Op values greater than 1.82 x 10-15 atm., the oxidation of Fe3O4 into

Fe2O3 will occur. Similar analysis yields a value of 1.41 x 10-26 atm. for the equilibrium

11

oxygen partial pressure needed for the oxidation of Fe into Fe2O3 [21]; therefore, at 2Op

greater than 1.41 x 10-26 atm. oxidation of Fe into Fe2O3 will take place.

As can be readily seen from reactions shown in Eq. (3) and (4), the oxidation of Fe or

Fe3O4 into Fe2O3 is attended by the reduction of CO2 into CO. The partial pressure of

oxygen due to the coexistence of CO2 and CO by virtue of the reaction:

)()(21)( 22 gCOgOgCO →+ (7)

could be computed from the following equation:

22

)7(

2

2

1

⎟⎟⎟

⎠

⎞

⎜⎜⎜

⎝

⎛⋅⎟⎟

⎠

⎞⎜⎜⎝

⎛=

∆−

RTG

CO

COO o

epp

p (8)

According to Eq. (8), it is possible to control the partial pressure of oxygen at a given

temperature very precisely by controlling the ratio of the concentration of CO2 and CO in

the mixture. The 2Op values computed at 853 K in some CO2/CO mixtures are shown in

Table I.

12

Table I. Equilibrium2Op values at 853 K in some representative CO2/CO and

H2O/H2 mixtures.

Mixture 2Op (atm.) Mixture

2Op (atm.) CO2 – 100 ppm CO 1810445.1 −× H2O – 100 ppm H2 19106.4 −× CO2 – 10 ppm CO 1610445.1 −× H2O – 10 ppm H2 17106.4 −× CO2 – 1 ppm CO 1410445.1 −× H2O – 1 ppm H2 15106.4 −× CO2 – 0.1 ppm CO 1210445.1 −× H2O – 0.1 ppm H2 13106.4 −×

As stated above, at 853 K the equilibrium2Op values for the individual Fe3O4/Fe2O3 and

Fe/Fe2O3 coexistences is 1.82 x 10-15 atm. and 1.41 x 10-26 atm., respectively. Both these

values must be exceeded in order to oxidize Fe3O4 or Fe to Fe2O3. From Table I it can be

readily seen that the 2Op generated by the CO2-CO mixture in the last row (* ~ practically

pure CO2 stream) exceeds both these values by several orders of magnitude. It can,

therefore, be argued that pure carbon dioxide is capable of oxidizing Fe3O4 and Fe both to

Fe2O3; as seen from Eq. (3) and Eq. (4), CO2 would in turn be reduced to CO.

Likewise, to demonstrate the feasibility of the reaction between water vapor and the solid

metal or metal oxide, the 2Op values at 853 K for various H2O/H2 mixtures were

computed for the reaction:

OHOH 222 21

→+ (9)

Some typical 2Op values in the H2O/H2 mixtures are also summarized in Table I.

13

As can be seen, the 2Op in the H2O – 0.1 ppm H2 mixture (*~ pure H2O stream) is also

more than several orders of magnitude higher than that required theoretically for the

oxidation of Fe3O4 or Fe to Fe2O3 as per the reactions:

)()(3)(2)( 232432 gHsOFegOFegOH +→+ (10)

)(3)()(2)(3 2322 gHsOFesFegOH +→+ (11)

Therefore, like carbon dioxide, water vapor also is capable of oxidizing Fe3O4 and Fe to

Fe2O3; accordingly, like CO2, H2O would also be reduced to H2.

Thus, the proposed technology offers two feasible pathways to utilize the syngas thus

produced, viz., as (i) a direct feed to a solid oxide fuel cell and/or (ii) conversion into

liquid fuels such as hydrocarbons, ethanol, butanol, etc., via Fischer-Tropsch synthesis.

These pathways were depicted in Fig. 1 above.

Experimental Details

Commercial grade steel mill waste powder (particle size ≤ 45 µm; hereafter referred to as

‘mill scale’) from Midrex Technologies Incorporated (Charlotte, NC) and two types of

iron powders (99%; average particle size 1-3 µm and 80 µm) from Alfa-Aesar were used.

In some cases nanoscale iron powder produced by the solvothermal conversion of mill-

scale waste in a 1-L capacity autoclave (Autoclave Engineers, PA) by a method

14

developed and described in detail elsewhere [22] and stated briefly here. 10 g of the mill-

scale waste (predominantly Fe3O4) powder was added to 50 mL of ethanol and 160 mL of

hydrazine monohydrate (99%, Alfa-Aesar). Reagent grade sodium hydroxide pellets

(~10g; Alfa Aesar) were added to maintain the pH (>12) of the mixture. This mixture

was placed in autoclave vessel and purged with high purity nitrogen to displace air in the

system after which the system was pressurized to about 1000 psig. The reaction was

conducted at 100°C for a reaction time of 6 h. The product was washed thoroughly using

ethanol, water, and acetone as the washing media in a centrifuge. All the samples were

systematically characterized for their phase and morphological features before and after

the conversion reactions using XRD, SEM/EDS and VSM.

In each case, the powder was packed in a stainless steel filter with 5µm size pores from

Swagelok™, which was placed in the center of a quartz reactor located in the uniform

temperature zone of a Lindberg Minimite furnace (Ashville, NC). The sample was first

heated up to 580°C in high purity nitrogen and switched to pure CO2 flowing at 100

Sccm. In the case of experiments involving water vapors, CO2 was bubbled through a

water bath prior to entry in the reactor; the moisture level in the gas stream was varied by

changing the temperature of the bath. The sample size was varied from 2 to15 g.

The conversion yield was quantified in terms of signals for the gaseous species in the exit

stream of the reactor quantified by a Model 2010 Shimadzu gas chromatograph. An

assessment of the quality and quantity of the product was also carried out in terms of the

performance of a single-button Hionic™ solid oxide fuel cell (Ni-GDC || YSZ || LSM-

15

GDC) from NexTech Materials (Lewis Center, OH) by monitoring the open circuit

voltage (OCV; no load condition) using the CO, H2 and/or (CO+H2) thus produced and,

comparing it with that obtained when the fuel cell was run with pure H2 and/or CO under

identical conditions.

To further evaluate the validity of the concept in a real-world like conditions, exhaust of a

600 cc Honda engine was fed to a reactor packed with the magnetite powder and the exit

stream was led into the anode of a single button SOFC operated at 650°C. The

experimental set up used for this evaluation is shown in Fig. 3.

Fig. 3. Experimental set up used for the validation of the concept using car exhaust feed.

Engine exhaust

Catalytic reactor

SOFC

16

Results and Discussion

Literature is abundant with several suggestions with regard to CO2 storage (also known

as sequestration), among which geologic formation and terrestrial ecosystems are most

widely discussed. In the first, CO2 storage is envisaged to occur between porous rock

layers, capped by a layer or multiple layers of non-porous rocks above them. In the

second, CO2 uptake by plants growing on land and in freshwater and is expected to be

achieved through tree-planting, no-till farming, forest preservation and understanding and

deciphering the genomes of carbon-storing soil microbes. Other possibilities that have

been looked at and pursued to a great degree are the dissolution of CO2 in brine [23],

reaction with alkali- or alkaline-earth minerals to form solid carbonates [24]. Other

interesting concepts for CO2 sequestration are through self-assembled organic

nanochannels [25], zeolites [26] or MCM-41 materials [27], its reduction to CO in the

presence of hydrogen and methane over magnesia (MgO) [28]. The photoreduction of

CO2 with water in gas phase over titania dispersed on silica and zeolites has also been

reported [29], but with very low activity; the evolution rate of the products (in this case

methane and methanol) per hour was few nanomoles/g catalyst. Moreover, in all these

instances, the end products are most of the time, methanol, formaldehyde, formic acid

and methanol and not syngas, which calls for elaborate separation protocols. Bidrawn et

al. [30] have described the reduction of CO2 at 800°C in a solid oxide electrolyzer, by

operating an SOFC in reverse using CO-CO2 mixture as the fuel and chromium-

substituted LSM (La0.8Sr0.2Cr0.5Mn0.5O3) in conjunction with (yttria + zirconia)-doped

ceria and palladium as the redox stable anode. Evidently, the fabrication of this complex

electrolyzer is quite elaborate.

17

Figure 4 shows the evidence of CO2 conversion into CO as monitored by the on-line GC

signals with 6 g of magnetite, 2 g of commercial iron, and 15 g of nanoscale iron derived

from commercial mill-scale (MIDREX) sample, respectively; the CO yield in all these

cases was found to be very close to the theoretical value as per equation (3) and (4),

within the permissible experimental errors. Several subsequent runs under identical

experimental conditions yielded similar results, bolstering the theoretical concept

outlined above. The monotonic decrease in the concentration of the product gas (CO) in

the exit stream with time is simply due to the gradual depletion of the active component

(Fe or Fe3O4) with the progress of reaction.

a

18

Fig. 4. Time dependence of CO generation from CO2 with: (a) Fe3O4, (b) commercial microscale Fe and (c) nanoscale Fe powder.

c

b

19

A direct evidence of the propensity of reduction of water vapors into hydrogen as per the

proposed concept (Eq. (10)) was seen in the functionality of a solid oxide button fuel cell

is and presented in Fig. 5. The OCV responded sharply to the increase in temperature of

the water bath. This is expected; as the vaporization and hence steam content of the

stream increased steadily with rise in the bath temperature, the hydrogen level also

increased. The steady decline in the cell output with time is due to the constant

consumption of the active catalytic component (magnetite). Generation of hydrogen via

metal-steam reforming reaction as per Eq. (10) has been amply demonstrated [31-34]:

Fig. 5. Evidence of H2 generation from pure water stream in contact with the magnetite.

That CO2 was indeed converted into CO as per reaction (3) or (4) was further verified by

connecting the exit stream to an SOFC at 650°C. This is shown in Fig. 6 for both dry and

20

humidified CO2 streams; for comparison the fuel cell output with pure hydrogen and pure

carbon monoxide fuels are also included.

Fig. 6. Comparison of a single-button fuel cell outpput with different feeds upon interaction of CO2 with magnetite.

A comparison of the fuel cell output under engine exhaust conditions with those using

other feeds is presented in Fig. 7; the fuel cell operated at 83.9% efficiency with the

engine exhaust.

21

90.7% 86.8% 89.8%83.9%

99.9%

0%

20%

40%

60%

80%

100%

Pure

H2

Pure

CO

CO

2Ex

haus

t

(n

on-h

umid

ified

)

CO

2Ex

haus

t

(h

umid

ified

)

Engi

ne E

xhau

st

% E

ffici

ency

90.7% 86.8% 89.8%83.9%

99.9%

0%

20%

40%

60%

80%

100%

Pure

H2

Pure

CO

CO

2Ex

haus

t

(n

on-h

umid

ified

)

CO

2Ex

haus

t

(h

umid

ified

)

Engi

ne E

xhau

st

90.7% 86.8% 89.8%83.9%

99.9%

0%

20%

40%

60%

80%

100%

Pure

H2

Pure

CO

CO

2Ex

haus

t

(n

on-h

umid

ified

)

CO

2Ex

haus

t

(h

umid

ified

)

Engi

ne E

xhau

st

% E

ffici

ency

Fig. 7. Comparison of the efficiency of a single-button fuel cell at 650°C on different feeds; the efficiency is defined as the ratio of ( )fuelGHGOCV − to ( )2HOCV .

In all the above cases, the fuel cell response under ‘no load’ condition (OCV) was

continuously monitored, and as could be seen, it shows a good correlation of the

operational efficiency of the reactant. The higher OCV in the case of humidified CO2

stream could be attributed to the reduction of water vapors; presence of H2 in the gas

stream, though in small concentration, accentuated the cell performance. Thus, it is

reasonable to conclude that a mixture of CO2 and H2O (both of which are potent

greenhouse gases and are the final products of a combustion process) could be converted

into syngas (CO+H2) as per the reactions:

)()()(6)()()(4 2322243 gHgCOsOFegOHgCOsOFe ++→++ (12)

22

)(3)(3)(2)(3)(3)(4 23222 gHgCOsOFegOHgCOsFe ++→++ (13)

The syngas stream could either be used as a combined feed for SOFC or can be catalyzed

into liquid fuel via Fischer-Tropsh synthesis. The latter possibility is currently under

investigation in our laboratory. Also included in Fig.6 is the fuel cell output under engine

exhaust conditions; the fuel cell operated at 83.9% efficiency with the engine exhaust.

It should be pointed out that as per Eqs. (3-4), Eqs. (10-11) and Eqs. (12-13), Fe3O4 and

Fe would transform theoretically into Fe2O3 upon interaction with CO2, H2O and their

mixtures. Intuitively, this could be hematite. However, a comparison of the magnetic

measurements before and after CO2 reduction in this work showed that this was not the

case. In fact, maghemite (instead of hematite) was formed, as can be seen from the

pronounced M-H behavior of the post-reduced sample from the magnetization (M) vs.

applied magnetic field (H) curves shown in Figs. 8 and 9.

Fig. 8. Fe3O4 magnetization curve before and after exposure to CO2 gas at 580°C for 4h.

23

Fig. 9. Magnetization curve of iron before and after exposure to CO2 gas at 580°C for 4h.

This can be explained as follows. Because metal oxides tend to be crystalline substances,

the reactions they undergo are either reconstructive or topotactic. In reconstructive

reactions, native crystals disassociate, react and, crystallize again. In topotactic reactions,

the crystal structure of each particle remains intact throughout the reaction.

Maghemite ((γ-Fe2O3) is ferrimagnetic and isostructural with spinel magnetite (Fe3O4)

with cationic deficiency at the octahedral B-site in the lattice. It is also known to form by

the topotactic transformation of magnetite attended by internal atomic displacements and

defect formation, so that the initial and final lattices are in coherence [35]. Diffusion

24

studies with radioactive iron tracers in magnetite particles have been used at high

temperatures and relatively low2Op to confirm it [36]. This observation correlates the

changes in the diffusion coefficients due to the changes in oxygen partial pressures with

reactions that involve the formation of vacancies in the octahedral lattice [37]. Formation

of these vacancies is important for diffusion [38] and oxidation as it electrically

compensates for the conversion of Fe2+ to Fe3+. We believe the experimental conditions

employed in this work are conducive to the observed effect. Similar curves were obtained

for water-treated samples as well.

The highly magnetic spent oxide (maghemite, γ-Fe2O3) was converted into elemental iron

by carbothermic reduction according to the following stoichiometric reaction with

graphite powder at 600ºC:

)(3),(2)(),(32 gCOsFesCsOFe +→+ αγ (14)

That the conversion did occur is evidenced from the XRD patterns on the sample before

and after the reduction. This is shown in Fig. 10.

25

Fig. 10. Structural evidence of reconversion of maghemite into iron via carbothermic reduction.

26

Conclusion

Two inexpensive precursors - a metal and an oxide - capable of converting CO2 and H2O

into CO and H2, respectively, on a 1:1 molar basis, at mild temperature of 580°C and 1

atm. pressure, have been identified and used in the beneficiation of the two greenhouse

gases. After reduction, these streams when fed into a solid oxide fuel cell at 650°C

created a open circuit voltage, quite comparable to that of the same run with pure H2

(ideal fuel). The energy sector is expected to benefit from this approach as it would

provide a means of curbing the carbon footprints from fossil fuels in an innovative way.

NASA is currently looking at non-petroleum-based jet fuels in the pursuit of alternative

fuels that can power commercial jets and address rising oil costs. A greenhouse gas-

derived F-T fuel could respond to that quest. Experiments to explore this option are in

progress.

This concept will advance the state-of-the-art in the pursuit of carbon capture and

sequestration, in several ways. First, as mentioned above, though several methodologies

have been proposed, none of them seem capable of selectively and quantitatively

converting CO2 to CO – one of the two valuable precursors for Fischer-Tropsch

synthesis. Therefore, the theoretical concept and experimental results shown here are

perhaps the first of their kind that ensure selective, quantitative and targeted conversion

of CO2 into CO. Second, the reduction of the CO2 provides a ‘green’ and environmentally

benign process of combating the ever-increasing greenhouse gas effect. Furthermore, it

shows that if the mill scale waste of steel industry could be employed as an effective

agent for the envisaged conversion, it would add great value towards the remediation of

27

another industrial waste, more than 80% of which is currently landfill destined. Nearly 20

million metric tons (MMt) of mill-scale waste is produced annually by the steel industry

worldwide; the United States’ share of this waste generation is 1.5 MMt per year.

Calculations based on the stoichiometry of reaction (3) indicate that at the current level of

this waste available worldwide, about 2 MMt of CO2 could be converted into 1.2 MMt of

CO annually. In the case of elemental iron (unlimited supply), every 1 MMt of metal

would remediate 1.2 MMt of CO2, producing 0.754 MMt of CO. In both the cases, the

spent oxide could be regenerated by carbothermic reduction under mild experimental

conditions or by using low quality gasified biomass. Alternatively, it could be utilized in

the manufacturing of ferrite (MFe2O4) components, due to its highly magnetic behavior.

References

1. Climate Change 2007: The Physical Science Basis, IPCC Summary for Policymakers,

May 2009.

2. Greenhouse gases, climate change, and energy, Report # DOE/EIA X012, May 2008.

3. D. Archer, “Fate of fossil fuel CO2 in geologic time,” J. Geophys. Res., 110 (2005)

C09S05.

4. N. Shah, D. Panjala and G.P. Huffman, “Hydrogen production by catalytic

decomposition of methane,” Energy Fuels, 15 (2001) 1528-534.

5. N. Shah, Y. Wang, D. Panjala and G.P. Huffman, “Production of hydrogen and

carbon nanostructures by non-oxidative catalytic dehydrogenation of ethane and

propane,” Energy Fuels, 18 (2004) 727-735.

6. US Dept. of Energy, Carbon sequestration technology road-map and program plan,

DOE Office of Fossil Energy’s National Energy Technology Laboratory, Pittsburgh,

PA (2007).

28

7. J.P. Ciferno, T.E. Fout, A.P. Jones and J.T. Murphy, “Capturing carbon from existing

coal-fired power plants,” Chem. Engg. Progress, 105 (2009) 33-41.

8. A. Lyngfelt and H. Thunman, “Construction and 100h of operational experience of a

10-kW chemical-looping combustor,” Carbon Dioxide Capture for Storage in Deep

Geological Formations – Results from the CO2 Capture Project, 1 (2005) 625-645.

9. E. Jerndal, T. Mattisson and A. Lyngfelt, “Investigation o f different NiO/NiAl2O4

particles as oxygen carriers for chemical-looping combustion,” Energy & Fuels, 23

(2009) 665-676.

10. H. Leion, Y. Larring, E. Bakken, T. Mattisson, R. Bredesen and A.Lyngfelt, “The use

of CaMn0.875Ti0.125O3 as oxygen carrier in chemical-looping with oxygen uncoupling

(CLOU),” private communication.

11. W.K. Lewis and E.R. Gilliland, “Production of pure carbon dioxide,” US patent,

2665971 (1954).

12. N. Berguerand and A. Lyngfelt, “The use of petroleum coke as fuel in a 10 kWh

chemical-looping combustor,” Int. J. Greenhouse Gas Control, 2 (2008) 169-179.

13. T. Mattisson, A. Lyngfelt and H. Leion, “Chemical-looping with oxygen uncoupling

for combustion of solid fuels,” Int. J. Greenhouse Gas Control, 3 (2009) 11-19.

14. T. Mattisson, H. Leion and A. Lyngfelt, “Chemical-looping with oxygen coupling

using CuO/ZrO2 with petroleum coke,” Fuel, 88 (2009) 683-690.

15. A. Shulman, E. Cleverstam, T. Mattisson and A. Lyngfelt, “Mn/Fe, Mn/Ni and Mn/Si

oxides used in chemical loping with oxygen uncoupling (CLOU) for combustion of

methane,” submitted for publication (2009).

16. http://oasys2.confex.com/acs/238nm/techprogram/S30544.HTM

17. M.E. Dry, “The Fischer-Tropsch process: 1950–2000,” Catalysis Today, 71 (2002)

227-241.

18. A.-M. Azad, “Method of making syngas from greenhouse gases,” US Patent, Ser. No.

12/705,759, filed February 15, 2010.

19. R. Hasegawa, “Design and fabrication of new soft magnetic materials,” J. Non-Cryst.

Solids, 329 (2003) 1–7.

20. J.M. Smith, H.V. Van Ness and M.M. Abbott, Introduction to Chemical Engineering

Thermodynamics, Chapter 13, 7th ed. McGraw Hill, New York, NY (2005).

29

21. O. Kubaschewski and C.B. Alcock, Metallurgical Thermochemistry, 5th Edition,

Pergamon (New York, 1979), pp. 378-384.

22. S. Al-Batty, Utilization of CO2 to Mitigate Greenhouse Gas Effect, MS Thesis, Dept.

of Chemical Engineering, UT (May 2010).

23. http://www.netl.doe.gov/technologies/carbon_seq/core_rd/storage.html.

24. H. Hayashi, J. Taniuchi, N. Furuyashiki, S. Sugiyama, S. Hirano and T. Nonaka,

“Efficient recovery of carbon dioxide from flue gases of coal-fired power plants by

cyclic fixed-bed operations over K2CO3-on-carbon,” Ind. Eng. Chem. Res. 37 (1998)

185-91.

25. P.K. Thallapally, B.P. McGrail, J.L. Atwood, C. Gaeta, C. Tedesco and P. Neri,

“Carbon dioxide capture in a self-assembled organic nanochannels,” Chem. Mater. 19

(2007) 3355-57.

26. L.M. Mulloth and J.E. Finn, Report, NASA/TM-1998-208752 (November 1998) pp.

10.

27. S.C. Shen, X. Chen and S. Kawi, “CO2 adsorption over Si-MCM-41 materials having

basic sites created by postmodification with La2O3,” Langmuir, 20 (2004) 9130-37.

28. K. Teramura, T. Tanaka, H. Ishikawa, Y. Kohno and T. Funabiki, “Photocatalytic

reduction of CO2 to CO in the presence of H2 or CH4 as a reductant over MgO,” J.

Phys. Chem. B, 108 (2004) 346-54.

29. K. Ikeue, S. Nozaki, M. Ogawa and M. Anpo, “Photocatalytic reduction of CO2 with

H2O on Ti-containing porous silica thin film photocatalysts,” Catal. Lett. 80 (2002)

111-14.

30. F. Bidrawn, G. Kim, G. Corre, J.T.S. Irvine, J.M. Vohs and R.J. Gorte, “Efficient

reduction of CO2 in a solid oxide electrolyzer,” Electrochem. Solid-Sate Lett. 11

(2008) B167-B170.

31. A.-M. Azad, “Carbon dioxide and water based fuels and fuel cells,” US provisional

patent, serial number 61/152,902, filed on February 16, 2009.

32. Sathees Kesavan and A.-M. Azad, “Conversion of steel mill waste into nanoscale

zero valent iron (nZVI) particles for hydrogen generation via metal-steam reforming,”

International Journal of Hydrogen Energy, 33 (2008) 1232-1242.

30

33. A.-M. Azad, Sathees Kesavan and Sirhan Al-Batty, “A closed-loop proposal for

hydrogen generation using steel waste and a prototype solar concentrator,”

International Journal of Energy Research, 33 (2009) 481-498.

34. A.-M. Azad, Sathees Kesavan and Sirhan Al-Batty, “Redemption of microscale mill

waste into commercial nanoscale asset,” Key Engineering Materials, 380 (2008) 229-

255.

35. J. Tang, M. Myers, K.A. Bosnick and L.E. Brus, “Magnetite Fe3O4 nanocrystals:

Spectroscopic observation of aqueous oxidation kinetics,” J. Phys. Chem. B, 107

(2003) 7501-7506.

36. P.S. Sidhu, R.J. Gilkes, and A.M. Posner, “Mechanism of the low temperature

oxidation of synthetic magnetites,” J. Inorg. Nucl.Chem. 39 (1977) 1953-1958.

37. N.L. Peterson, W.K. Chen and D. Wolf, “Correlation and isotope effects for cation

diffusion in magnetite,” J. Phys. Chem. Solids, 41 (1980) 709-719.

38. W. Jost, Diffusion in solids, liquids and gases, Academic Press, New York (1990) p.

135.