close

Вход

Забыли?

вход по аккаунту

?

Soot oxidation and diesel particulate filter regeneration by microwave heating

код для вставкиСкачать
SOOT OXID A TION AND D IESEL PARTICULATE FILTER REGENERATION
BY M ICROW AVE HEA TING
A Thesis
Presented to
The Faculty o f the College o f Graduate Studies
Lam ar University
In Partial Fulfillment
o f the Requirem ent for the Degree
M aster o f Environmental Studies
by
Daniel A. Rutm an
D ecem ber 2006
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1446361
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 1446361
Copyright 2007 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SOOT O X ID A TION A ND D IESEL PARTICULATE FILTER REGEN ERA TIO N BY
M ICROW AVE HEATING
DANIEL A. RUTM AN
Approved:
C he-Je^jpcrry) Lin
Supervising Professor
Thom as C. Ho
Co-Supervising Professor
R afail Tadm or
Com m ittee M ember
Tae Hoon Kim
Com m ittee M ember
Robert L.Yuan
Chair, D epartm ent o f Civil Engineering
Jaqk K. H opper
1 , College o f Engineering
Jerry W. Bradley
Dean, College o f Graduate Studies
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
© 2006 by D aniel A. Rutm an
N o part o f this w ork can be reproduced w ithout perm ission except as indicated by the
“Fair U se” clause o f the copyright law. Passages, images, or ideas taken from this work
m ust be properly credited in any written or published materials.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Soot O xidation and Diesel Particulate Filter Regeneration by M icrow ave Heating
By
Daniel A. Rutm an
W all-flow diesel particulate filters (DPFs) are considered the m ost effective devices
for reducing particulate m atter emissions from diesel exhausts. Periodic regeneration o f
the filters is required to rem ove the collected particulate m atter from the loaded filters to
restore their soot collection capacity, either through passive or active means. While
m icrowave heating is an effective active regeneration method, past investigations have
reported the generation o f hotspots because o f uneven m icrowave heating resulting in
physical filter dam age and eventual failure.
In this study the research team developed a w aveguide to elim inate DPF therm al
failure. The technology was dem onstrated in a well-equipped laboratory com posed o f a
diesel generator, an exhaust flow control system, a silicon carbide DPF, a soot sampling
system, and a differential pressure and temperature m easurem ent system. Upon
m icrowave heating the filter rapidly reaches 700° C w ith no apparent hot spots and
temperature varies no m ore than 100° C in both the radial and vertical positions. After
the desired soot loading, the regeneration took place off-line for five minutes under a
flow o f air, and the return to initial pressure dem onstrates effective regeneration.
M ultiple regeneration cycles w ith no evidence o f DPF failure dem onstrates an effective
system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I am grateful to my fam ily for their unending support and great encouragem ent to
pursue higher education and to achieve excellence in my academ ic career. I w ould like to
express deep gratitude to my research advisor Dr. Thom as C. Ho for providing me with
this opportunity to carry out the research under his guidance. I w ish to thank Dr. Tae
Hoon Kim for his active involvem ent in our project and always instilling creative ideas to
my work. I thank him for being encouraging and improving m y understanding o f the
subject by instilling his knowledge. I would also like to thank com m ittee members Dr.
Jerry Lin and Dr. Rafael Tadm or for sharing their broad know ledge and valuable
suggestions to com plete this work. I am also thankful to the staff o f the Chemical
Engineering D epartm ent for their help. I appreciate the help and cooperation given by
group m em bers Sameer Pallavkar and Suraj Shetty. Finally I gratefully acknowledge
financial support o f this study from the Texas C om m ission on Environm ental Quality
through the N ew Technology Research and D evelopm ent (NTRD) Program (Grant No.
582-5-70807-0007).
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
List o f Tables
vii
List o f Figures
viii
Chapter
1. Introduction
2.
1
1.1 Sources o f Particulate M atter
2
1.2 U.S. EPA PM Standards
3
1.3 European U nion Standards
4
1.4 H ealth Effects o f D PM
5
1.5 Diesel Particulate Control
6
1.6 Objective
7
Literature R eview
8
2.1 Introduction to D PM
8
2.2 C om position o f DPM
8
2.3 Diesel Particulate Filtration
10
2.4 Catalyzed Diesel Filters
13
2.5 Introduction to M icrowave Theory
15
2.5.1 M icrow ave Heating
16
2.5.2 RF Field Interaction with Substrate
19
2.5.3 Safety Considerations
20
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page
2.5.4 Com m ercial Components and Costs
2.6 Introduction to DPF Regeneration
3.
20
21
2.6.1 M icrow ave Regeneration
23
2.6.2 Factors that Affect Regeneration
23
2.7 Soot O xidation Kinetics and M echanism
24
Experim ental
25
3.1 Experim ental Setup
25
3.2 Diesel G enerator and Exhaust Flow Control
25
3.3 D iesel Particulate Filter and M icrowave
26
3.4 DPF Tem perature Profile
30
3.5 PM Sampling System
30
3.6 Periodic Off-line Regeneration and Soot Loading
31
3.7 M easurem ent Electronics
34
4. Results and D iscussion
4.1
4.2
35
Tem perature Profiling o f DPF
35
DPF R egeneration
39
4.2.1 Introduction to Regeneration
39
4.2.2 Off-Line Regeneration
39
4.3
Relationship o f DPF Tem perature and Pressure Drop
40
4.4
Relationship o f CO Formation and Pressure Drop
40
4.5
Relationship o f CO Formation and DPF Tem perature
41
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.6
5.
Online Regeneration
43
4.7
Soot Rem oval and Regeneration Efficiency
46
4.8
Soot O xidation Kinetics and M echanism
47
4.9
Elem entary Reaction M echanism
47
4.10
D efinition o f V ariables and Derivation o f Overall Rate Law
48
4.11
Experim ental Rate M odel Comparison
49
4.12
H eat Balance
50
Conclusions and Recom m endations
51
5.1 Conclusion
51
5.2 Recom m endations for Future W ork
52
References
53
Appendices
59
I.
Diesel Fuel Specifications
59
II.
Com ponent Sources
60
III.
DPF Specifications
61
Biographical N ote
62
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
T able
Page
1.1 EPA Em ission Standards for Heavy Duty Trucks
4
1.2 Heavy D uty Diesel Em issions EU Standards
5
2.1 Table o f D ielectric Factors
19
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figures
Page
1.1
PM Characteristics and M odal D istribution
2
2.1
Types o f W all Filters
11
2.2
Diffusional D eposition
11
2.3
Interception D eposition
12
2 .4
Electrom agnetic Spectrum
16
2.5
M icrowave vs. Conventional Heating
18
2.6
Regeneration Cycle
22
3.1
Schem atic o f Diesel Test Unit
27
3.2
D iagram o f DPF A ssem bly
28
3.3
DPF A ssem bly and O ven
29
3.4
W aveguide Installation
29
3.5
PM Sampling Loop
31
3.6
O verview o f Em issions Diesel Test U nit
32
3.7
Research Control V alve and Transmitter
33
3.8
Process Instrum entation Console
34
4.1
Vertical DPF Tem perature Profile
36
4.2
Radial (Parallel) DPF Tem perature Profile
37
4.3
Radial DPF Tem perature Profile
37
4.4
Effect o f W aveguide on Vertical DPF Tem perature Profiles
38
4.5
DPF Tem perature and Pressure Drop vs. Time
41
4.6
Pressure D rop and CO Concentration vs. Tim e
42
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.7
CO Concentration and DPF Tem perature vs. Time
42
4.8
Exhaust Flow Rate vs. Time
44
4.9
DPF Tem perature vs. Time
44
4.10
Pressure Drop vs. Time
45
4.11
DPF Tem perature vs. Time
45
4.12
Rate M odel Com parison
49
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
NOMENCLATURE
Roman Symbols
Cp
H eat Capacity
DPF
D iesel Particulate Filter
DPM
Diesel Particulate M atter
EPA
Environm ental Protection Agency
k
Rate Constant
KW
Kilow att
LPM
Liters Per M inute
Mw
M olecular W eight
N
R egeneration Efficiency
P
Pow er in W atts
Pa
Filter Pressure after Regeneration
Pb
Filter Pressure before Regeneration
P0
Clean Filter Pressure
PMio
Particulate M atter w ith Aerodynamic D iam eter 10 pm or less
PM 2.5
Particulate M atter w ith A erodynam ic D iam eter 2.5 pm or less
Q
Flow in m 3/hr
R
Ideal Gas Constant
RF
Radio Frequency
SiC
Silicon Carbide
T
Tem perature
TCEQ
Texas Com m ission for Environmental Quality
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 1
CHAPTER 1
Introduction
Over the last decade a m ajor effort to control diesel exhaust em issions has been
undertaken by European nations and the United States. To improve fuel econom y, many
people are switching to diesel fuel as an alternative. Some o f the advantages include
lower volatile organic com pounds (VOC) em issions and lower carbon dioxide (CO 2 ) and
other greenhouse gases. Two o f the m ajor drawbacks to diesel technology are increased
soot and N O x em issions produced by both stationary and m obile sources. Stationary
diesel generators are found widespread in the pow er generation industry. To improve air
quality, the regulatory bodies are imposing strict em ission standards. The prim ary
pollutants o f interest are diesel particulate m atter (DPM ), nitrogen oxides (NOx), and
sulfur oxides that are com m only present in com bustion gases and are strictly regulated by
the Clean A ir Act. D iesel exhaust has been long im plicated in serious health problems.
Some o f the possible exposure effects are cancer and cardiovascular and pulm onary
diseases. Current em issions reduction technology is sufficient for gasoline engines but
not suited for m eeting the strict new diesel standards.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 2
1.1 Sources of Particulate Matter
Particulate m atter or PM can be created by m an-m ade or natural sources. Natural
sources include com bustion products from w ildfire and geologic sources. M obile
emissions from vehicles are a m ajor contribution; other m an-m ade sources such as diesel
generators are becom e m ore prevalent. The characteristics o f PM are shown in Figure
1.1 (Kittelson 1998a).
0.2
Fine
Particles
Dp < 2.5 m
Nanoparticles
Dp < 50 nm
a °18
a
at
Ultrafine Particles
Dp < 1 0 0 nm
O 0.16
G
0.14
™
e
>
.. 0.7
o
PM10
D p< 10 m
j F r a c tio n a l d e & o srb o n o f p a r t c t e
s u iif h r t a n t i n f n f 1
- - 0.5
0.08
■■0.4
?a
8
it
■n
s
s
o.oe
5'
3
- •
Nuclei
Mode
0.02
0.001
Accumulation
Mode
0.010
0.100
0.2
o.t
1000
10.000
Diam eter |im)
M ass Weighting
Number Weighting
Alveolar Deposition Fraction
Figure 1.1 PM Characteristics and Modal Distribution
The above figure shows the diesel aerosol num ber and m ass-w eighted size distributions
and are trim odal and lognorm al in form. The concentration o f particles is proportional to
the area under the corresponding curve.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 3
M ost o f the m ass is in the accum ulation mode and consists o f volatile compounds. The
nuclei m ode contains 1-20 % o f the mass and 90 % o f the particle number. The coarse
mode particles are 5-20 % o f the particle mass. Also shown in Figure 1.1 are the
diameters o f the particle. The PM 2.5 particles are believed to cause the m ost detrimental
effects on health. The other sizes have research interest but are not considered significant
health risks. Secondary PM 2.5 from heterogeneous reactions convert some gaseous
pollutants (i.e. sulfur and nitrogen) to particles. In France, Belgium, and A ustria diesel
autos account for 60% o f new sales (W alsh 1999). Carbon based PM 2.5 accounts for the
majority o f the inventory. Roughly 50% o f the total PM 2.5 inventory is carbon based.
Recent data studies show that diesel exhaust contributes 50 -7 0 % o f total PM 2.5
(W atson, Fujita, Zielinska, Richards, N eff and D ietrich 1998).
1.2 U.S. EPA PM Standards
The Clean A ir A ct o f 1967 requires the EPA to set standards for six criteria pollutants
(Code o f Federal Regulations 1997). The m ost recent revision cam e in Fall 2006. The
PM 2.5 particle daily standard was lowered from 65 to 35 pg/m 3. To m eet the daily
standard, the three-year average o f the 98th percentile o f 24-hour concentrations at each
population-oriented m onitor m ust not exceed 35 pg/m 3. The annual P M 10 standard was
dropped because o f lack o f clinical evidence (Code o f Federal Regulations 1997). This
lowering o f standard facilitates the need for innovative diesel PM em issions control.
Table 1.1 docum ents the EPA (Code o f Federal Regulations 1997) em ission standards
(Nauss 1995).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 4
Table 1.1 EPA Emission Standards for Heavy Duty Trucks
Engine Year
NOx in g/bhp-hr
Truck and Bus
PM in g/bhp-hr
Truck
Bus
1985
10.7
None
N one
1988
10.7
0 .6
0 .6
1990
6
0 .6
0 .6
1991
5
0.25
0.25
1993
5
0.25
0 .1
1994
5
0 .1
0.07
1996
5
0 .1
0.05
1998
4
0 .1
0.05
1.3 European Union Standards
Low sulfur diesel and gasoline fuels (< 1 0 ppm S) were made available in 2005 and
become m andatory in 2009. European emission regulations for new heavy-duty diesel
engines are com m only referred to as Euro I, II, III, IV, and V (European Union
Com m ission Directive 2001). Sometimes Arabic num erals are also used (Euro 1 to 5).
Since the Euro 2 stage, EU regulations introduce different em ission limits for diesel and
gasoline vehicles. D iesels have more stringent CO standards but are allowed higher
NOx. For PM m easurem ents, particle num ber limits have been considered at the Euro 5
level. A t the tim e o f the publication o f the Euro 5 proposal, closed particulate filters
could m eet only its m ass-based PM em ission limits. On -road heavy-duty diesel emission
standards are listed in Table 1.2 (E. U. Com m ission Directive 2001).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 5
Table 1.2 Heavy Duty Diesel Emissions EU Standards (g/kWh)
TIER
YEAR
CO
HC
NOx
PM
Euro 1
1992
4.5
1 .1
8 .0
0.36
Euro II
1996
4.0
1 .1
7.0
0.25
Euro II
1998
4.0
1 .1
7.0
0.15
Euro III
2000
2 .1
0 .6 6
5.0
0 .1 0
Euro IV
2005
1.5
0.46
3.5
0 .0 2
Euro V
2008
1.5
0.46
2 .0
0 .0 2
1.4 Health Effects of DPM
Diesel em issions are highly com plex compounds. Public heath concern has come to
the forefront for the following reasons:
1. M ost o f the particles are highly respirable w ith a size less that 10 microns;
2. M any carcinogenic organic com pounds are adsorbed onto the surface;
3. The gas phase consists o f irritants and toxic chemicals.
Diesel particles or diesel particulate matter (DPM ) is thought to cause cancer and other
cardiovascular diseases. It is difficult to distinguish diesel exhaust PM 2.5 from other
am bient sources o f PM (H ealth Effects Institute 1995) because there are too many
outlining external sources and factors to blame PM 2.5 for adverse health. Since PM 2.5 is
higher in urban areas, m ost clinical studies are done w ith samples from urban areas.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 6
Recent exposure studies to elevated (23-311 ug/m ) concentrations o f PM 2.5 for two
hours induced transient pulm onary irritation in test subjects, but no chronic effects were
seen. It appears that the healthy subject’s defense m echanism is able to cope with
ambient PM 2.5 concentrations (Ghio, Kim and Devlin 2000). However, the elderly and
young m ay be greater affected (Schwartz 1994). M ortality studies are widespread in
Europe (W orld H ealth O rganization 2002). A random study done in the Netherlands
concluded that exposure to traffic related PM 2.5 pollution may shorten lifespan (Hoek,
Brunekreef, Goldbohm, and Fischer 2002). However, when studied here in the United
States, an increased but statistical non-significant risk for cancer was established
(Dockery, Pope, Xu, Spengler, Fay, and Spiezer 1993). A detailed A m erican Cancer
Society study did find data that support shortened lifespan because o f lung cancer
mortality (Pope, Burnett, Thun, and Calle 2002).
1.5 Diesel Particulate Control
Comm ercial diesel particulate control systems are com binations o f filters and
regeneration methods. The majority o f these use therm al m ethods for regeneration. The
first is utilized in passive systems, the latter in active systems. The passive method
involves an oxidation catalyst that is discussed later in the chapter on oxidation. The
m ost com mon types include fuel additive systems, catalyzed filters, and filter + catalyst
configurations. A m ajor problem is that diesel exhaust tem peratures are generally low
compared to soot ignition temperatures. To supply additional heat, m icrowave absorbing
ceramic filters have been tested w ith high regeneration efficiency. However, hotspots
from unequal m icrowave heating result in physical filter damage.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 7
1.6 Objective
The objective o f this study was to develop and dem onstrate the use o f microwave
absorbing silicon carbide for the reliable regeneration o f DPFs. In the study a diesel
emission control laboratory equipped w ith 7.4 kW generator, exhaust flow system, DPF
system, soot sam pling system, differential pressure m easurem ent system, temperature
measurement, and a data acquisition system was established to conduct the experiments.
In this thesis Chapter 1 describes the sources and standards o f particulate matter,
Chapter 2 presents a literature survey; in Chapter 3 the experim ental setup is presented in
full detail; and Chapter 4 presents the results o f this study. R ecom m endations and
conclusions are presented in Chapter 5.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 8
CHAPTER 2
Literature Review
This chapter review s literature related to this study.
2.1 Introduction to DPM
Diesel particulate m atter or DPM is the most com plex em ission product o f diesel fuel
combustion. This definition includes solids as well as liquid m aterial that condenses
during the exhaust formation. Opposite to gaseous em issions, D PM is not a w ell-defined
species. The sam pling process is essential; any changes here will affect results
(Burtscher 2001). For sound analytical results the sampling m ust be standardized
(Kittelson 1998a). D PM as specified by National Institute o f O ccupational Safety and
Health (NIOSH) is sampled by filtering exhaust at tem peratures no higher than 52° C
(NIOSH 1996). The exhaust stream is proportionally diluted w ith air, and this m ethod is
said to simulate D PM release from vehicles into the atmosphere.
2.2 Composition of DPM
DPM is com posed o f elemental carbon clusters that adsorb other species to form
complex heterogeneous structures. DPM has a bim odal size distribution consisting o f
nuclei and accum ulation modes. N uclei particles are very small and are form ed from gas
precursors that form from cooling. They com pose the m ajority o f the particle number
(90 %) but only a few percent o f the total DPM mass. A ccum ulation m ode particles are
formed from carbon clustering and other solids present in the exhaust. A liquid layer is
formed by condensation o f vapors on the solid clusters. This fraction is m ost o f the total
mass o f the DPM.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 9
DPM can be characterized by three fractions. The first fraction is the (SOL) solid
fraction com posed o f ash and elemental carbon. The second fraction is the (SOF) soluble
organic fraction derived by lube oil and organic fuel content. The final fraction is the
sulfate (SO 4 ) particulates com posed o f sulfuric acid and water. From this the total
particulate m atter or TPM can be calculated.
TPM = SOL + SOF + S 0 4
The solid fraction is m ainly elemental carbon that is not chem ically bound to other
species. This fraction results from heterogeneous combustion. The carbon atoms are
arranged in hexagonal arrays (Broome and Khan 1971). These arrays form platelet like
structures. These platelets agglomerate and form more com plex structures. These
structures have varying lengths and diameters. A sh particles are the second component
o f the solid fraction. N ew engines tend to produce m ore ash, and it poses a corrosion
problem (Abdul-K halek, Kittleson, Graskow, Wei, and Brear 1998). A sh is composed o f
sulfates, phosphates oxides o f zinc, calcium, magnesium , and other m etals present in the
lube additives (M erkel, Cutler, and W arren 2001). M etal w ear and tear particles also
contribute to ash. A sh does not oxidize further and can eventually end the filter’s life.
The soluble organic fraction is com posed o f hydrocarbons that adsorb onto the surface o f
the carbon particles. The SOF has a vapor and a liquid state depending on temperature.
The SOF is form ed from lube oil hydrocarbons (Voss 1995). Particulates w ith low SOF
content are called dry particulates. Particles o f high SOF content are referred to as wet
particulates. Here the organic fraction is over 50 % o f the total PM. The SOF content
depends on engine conditions (W achter 1990).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 10
If the exhaust tem perature is low the SOF content is high. This range is from idle to
200° C. W hen the tem perature is above 400° C, the SOF drops to less than 5 %. The
SOF also contains cyclic aromatic com ponents that are thought to be carcinogenic.
Sulfate particles are com posed o f sulfuric acid, and the reaction betw een w ater and acid
is modeled as a heterom olecular nucleation (Baum gard and Johnson 1996). It is believed
by Baum gard and Johnson that the sulfate particles are separate from the carbon particles
and are present in the exhaust gas prim arily as nuclei m ode particles. The presence o f
sulfate salts com es from cations present in lube oil additives. TPM em ission is calculated
from the w eight o f the total mass collected on a filter.
2.3 Diesel Particulate Filtration
Diesel exhaust sources include both stationary and mobile sources. Each source
presents a set o f em ission reduction and control issues. Exhaust em issions from mobile
sources such as autos and commercial vehicles can be reduced by a converter system very
similar to that found in regular gasoline automobiles. M odem control technologies either
oxidize (bum ) or physically remove soot. Paper filters for low-tem perature service or
ceramic fiber filters for high-tem perature service rem ove soot from the exhaust stream as
it flows through the filter. H igh removal efficiencies o f over 80 % are achievable with
soot filters. Typical service lifetimes for these filters are short because o f clogging,
usually less than 50 hours. This method generates solid waste and requires investment in
consumable goods. Diesel particulate filters (DPF) capture particle em issions through a
combination o f surface and bed filtration. The filters accum ulate a large mass o f soot
because o f the low bulk density o f soot. To restore the flow o f the filter, oxidation o f the
soot m ust take place.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 11
The m ethod o f therm al regeneration is common. The m ost com m on filter for thermal
regeneration is the wall flow monolith. The material o f choice is silicon carbide (SiC).
Alternate channels are plugged, forcing the diffusion o f gas through the walls as shown in
Figure 2.1 (H aralam pous and Koltsakis 2002).
Wall-flow
|
■
Flow-through
^
Z <1
^
-—*
< rra
—»
—
—
..............=
Figure 2.1 Types of Wall Filters
Cake and depth filtration are the main mechanisms o f w all filter filtration. The two
mechanisms are diffusional deposition and interception. In diffusional deposition
Brownian m otion is exhibited, and the particles do not m ove uniform ly in the
streamlines. Particle concentration near the fiber surface is zero, and a concentration
gradient in the exhaust gas is created driving diffusion. For an illustration o f diffusional
deposition refer to Figure 2.2.
Figure 2.2 Diffusional Deposition
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 12
For interception, the stream lines pass close to the m edia at a distance less than the
particle’s radius and result in particles striking the m edia and depositing. Flow
interception deposition is illustrated in Figure 2.3.
Figure 2.3 Interception Deposition
The DPF is efficient w ith solids but not w ith liquid particulate fractions. The sulfates and
organic fraction (SOF) are difficult to remove. Certain catalyst system s generate sulfates
and contribute to em issions. To offset this problem , systems are designed to run on less
than 50 PPM sulfur fuels only. A nother problem is the filtration m echanism tends to
increase the form ation o f nanoparticles through nucleation. The m ost important
commercial factor is reliable regeneration and service span. The EPA regulations o f 0.02
g/mi for light duty (LD) vehicles and 0.01 g/bhp-hr for heavy duty (HD) vehicles will lead
to a boom in DPF im plem entation (CFR 1997). The first use w as by M ercedes in 1985
for California vehicles but had problem s and was rem oved from the m arket (Abthoff,
Schuster, and Loose 1985).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutman 13
The m ost popular systems can be divided into two groups: regeneration during engine
operation and disposable filter systems. The first case is active regeneration (engine
operating) by heating the gas exhaust stream w ith a fuel burner. A n alternate technology
is to use electricity to heat gas and filter media. High pow er consum ption is a major
disadvantage. U sing catalysts is another popular method; this is discussed in detail in
next section. Some use fuel additives as a passive regeneration method. Its main
disadvantage is the form ation o f additional compounds. The m ost popular commercial
technology is by Johnson M atthey (Chatterjee, Conway, V isw anathan and Jacobs 2004).
The continuous regenerating trap or CRT ™ uses nitrous oxide or N O present in the
exhaust stream to convert gas to N O 2 . The passive system does have problem s with low
N Ox/PM ratio streams. H eat m ust be supplied to exhaust streams that are below 260° C.
The disposable filter system s are simple and involve changing dirty filters for new ones.
Pollution concerns and the issue o f flam mability exist w ith disposable filters.
2.4 Catalyzed Diesel Filters
A catalytic diesel filter is im pregnated w ith oxidation catalysts, usually platinum and/
or palladium. W hen a hot exhaust stream passes through the filter structure, the catalyst
promotes the reaction o f soot, CO, and volatile organic com pounds (VOC) w ith oxygen
in the exhaust gases. Catalytic filter systems are m ost effective w hen at a high
temperature w ith optim al perform ance near 400° C. However, the diesel engine
generates exhaust tem peratures below the soot ignition tem perature w ith large amounts o f
soot that quickly plug the filter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 14
In catalytic oxidation the exhaust gas is sent through a catalyst elem ent that traps and
oxides the soot. Rem oval efficiencies o f up to 50% are achievable w ith this process.
Generally, higher rem oval efficiencies are not possible because o f the low tem perature o f
the diesel exhaust stream. M any catalytic systems have been developed for regular
gasoline em issions, but until late 1990 only a few were available for diesel engines. By
using iron and copper as a catalyst, the diesel soot ignition tem perature can be reduced by
200° C (Ma, Fang, Li, Zhu, Lu, and Lau 1997). M a also investigated the use o f Pd for
improving diesel soot oxidation. Later the shift to use La MnC>3 perovskite coated filters
come to the forefront (Zhang-Steenwinkel, V an der Zande, Castricum, Blick, V an der
Brink, and Elzinga 2005). In this study the selectivity tow ards CO 2 was close to 100 %.
The soot bum o ff tem perature was approxim ately 500° C with this catalyst. The method
o f thermal heating used was m icrowave energy. Current com posite technology allows
ceramics to be used for filter design. Pr doped cerium oxide catalysts have been
dem onstrated for autom otive use by Fino (Fino and Specchia 2004). The soot bum off
temperature is now reduced to 400° C. The use o f M o /V on A lum ina substrates have
been investigated by Leocadio et al. (Leocadio, Braun, and Schmal 2004). The
mechanism o f oxidation is form ation and decom position o f carbonate species to give
CO 2 . In 2005 advanced tem poral analysis further enhanced the m echanism o f soot
oxidation (Bueno-Lopez, M akkee, Krishna, and M oulijn 2005). In the presence o f a La3+
doped CeC>2 catalyst, labeled oxygen replaces non-labeled lattice oxygen.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 15
This highly reactive lattice oxygen reacts with soot to give CO 2 . The bum off
tem perature is 400° C. The current area o f research is w ith zirconium oxide catalysts
doped w ith Cu and or K (Laversin, Courcot, Zhilinskaya, Cousin, and Aboukais 2006).
W ith these oxides the soot bum o ff temperature is reduced to 375° C. Carrascull and
associates did a sim ilar study w ith nitric acid on zirconium oxides (Carrascull, Ponzi, and
Ponzi 2003). The acid-doped catalyst has the m ost enhanced rates. To further
understand the role o f K precursors, Jimenez and fellows studied K/M gO oxides
(Jimenez, Garcia, Cellier, Ruiz, and Gordon 2006). Carbon black was used in situ o f
diesel exhaust soot, and no filter system was used to actually test catalyst.
2.5 Introduction to Microwave Theory
M icrowaves are electrom agnetic waves w ith frequencies in the range o f 100 M Hz to
100 GHz as shown in Figure 2.4 (M etaxas and M eredith 1983). M icrow aves have been
used prim arily in the field o f telecom m unications and industry. In d u strial p ro cesses
use freq u en cies o f 915 and 2450 M Hz.
The FCC has reserved 915 M H z and 2450
MHz, among other frequencies, for industrial applications. 2450 M Hz microwaves are
common in com m ercial systems and have a wavelength o f 4.8" in air.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 16
Frequency,
103
M
l ;
|
i
i
A —c p o w e r i
I I I ! ! !
3 X 1 0 ’°
3 X IQ7
Hz
109
106
ill
i 111 1
I I III
|T V |j| |
3 x I ff1
30
i
i
jR adiO i
1M i c r o w a v e
10
1
1
i
1
12
1]
..
Infra
red
1
3 x 10 “ 2
1 0 15
1
i i!
i
| V isible
!
i i!
1
t!
i
'
1 0 18
'1
!X
f
rays
•
I
3 X 10- 5
W avelength,- c m
Figure 2.4 Electromagnetic Spectrum
M icrowave pow er is usually m easured in kilowatts. A t room tem perature and 1
atmosphere pressure, 1 kilow att o f m icrowave energy will evaporate approxim ately 2.5
pounds o f w ater in 1 hour. After the mass production o f the m agnetron, an efficient high
frequency m icrowave generator, microwaves have been increasingly used for heating
applications.
2.5.1 Microwave Heating
Heating is based on absorption o f radiofrequency (RF) energy by dielectric materials.
Polar molecules, called susceptors, vibrate when placed in electrom agnetic fields
(M etaxas and M eredith 1983). This vibration causes an increase in the molecule's
kinetic energy, w hich is dissipated as heat. The dielectric properties o f a material are
characterized by its com plex perm ittivity s:
e = s' - i s"
(2.1)
where the real part (s') is the dielectric constant and the im aginary part (s") is the
dielectric loss factor.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutman 17
The dielectric constant is an indication o f the am ount o f energy that can be stored in a
material in the form o f an electric field, while the dielectric loss factor is a direct measure
o f how m uch energy a material can dissipate in the form o f heat (M eredith 1998). W hen
a material is heated by m icrowave irradiation, the am ount o f RF pow er converted to heat
per unit volum e o f a susceptor is a function o f the electric field frequency, its intensity,
and the dielectric loss factor, as given by equation 2.2:
P = 2 7t f s0 e" E2
(2.2)
P power, W /m3
f - field frequency, Hz
so - absolute perm ittivity, 8.854-10'12 F/m
s" - dielectric loss factor, dimensionless
E - electric field intensity, V/m
M aterials that exhibit m agnetic properties can be heated by both the electric (E) and
magnetic (H) com ponents o f an electrom agnetic field. RF pow er associated with the Hfield com ponent can be expressed by a relationship sim ilar to Eq.2.2 in which the
dielectric loss factor (s") and the electric field intensity (E) are replaced by magnetic loss
factor (p") and m agnetic field intensity (H) respectively. M icrow ave heating differs
significantly from conventional heating. During conventional heating o f a solid material,
the heat is first transferred to its surface, typically by a com bination o f convection and
radiation mechanisms. The inside o f the material is then heated through conduction from
the surface. As illustrated in Figure 2.5, with m icrowave heating the energy deposition is
concentrated in the m aterial itself, resulting in heating from inside out. A nother
characteristic feature o f m icrowave heating is its selectivity. Some materials are strong
absorbers o f RF energy, while others are not.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 18
I
_
^
I C onventional
H eating
Figure 2.5 Microwave vs. Conventional Heating
M icrowaves m ost effectively heat m aterials o f high dielectric constants and high
dielectric loss factors. Exam ples o f such materials include water, carbon black, and SiC
as well as diesel soot. M etals, on the other hand, are nearly perfect reflectors o f
m icrowave energy. Therefore, metal ducts can be used as w aveguides for RF energy.
Dielectric properties o f some selected materials, including diesel soot, are listed in Table
2.1 (M a et al. 1997). These values confirm why diesel particulates are an RF absorber,
while cordierite is not (Babu, Farinash, and Seehra 1995). However, if SiC is substituted
for the cordierite, it is an excellent RF absorber.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 19
2.1 Table of Dielectric Factors
Material
Dielectric constant s’
Dielectric loss factor e”
Diesel soot
10.695
3.561
Quartz
3.78
0.001
Cordierite
2.873
0.138
A120 3
3.006
0.170
SiC
30
11
Z r0 2
4.214
0.186
2.5.2 RF Field Interaction with Substrate
The deposition o f m icrowave energy w ithin a heated m edium is dependent on the
m edium geometry, som etim es called an RF cavity. If RF energy launched dow n a metal
waveguide strikes a metal wall, the energy will be reflected o ff the wall, setting up a
series o f E and H -field standing waves. The E-field distribution in a microwave oven can
be pictured as a tw o-dim ensional array o f E-field peaks o f high electrical field intensity
and valleys w here the field intensity is low. M aterials w ith dielectric loss properties will
be heated only w hen placed in an E-field standing wave peak in phase. Consumer
microwave oven m anufacturers attem pt to com pensate for this uneven heating by moving
the food through the electrical field on a rotating plate.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 20
2.5.3 Safety Considerations
M icrowave heating equipm ent m ust be designed, constructed, and operated to provide
adequate protection against radiation hazards. On the basis o f current Occupational
Safety and H ealth A dm inistration (OSHA) research, the m icrowave leakage power
density shall not exceed a pow er density o f 50 W /m2 (5 m W /cm 2) at any accessible
location 0.05 m from any portion o f the equipm ent under conditions designated as normal
operation. In addition, the m icrowave leakage shall not exceed a pow er density o f 100
W /m at any accessible location 0.05 m from any portion o f the equipm ent under
conditions designated as abnormal operation. Electrom agnetic radiation protection
guidelines are defined under 29 CFR Ch. XVII o f the O ccupational Safety and Health
Admin, Labor, § 1910.97 for non-ionizing radiation.
2.5.4 Commercial Components and Costs
M icrowave ovens are m ass-produced so com ponent costs are relatively low. For
example, the m anufacturer costs for a typical 900 W m agnetron were reported to be only
about $15 in quantity w ith the associated pow er supply unit at about $20 - $40. A 900 W
magnetron has an average service life o f about 5,000 hours.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 21
2.6 Introduction to DPF Regeneration
The dynam ic regeneration o f D PF’s is characterized by a dynam ic equilibrium
between the soot produced and the soot oxidized by the process. Soot oxidation rates
mainly depend on DPF tem perature and soot loading. Therm al oxidation o f soot involves
the oxidation o f solid particulates to gaseous products.
C+ 0 2 -»> C 0 2
C + V2 O2 -*■ CO
The tem perature at w hich soot starts oxidizing is the soot ignition temperature. The exact
value is highly dependent on the system used. C 0 2 is the preferred product, but in an
oxygen deficient zone CO is produced. In a standard flow through catalytic converter the
residence tim e is too short, so soot is not effectively oxidized. A t low tem peratures soot
is oxidized slowly; only at 600° C is oxidation fast and complete. As soot loading
increases, so does rate. The regeneration o f the filter is a continuous equilibrium process.
The first term in the equation represents the accum ulation o f soot; the rem aining two
term s describe the oxidation and deposition o f fresh soot on the filter.
dM/dx + M-k(T) - eF-m = 0
(2.3)
M - particulate mass on the filter, kg
x - time, s
k(T) - reaction rate constant for particulate oxidation, 1/s
T - temperature
eF - filter collection efficiency, dimensionless
m - particulate m ass flow from the engine, kg/s
Depending on the sign o f the accum ulation term, the filter may be in one o f the following
operation modes. R efer to Figure 2.6 for an illustration o f the stages.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 22
1. A ccum ulation {dM/dx > 0)
2. Equilibrium {dM/dx = 0)
3. R egeneration {dM/dx < 0).
AP
Time
Figure 2.6 Regeneration Cycle
The first zone is the accum ulation phase at a low tem perature and low soot oxidation rate.
As the exhaust tem perature is raised, the pressure increases, and the rate o f oxidation
increase. Eventually equilibrium occurs in zone two. This is the balance temperature.
As the tem perature increases, the filter enters the regeneration zone. Here the rate o f
oxidation is greater than the rate o f accumulation. The pressure drop now decreases and
is zero at very high tem peratures. The second variable is soot loading; as a consequence
the DPF can regenerate at different loading and tem peratures. A t low tem perature the
soot loading increases, and pressure drop increases.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 23
2.6.1 Microwave Regeneration
M icrowave regeneration uses volum etric heating to heat a filter substrate. G am er and
Dent did the first microwave regeneration study in 1989 (G am er and D ent 1989). This
attempt at regeneration was not too efficient because o f poor w aveguide and diffuser
design. In a vehicle prototype study, G am er and associates discovered the problem o f
poor regeneration because o f flow control (G am er and D ent 1990). G am er and
associates found that preheating time had to be long to ensure com plete regeneration. To
optimize m icrowave energy, W alton and associates fitted the ceram ic filter cavity with
ferrite plugs to optim ize the interaction with the electrom agnetic field (W alton, Hayward,
and W ren 1990). In 1994 m athematical models o f regeneration started to be refined. A
two-dim ensional transient model utilizing ceramic foam further confirm ed the concept o f
selective dielectric heating o f ceramics by m icrowaves (Chunrun, Jiayi, Jiahua, Lunhui,
Junmin, and Chengbin 1994).
2.6.2 Factors that Affect Regeneration
M icrowaves tend to heat unevenly, leading to hot spots. If these hot zones go
unregulated, therm al stress will lead to cracking and/or melting. M ost m ajor failures are
caused by either slow or uncontrolled rapid regeneration (Popuri and Hendrichsen 2001).
Exhaust flow is another variable that affects regeneration efficiency; if flow is too high,
the tem perature m ay not be high enough for com bustion to occur com pletely, and soot
loading m ay cause an increase in backpressure. This reduction in tem perature is a reason
that online regeneration is difficult to achieve in our test unit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutman 24
To com pensate for high flow, high power levels are needed, or w aveguide design m ust be
optimized. The issue o f soot loading has to be addressed since high levels o f soot can
lead to runaw ay regeneration. This happens when the additional heat supplied by soot
com bustion causes the tem perature to rise rapidly and crack or m elt the filter.
2.7 Soot Oxidation Kinetics and Mechanism
Soot oxidation is a very com plex process and does not always yield the classical
com bustion product CO 2 . W hen m odeling soot oxidation, two assum ptions are made to
simplify the model. It is assumed that all filter channels behave the same with uniform
soot deposition and tem perature is uniform across the filter (H aralam pous and Koltsakis
2002). The rate o f carbon oxidation is dependent on the partial pressure o f O 2 (Field,
Gill, M organ, and H aw ksley 1967). W ith the advent o f CFD, m any new m ulti channel
regeneration m odels have surfaced (Kostoglou, Housiada, and Konstandopoulos 2003).
These studies model the m igration o f a com bustion zone traveling through the DPF. In
the two-step model the form ation o f the soot and the oxidation step are the only reaction
steps (Hiroyasu, Kadota, and Arai 1983). The m ost plausible m echanism appears to be
the N agle and Strickland-Constable or N SC model (Nagle and Strickland-Constable
1962). This model involves active sites on the surface o f a soot particle. There also
appear to be different sites w ith lesser selectivity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 25
CHAPTER 3
Experimental
The experim ental aspect o f the study is discussed in detail in this chapter. In this
study a diesel em ission control laboratory equipped w ith 7.4 kW generator, exhaust flow
system, DPF system, soot sampling system, differential pressure m easurem ent system,
temperature m easurem ent, and a data acquisition system was established to conduct the
experiments. A 1.4 kW com mercial oven was m odified for use in this study. The
exhaust flow was lined up to flow through the filter at a fixed rate. The setup and
procedures are split into subsections and discussed in depth.
3.1 Experimental Setup
The test setup is divided into four m ain sections:
1. Diesel generator and exhaust flow control
2. DPF and m icrowave
3. PM sampling and exhaust gas analysis
4. Instrum entation
3.2 Diesel Generator and Exhaust Flow Control
A 7.3 kW generator driven by a single- cycle diesel engine (Lom bardi M odel 15LD
400) was installed in the w alk-in hood, and engine exhaust was connected to a DPF
through the flow control valve (Badger) and flow m eter (A sea Brow n Boveri). An
overview o f the entire apparatus can be found in Figure 3.1.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 26
Three resistive loads were then connected to the outlets, and the load o f 36.36 amps (60%
o f total capacity) was m easured with a wattmeter. A carbon m onoxide detector was used
to verify that harm ful levels were not present in the operating area, and adequate hearing
protection was required when the engine was operating. The Diesel fuel (Citgo Clear No.
2) was purchased in bulk from a local bulk supplier to ensure the consistency o f fuel
quality. (See fuel specifications in Appendix 1).
3.3 Diesel Particulate Filter and Microwave
A silicon carbide (SiC, Ceramic Techniques et Industrielles) w all-flow monolith filter
(50 m m diam eter x 150 m m length, cell density = 200 cpi, pore size = 20 microns) is
enclosed in a custom -m ade quartz holder (Technical Glass Products), w hich is insulated
and sealed by using Fiberfax alum ina blanket and Interam m at (3M ) to insulate and
protect the quartz holder from therm al expansion stress o f the filter elem ent as shown in
Figure 3.2. Two high tem perature flange gaskets are also used to seal between the filter
element and the holder. A 1400W Sharp consumer oven was m odified to concentrate
microwave energy to the SiC filter by adding a wave-guide. The oven as shown in
Figure 3.3 has the cavity m odified to allow the filter assem bly to mount.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 27
V-2
wr
©©
Diesel
Diesel Eng ne/Generator
V-1
Diesel
RarticUate
filter
Air
FT- HowTransfritter
TT- Tenperafare Transrritter
FT - Pressure Transmitter
DP-Deferential FYesstxeCell
WT-Vthtt Meter
AX-Analyzer
V1,V2,V3,V4-^ves
*
V-3
0
----------
PMSampler
Row Controller
Figure 3.1 Schematic of Diesel Test Unit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SarrplePirrp
Rutm an 28
► Diesel Engine Exhaust - OUT
High Temperature
Gasket
Ceramic Wool / Interam Mat
Diesel Particulate Filter
(SIC)
Gasket
Faraday Screen
Diesel Engine Exhaust - IN
Figure 3.2 Diagram of DPF Assembly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 29
Figure 3.3 DPF Assembly and Oven
Figure 3.4 Waveguide Installation
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
Rutm an 30
3.4 DPF Temperature Profile
Uniform heating o f filter verifies the effectiveness o f the w aveguide design.
The research team perform ed vertical and radial tem perature profiles to discover any
variations in heating. The critical tem perature for soot oxidation is approxim ately 400°
C. Ideally the filter heating should be gradient free to avoid incom plete soot oxidation
and possible therm al stressing that leads to m elting and or cracking. Sub miniature
junction therm ocouples (Omega) are used to reduce m icrowave interference.
The research team placed three therm ocouples inside the m iddle o f a new DPF at 5, 9,
and 13 cm depths, and the m icrowave was turned on for thirty minutes. Tem perature and
time were logged w ith a data acquisition system com m ercially available from National
Instruments (N.I.). For radial profile, three therm ocouples were placed in a line from the
edge o f the filter to opposite edge at a depth o f 8 cm. The filter is rotated 90 degrees for
other orientation, and the procedure was repeated.
3.5 PM Sampling System
The soot sam pling system is consisted o f sample selector valves, a filter holder (SKC
M odel LS-47), and a sam pling pum p (SKC M odel Hi Lite 30) w ith rotam eter to monitor
a sample flow rate o f 30 cc/minute. The valves (Parker) and pum p (SKC) are controlled
by the switches located in the control panel to ensure the right sam pling sequence. The
sample lines and the filter holder are traced with electric heating tape as shown in Figure
3.5 to m aintain tem perature at 80 - 90° C to avoid the condensation o f w ater in the line
and the filter holder. A fter collecting soot for a preset time, the filter is allowed to dry in
a desiccator for 24 hours and w eighed on a m icrobalance.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 31
Figure 3.5 PM Sampling Loop
3.6 Periodic Off -line Regeneration and Soot Loading
The off-line regeneration setup as shown in Figure 3.6 consists o f a m odified
microwave oven, filter holder, flow control valve, and differential pressure transmitter.
To avoid exposure to microwave energy and hot surfaces, all operation o f valves are done
remotely through rack-m ounted instrum entation controls.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 32
Figure 3.6 Overview of Diesel Emissions Test Unit
As shown in Figure 3.7, a Research control valve (Badger) and A sea Brow n Boveri
(ABB) flow control transm itter were installed to m aintain constant exhaust flow and
pressure.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 33
Figure 3.7 Research Control Valve and Transmitter
To dem onstrate off-line regeneration, the DPF and engine m ust be run under controlled
conditions. Initial DPF pressure is recorded at 10 liters/m in at am bient temperature. The
research team ran the engine under load and set the exhaust flow rate at 5 m 3/hr.
Regeneration was started when pressure reaches 50 inches o f water. A ir flow is set to 10
Liters/min, and exhaust flow is bypassed. M icrowave is switched on and o ff to maintain
DPF tem perature above 675° C for five minutes. If filter tem perature exceeds 750° C, the
microwave is cycled o ff to avoid therm ocouple damage.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 34
3.7 Measurement Electronics
A cromag A PM 765 m eters display the pressure, temperature, and flow data as shown
in Figure 3.8. N.I. software and PCI-232 data acquisition hardware w ere used to log all
data in real time. M ultiple therm ocouples were used to log tem perature through the N.I.
U SB -9211 interface hardware. A Testo 350XL em issions analyzer was also used in
conjunction to log O 2 , N O x, total hydrocarbons, CO, and SOx. Online CO 2
measurements w ere taken with a CEA Instruments GD444 portable analyzer. All
pressure-m easuring and tem perature devices were calibrated w ith a calibrator.
Figure 3.8 Process Instrumentation Console
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 35
CHAPTER 4
Results and Discussion
This chapter reports experim ental data, provides discussion o f the observed data, and
is divided into the following subsections: tem perature profiles o f DPF, DPF
regeneration, the effects o f flow on DPF tem perature, and finally soot removal efficiency
and soot oxidation kinetics.
4.1 Temperature Profiling of DPF
The first phase o f the experim ent was to m easure tem perature profile o f the DPF to
see if the m icrowaves are evenly absorbed. Uneven m icrowave heating w ould result in
local hotspots resulting in physical damage to the DPF. The filter was fitted with three
therm ocouples, and m icrowave pow er was applied. The two orientations (radial and
vertical) were m easured. In the vertical profile the therm ocouple leads were placed at a
depth o f 50, 75, and 100 mm. For the radial profile the leads were placed in line at a
fixed depth o f 75 mm. The filter was then rotated 90 degrees and repeated. We
conducted a study to see if a waveguide could effectively increase the DPF temperature.
The com m ercial m icrowave oven has no adjustable tuning circuit or dumm y load to
absorb reverse pow er that could potentially dam age the magnetron; however, it has a
reverse interlock to protect the magnetron. Good tem perature profiles as shown in
Figures 4.1 to 4.3 w ere obtained w ithout interference by using thinner gauge
therm ocouple wire (0.8m m J-type) and placing the wire leads inside the SiC channels.
RF interference is norm ally experienced but is elim inated by this technique.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 36
The high therm al conductivity o f the SiC tends to reduce the tem perature gradients and
the waveguide enhances the DPF tem perature as dem onstrated in Figure 4.4 by as much
as 300° C. W ithout the waveguide the average tem perature is only 400° C, w hich is
insufficient for com plete soot combustion. In the vertical profile the tem perature ranges
from 690° C to 830° C, and the deviation appears to be related to the relative position o f
the DPF to the waveguide. For the radial profile the deviation is greater because o f heat
losses near the edges o f the DPF.
900
800
830° C
757° C
700
690° C
u
'J*-iT 600
|
50 mm
500
MW 900W
!«*>
H
£ 300
Q
150 mm
200
100
0
2
4
6
8
Time (minutes)
10
12
Figure 4.1 Vertical DPF Temperature Profile
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Rutm an 37
800
710'
DPF Temperature (C)
700
660'
600
585
500
400
300
ooo
MW 900 W
100
2
0
4
6
12
10
8
14
T im e ( m in u te s )
Figure 4.2 Radial (Parallel) DPF Temperature Profile
800
779 °C
671° C
DPF Temperature (C)
700
642° C
500
400
300
MW 900W
200
100
0
2
4
6
8
10
12
Time (minutes)
Figure 4.3 Radial DPF Temperature Profile
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Rutm an 38
900
With waveguide
835° C
800
MW 900W
U 700
Without waveguide
528° C
600
I 500
a
& 400
4)
£ 300
Pu
Q 200
100
0
2
4
6
8
10
12
14
T im e (m inutes)
Figure 4.4 Effect of Waveguide on Vertical DPF Temperature Profiles
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 39
4.2 DPF Regeneration
4.2.1 Introduction to Regeneration
Regeneration can be carried out off-line or on-line. Off-line regeneration differs
because the exhaust flow is diverted from the DPF while regenerating. Off-line
regeneration is easier to reach soot light o ff tem perature since airflow has negligible
effect on DPF heating. For on-line regeneration to be successful, m ore m icrowave power
is needed to oxidize the soot because o f heat losses. Both have a problem w ith runaway
regeneration and therm al stressing if too m uch soot is accumulated.
4.2.2 Off-Line Regeneration
The off-line regeneration study involves diverting the engine exhaust flow from the
DPF once a specified pressure drop across the DPF is reached. The soot m ust be loaded
at a rate w here uncontrolled regeneration does not take place. A flow control valve and
controller m aintain constant flow during the soot-loading phase. The exhaust flow rate o f
5 m /hr was set to load the filter and takes approxim ately one hour to reach a pressure
drop o f 50 inches o f w ater under accelerated loading conditions o f high exhaust pressure.
To insure sufficient oxygen for combustion, a constant flow o f 10 1/min o f air at STP
conditions (21% O 2 ) was switched on before the m icrowave was turned on for five
minutes. In the regeneration experiment, the variables observed were DPF temperature,
exhaust pressure and flow, soot loading, and time. A t least four cycles o f loading and
regeneration were carried out, and data acquisition (N.I. Labview) was logged every
second for the entire event. The results are presented in the next sections.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 40
4.3 Relationship of DPF Temperature and Pressure Drop
The tem perature o f the system is fixed by the efficiency o f the m agnetron; 800° C is
the m axim um D PF tem perature obtained with 1400 W. The research team studied the
effects o f DPF tem perature and pressure as dem onstrated in Figure 4.5. As the engine
loads the filter w ith soot, the DPF AP rises. W hen the threshold o f 50 inches o f AP is
reached, the exhaust gas is diverted, the m icrowave is switched on, and the pressure falls
immediately. A fter five m inutes o f soot oxidation, the pressure returns to the clean state
and is ready to be loaded w ith soot again. The first cycle always loads up the quickest
since the PM content o f the exhaust is high w hen the engine is first started and not at
optimum operating temperature. A fter a short time the DPF tem perature reaches a steady
state o f 300° C. The DPF tem perature average is about 725° C, and the whole process is
reproducible over m ultiple runs.
4.4 Relationship of CO Formation and Pressure Drop
At 400° C, oxidation takes place, and carbon m onoxide (CO) is produced. As this
happens, the blocked filter’s pores and channels open up, and the pressure falls off.
Instantaneously the pressure rapidly returns to the initial pressure value as demonstrated
in Figure 4.6. Each AP profile is similar as expected, and the CO produced varies from
cycle to cycle.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 41
4.5 Relationship of CO Formation and DPF Temperature
As expected, soot com bustion generates H 2 O and CO 2 , but in an oxygen-deficient
environm ent CO is the expected product. As the tem perature rises above 400° C, soot
bum s and produces CO as shown in Fig 4.7. As dem onstrated, the CO production rapidly
increases during the soot bum phase and quickly returns to the baseline when the
microwave is switched off. The m axim um CO concentration corresponds to the
m axim um DPF temperature.
800
60
DPF T
700
- 50
0
-4—»
CD
5
600
o
£ 500
3
40
0
.c
o
c
C
- 30 0l
CD
0 400
Q.
E
0
Iu_ 300
Q
0
L_
1
U
20 w
CO
01
1
CL
Q
I
200
CL
LL
CL
100
Cl
Pressure
0
100
200
300
T im e (m in u te s)
Figure 4.5 DPF Temperature and Pressure Drop vs. Time
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 42
100,000
60.000
50.000
40.000
Pressure
30.000
20.000
(ppm )
70.000
Concentration
80.000
qq
Drop (inches of w ater)
90.000
10,000
i
0
i
50
0
r* i
100
150
200
250
300
Tim e (minutes)
Figure 4.6 Pressure Drop and CO Concentration vs. Time
- 8 0 ,0 0 0
700 -
500 -
- 5 0 ,0 0 0
400 -
- 4 0 ,0 0 0
3 0 0 --
- 3 0 ,0 0 0
200
- -
-
2 0,000
CO
100
10.000
-
0
30
60
90
120
150
180
210
240
270
0
30 0
T im e (m in u te s )
Figure 4.7 CO Concentration and DPF Temperature vs. Time
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO Concentration
(C)
DPF Temperature
- 6 0 ,0 0 0
(ppm)
7 0 .0 0 0
DPF T
Rutm an 43
4.6 Online Regeneration
The on-line regeneration experim ent differs from the off-line in the fact that no
external air source is used for com bustion and the exhaust is not bypassed. The on-line
regeneration has problem s with heat loss from the exhaust gas in the pipe, control valve,
and flow m eter and from not enough microwave power. U sing the heat content o f the
exhaust stream to boost the DPF temperature w ould be feasible, but the exhaust
temperatures are still less than 250 0 C after 80 minutes o f engine runtime. Improvements
including using fiberfax wool and glass fiber insulating tape w ere tried, but still too much
heat was lost. The DPF tem perature is related to the flow rate as illustrated in Figure 4.8
and 4.9. A flow rate o f 5 m 3/hr results in a DPF regeneration tem perature o f 278° C, and
a flow rate o f 2 m /hr resulted in a DPF regeneration tem perature o f only 554° C. In
separate trial the flow w as reduced to lm 3/hr as shown in Figure 4.11 for a maxim um
DPF heating o f 787° C. As shown in Figure 4.10 and 4.11, the DPF tem perature peaks as
the initial accum ulated soot bum s o ff and the pressure drop rises at the same mom ent as
the tem perature rises. D uring the accum ulation stage, the DPF tem perature is below the
light o ff point o f 400° C, so little oxidation takes place. W hen the m icrow ave is turned
on, the filter enters the regeneration mode. The tem perature is so high that the am ount o f
soot oxidized is higher than that captured in the filter per unit o f time. As a result, the
soot loading and the pressure drop decrease. The other stage is w hen equilibrium is
reached betw een fresh soot loading and oxidation. The filter operates w ith a constant
soot loading and at a constant pressure, and this tem perature is called the balance
temperature.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 44
Exhaust Flow in (m3/hr)
5 m3/hr
E x h a u st F lo w
„ .„
2 m /hr
|
0
20
40
60
80
100
1 m3/hr
12 0
140
Tim e (m inutes)
Figure 4.8 Exhaust Flow Rate vs. Time
700
622° C
DPF Temperature
(C)
600
554° C
500
400
273° C
DPF Temp
200
100
0
0
20
40
60
80
100
120
Time (minutes)
Figure 4.9 DPF Temperature vs. Time
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
Rutm an 45
60
<D
-4-»
DPF AP
o
<
>
o/>
sz 45
o
c
Equilibrium
between soot
accumulation
and oxidation
Q_
O 40
Q
CL
LL
CL
°
25
20
80
100
140
120
160
180
Tim e (m in u tes)
Figure 4.10 Pressure Drop vs. Time
900
777oC
800 -
MW off
700 -
MW off
O
724° C
<u 600 -
ZJ
5 500 -
a>
E 400 -
Q)
u- 300 0.
Q
DPF Temp
200
100
MW on
MW on
-
80
100
120
140
160
Time (minutes)
Figure 4.11 DPF Temperature vs. Time
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
180
Rutm an 46
4.7 Soot Removal and Regeneration Efficiency
The removal efficiency o f the DPF was m easured by taking the mass difference o f two
samples, one before (12.5492 mg) and after the DPF (1.3186 mg). The average removal
efficiency o f 90 % was obtained. The process o f soot sampling is com plex, but because
o f technical lim itations, only simple filtration could be done. The sample lines are traced
with electrical heating tape to ensure that condensation does not adversely affect the
collected soot m ass or chemistry. The dried samples were w eighed and mass averaged.
The mass errors included loss o f volatile organic fractions, loss o f particles smaller than
20 microns, and m ass loss during handling o f filters.
The regeneration efficiency is calculated by the following equation:
P b -P a
N=
---------------
(4.1)
Pb - Po
is pressure before regeneration
P a is pressure after regeneration
P 0 is clean filter pressure
Pb
After calculation the average efficiency for five runs is 81 %. The efficiency will drop as
ash buildup occurs over time. A nother approach to estim ating efficiency involves
calculating a percentage from the initial and final experim ental pressure measurements.
From one set o f m ultiple regeneration cycles, an average value o f 93 % is obtained.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutman 47
4.8 Soot Oxidation Kinetics and Mechanism
Soot oxidation is a very com plex process and does not always produce the classical
com bustion product CO 2 . W hen m odeling soot oxidation, two assum ptions are made to
simplify the model. We assume that all filter channels have uniform soot deposition and
temperature. In our model the diffusion o f oxygen and oxygen adsorption on a surface
site to form a com plex are the m ain steps. These active surface sites at elevated
temperature release CO and free up the surface site. In the proposed model the
assum ption is that not all the soot reacts or all the active sites are occupied by oxygen.
The assum ption is that CO formation is dependent on oxygen concentration and
temperature. A ccording to experimental analytical data, CO 2 is not a product. In all the
regeneration experim ents CO is present in each regeneration cycle as reported in Chapter
4. To verify this, on-line analysis did not detect the presence o f CO 2 .
4.9 Elementary Reaction Mechanism
O 2 + S —►O 2 -S
(4.2) Physical adsorption o f O 2
0 2 -S + S
(4.3) Rate Determ ining Step and surface reaction
CO + 2S
-^overall = £a[02][S]
(4.4) Rate for overall reaction
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 48
4.10 Definition of Variables and Derivation of Overall Rate Law
The overall site balance equation is defined below in equation 4.5. The soot loading
[C-S] is proportional to [C][S]. In this rate model [S] = vacant soot active site, [O 2 -S] is
the oxygen adsorption com plex, and [C-S] is soot deposited on SiC site during loading.
C (t) = [S] + [OrS] + [C-S]
(4.5)
After substitution o f (4.6) into (4.5):
[C-S] = kc[C][S]
(4.6)
N ow yields
C(t) = [S] + **[02] [S]
+ *c[C][S]
(4.7)
Simple substitution o f (4.7) into (4.4) yields the overall rate expression presented in the
next section.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 49
ka [02]
(4.8)
“ ^overall
(1 +
[02] + *, [C]) 2
1.272 x 107 [02]
(4.9)
overall
(1 +6.181 x 105[O2] + 1.994 x 103 [C] ) 2
4.11 Experimental Rate Model Comparison
The proposed m odel was tested against the experim ental data. As shown in Figure
4.12, the experim ental data are in close agreem ent w ith the theoretical model.
0.008
■uy
C
0.007
ai> 0.006
E:o 0.005
E
3 0.004
c
o“ 0.003
CD
CD
i_
H—
o 0.002
Reaction rate-Oxygen-model
a>
Reaction rate-Qxygen-experimental
0
5
10
15
20
Time (minutes)
Figure 4.12 Rate Model Comparison
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
30
Rutm an 50
4.12 Heat Balance
As the m icrow ave energy (1.4 KW ) is absorbed by the SiC, the tem perature increases
at a rate dependent upon a num ber o f factors. The required pow er is dependent on heat
capacity, mass flow, regeneration time, and DPF temperature.
P = M aC pA T
(4.10)
From the Ideal Gas Law and equation 4.10, a simple ratio (eq. 4.13) can be derived to
calculate the am ount o f pow er (P) needed per flow rate (Q).
P i=
[PrM w C p/R K Q O C A Ti/TO
(4.11)
P2 = [Pr M w Cp / R] (Q 2 XAT 2/T 2 )
(4.12)
P i/P 2 =
( 4 . 13)
(T 2 /T 1 X Q 1 /Q 2 X A T 1 /A T 2 )
P 1/P 2 = (896 K / 546 K)(5 m 3/hr / lm 3/hr)(546 K-298 K / 896 K-298 K)
P 1 /P 2 =
3.4
Therefore, the am ount o f pow er required to heat the substrate at the flow rate o f 5 m 3/hr
is 3.4 tim es greater than that required at lm 3/hr.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 51
CHAPTER 5
Conclusions and Recommendations
This thesis reports an experimental study o f DPF regeneration using microwaves.
Diesel exhaust was passed through a SiC DPF and exposed to m icrowaves, resulting in
selective heating o f the soot and ceramic substrate. M ultiple series o f regeneration cycles
were dem onstrated and the DPF pressure returns to a clean state after a five-m inute
regeneration. N o therm al stressing o f substrate was encountered. D ata obtained from the
emissions analyzer produced the kinetic and oxidation mechanism.
5.1 Conclusions
The study has resulted in the following conclusions:
1. The m icrow ave regeneration technique using SiC m onolith w all-flow filter is
effective and reliable. The temperature profile indicates that m icrowave
penetration to the filter is good and heat distribution is uniform w ith no hotspots.
The high therm al conductivity o f SiC also contributes to reduce the temperature
gradients and enhances m axim um DPF tem perature for soot oxidation.
2. A regeneration period o f about five m inutes is sufficient for filter regeneration
because heating the filter to soot ignition temperature o f greater than 500° C is
fast enough, and it prevents filter from overheating.
3. A proper w aveguide design enhances the selective heating o f the DPF and soot
and provides sufficient temperature to oxidize soot.
4. O xygen concentration has a significant effect on regeneration efficiency;
therefore, off-line regeneration by using air is m ost effective.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 52
5. On-line regeneration in the test unit has experienced heat loss from the
instrum entation including control valve, flow meters, and pipefittings. It requires
m ore m icrowave energy (KW) to heat exhaust gas to the soot ignition
temperature.
6. The proposed oxidation model fits the experim ental data, and the oxidation rate
depends on the am ount o f active sites (soot loading) and the am ount o f oxygen.
5.2 Recommendations for Future Work
The following are recom m ended for future study:
1. D esign a com m ercial full-size DPF system and m ount closer to engine exhaust to
m inim ize heat losses, w hich reduces m icrowave energy for regeneration;
2. D esign a com pact m icrowave applicator suited for com m ercial use;
3. D evelop a catalyzed DPF to reduce the soot ignition temperature.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 53
REFERENCES
Abdul-Khalek, I.S., D.B Kittelson, B.R. Graskow, Q. Wei. and F. Brear. 1998. Diesel
exhaust particle size: M easurem ent issues and trends. SAE 980525.
Abthoff, J., H. Schuster, and G. Loose. 1985. The regenerable trap oxidizer- an
em ission control technique for diesel engines. SAE 850115.
Babu, V.S., L. Farinash and M.S. Seehra. 1995. Carbon in diesel particulate matter:
Structure, m icrow ave absorption and oxidation. Journal o f Material
Research 10: 1075-1078.
Baumgard, K.J., and J.H. Johnson. 1996. The effect o f fuel and engine design on
diesel exhaust particle size distribution. SAE 960131.
Broome, D. and I. Khan. 1971. The mechanism o f soot release from combustion o f
hydrocarbon fuels. London: Institution o f M echanical Engineers.
Bueno-Lopez, A., M. M akkee, K. Krishna, and J. A. M oulijn. 2005. Enhanced soot
oxidation by lattice oxygen via La
doped Ce02- Journal o f Catalysis 230: 237-
248.
Burtscher, H. 2001. Literature study on tailpipe em issions m easurem ent for diesel
engines. D one for European Union Particle M easurem ent Program m e (PMP).
Carrascull, A.L., M.I. Ponzi, and E.N. Ponzi, 2003. Catalytic com bustion o f soot
on KNCE/ZrCE catalysts. Effect o f potassium nitrate loading on activity.
Industrial and Engineering Chemistry Research 42: 692-697.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 54
Chatterjee, S., R. Conway, S. Viswanathan, and T. Jacobs. 2004. Diesel
particulate filter technology for low tem perature and low N O x/PM
applications. Diesel Engine Em issions Reduction (DEER) Conference,
Coronado, CA.
Chunrun Z., M. Jiayi, C. Jiahua, L. Lunhui, L. Junmin, and L. Chengbin. 1994.
Studies on regeneration o f diesel exhaust particulate filters by m icrowave
energy. SAE 941774
Code o f Federal Regulations (CFR). 1997. National prim ary and secondary ambient
air quality standards, final rules. Title 40, Parts 50 - 53 and 58.
Dockery, D. W., C. Pope, X. Xu, J.D. Spengler, M. Fay, and F. Spiezer. 1993. An
association betw een air pollution and m ortality in six U.S. cities. New England
Journal o f Medicine 329: 1753-1759.
European U nion C om m ission Directive 2001/27/EC. Official Journal o f the European
Union, Brussels.
Field, M. A., D.W. Gill, B.B. M organ, and P.G. Hawksley. 1967. Combustion o f
pulverized coal. London: British Coal U tilization Research Association.
Fino, D. and V. Specchia. 2004. Compositional and structural optim um design o f a
nanostructured diesel soot com bustion catalyst for a fast regenerating
Trap. Chemical Engineering Science 59: 4825-4831.
Gamer, C.P. and J.C. Dent. 1989. M icrowave assisted regeneration o f diesel
particulate traps. SAE 890174.
Gamer, C.P. and J.C. Dent. 1990. D evelopm ent o f a m icrowave diesel particulate
trap regeneration system. SAE 905446.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 55
Ghio, A. J., C. Kim, and R. Devlin. 2000. Concentrated am bient air particles induce
m ild pulm onary inflam m ation in healthy hum an volunteers. American Journal
o f Respiratory and Critical Care Medicine 162: 981-988.
Haralampous, O. and G.C. Koltskis. 2002. Intra-layer tem perature gradients
during regeneration o f diesel particulate filters. Chemical Engineering
Science 57: 2345-2355.
H ealth Effects Institute. 1995. Diesel exhaust: A critical analysis o f emissions,
exposure and heath effects. Cambridge, MA: H ealth Effects Institute.
Hiroyasu, H., T. Kadota, and M. Arai. 1983. D evelopm ent and use o f a spray
com bustion m odeling to predict diesel engine efficiency and pollutant
em issions (Part I: Com bustion modeling). B ulletin o f the Japanese Society o f
M echanical Engineers 26: 569-575.
Hoek, G, S. Brunekreef, S. Goldbohm, and P. Fischer. 2002. The association between
m ortality and indicators o f traffic related air pollution in a dutch cohort study.
Lancet 360: 1203-1209.
Jimenez, R., X. Garcia, C. Cellier, P. Ruiz, and A.L. Gordon. 2006. Soot
com bustion w ith K/M gO as catalyst II. Effect o f K-precursor. Applied
Catalysis A 314: 81-88.
Kittelson, D.B. 1998a. Engines and nanoparticles: A review. Journal o f Aerosol Science
29: 575-588.
Kostoglou, M., P. Housiada, and A. G. Konstandopoulos. M ulti-channel simulation
o f regeneration in honeycom b monolithic diesel particulate filters.
Chemical Engineering Science 58: 3273-3283.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 56
Laversin, H., D. Courcot, E.A. Zhilinskaya, R. Cousin, and A. Aboukais. 2006.
Study o f A ctive Species o f Cu-K /Zr 0 2 Catalysts involved in the oxidation o f
soot. Journal o f Catalysis 241: 456-464.
Leocadio, I. C., S. Braun, and M. Schmal. 2004. Diesel soot com bustion on
M o/A l 2 0 3 and V/AI 2 O 3 catalysts: Investigation o f the active species. Journal o f
Catalysis 223: 114-121.
Ma, J., M. Fang, P. Li, B. Zhu, X. Lu, and N. T. Lau. 1997. M icrow ave assisted
catalytic com bustion o f diesel soot. Applied Catalysis A 159: 211-228.
Meredith, R.J. 1998. Engineers ’ handbook o f industrial microwave heating. London:
IEE Publication.
Merkel, G. A., W. A. Cutler, and C.J. Warren. 2001. Thermal durability o f wall flow
ceramic diesel particulate filters. SAE 2001-01-0190.
M etaxas, A.C. and R.J. Meredith. 1983. Industrial microwave heating, 1st edition.
London: IEE Press Inc., 2-9.
Nagle, J., and R. F. Strickland-Constable. 1962. O xidation o f carbon between
1 0 0 0 -2 0 0 0 ° C. Proceedings o f the Fifth Carbon Conference, Part I. 265-325.
Nauss, K. M. 1995. Critical issues in assessing the carcinogenicity o f diesel
exhaust: A synthesis o f current knowledge. Diesel Exhaust: A Critical
Analysis 11-61.
National Institute o f O ccupational Safety and H ealth (NIOSH) 1996. Elemental carbon
(diesel exhaust) 5040, NIOSH manual o f analytical methods, 4th edition.
W ashington, DC.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 57
Pope, C.A., R.T. Burnett, M. J. Thun, and E.E. Calle. 2002. Lung cancer,
cardiopulm onary m ortality and long term exposure to fine particulate air
pollution. Journal o f the American Medical Association 287 : 1132-1141.
Popouri, S and M. Hendrichsen. 2001. D evelopm ent o f a m icrow ave assisted
particulate filter regeneration system. Diesel Engine Em issions Reduction
(DEER) Conference, Detroit, MI.
Schwartz, J. 1994. W hat are people dying o f on high air pollution days?
Environmental Research 64: 26-35.
Voss, K.E. 1995. Catalytic oxidation o f diesel particulates with base metal
oxides. In Catalysis and automotive pollution control III. Elsevier: Amsterdam
Press, 499-505.
Wachter, W.F. 1990. A nalysis o f transient em ission data o f a 1991 m odel year
heavy duty diesel engine. SAE 900443.
Walsh, M.P. 1999. Global trends in diesel em ission control: A 1999 update.
SAE : 1999-01-0107.
W alton, F.B., P.J. Hayward, and D. J. Wren. 1990. Controlled energy deposition in
diesel particulate filters during regeneration by means o f m icrowave
irradiation. SAE 900327.
Watson, J.G., J.C. Fujita, L.W .Zielinska, W. Richards, W. Neff, and D. Dietrich.
1998. N orthern front range air quality study (NFRAQS) final report
executive summary. DRI Document 6580-685-8750.IF2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutman 58
W orld H ealth Organization. 2002. W orld health report. Geneva.
Zhang-Steenwinkel, Y., L.M. V an der Zande, H.L. Castricum, A. Blick,
R.W . V an D er Brink and G. D. Elzinga. 2005. M icrow ave assisted in situ
regeneration o f a perovskite coated diesel soot filter. Chemical Engineering
Science 60: 797- 804.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 59
Appendix I
Diesel Fuel Specifications
Boiling Range
150° C
Density
6.8 lb/gal
Viscosity
AP 3
voc
825 g/L
Flash point
52° C
Benzene
0.01 %
Toluene
0.05 %
Ash
1 .4 0 %
Sulfur
300 ppm
Cetane N um ber
43
Nonane
10%
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 60
Appendix II
Component Sources
Quantity
1
1
1
1
5
3
1
1
2
2
1
1
1
1
1
1
1
1
5
Description
Model
Spec
Generator
G PD 6000P
6.9 KW
D iesel Engine
15LD400
O ne cylinder
E m issions Analyzer
T esto-350
n/a
R esearch Control Valve 759-TLDA
Air
Digital Panel Meter
APM 765
n/a
P ressure Transmitter
600 T
n/a
MW Oven
R330EK
1400 W
PM S am ple Pump
Hi Lite 30
10-30 L/Min
Pneum atic valve
62 series
n/a
Rotam eter
n/a
320-530
PC
Precision 370
P4 3.2 GHz
SiC DPF
200 cpi
50 x 150 mm
Guard filter sam pling
NXDDS21
3" x 10"
line
Data Acquisition
LabView
Version 8
DPF Holder
Custom
Quartz
S S filter holder
LS-47
S S 306/ 47 mm
E/P Transducer
77-3
DC/air
Flow controller
840-L
0-10 L/min
T herm ocouples
JM Q SS-020
Mini junction
Maker
G enPro Power S ystem s
Lombardi
T esto
BadgerM eter
Acromag
ABB
Sharp C onsum er Electronics
SKC
Parker
SKC
Dell
CTI
Nett T echnologies
National Instruments
Technical G lass Products
Cole Palmer
Moore
Sierra Instruments
O m ega Engineering
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 61
Appendix III
DPF Specifications
Material
100% SiC
CPSI coding
150
CPSI - calculated cells per square inch
146
C PS cm - calculated cells per square centim eter
22.65
Cell size, wall width - mm
1.6 x 1.6
Wall thickness - mm / 1000/inch
0.5/20
Pore siz e - pm / 1000/mm
12-15
Filtration efficiency - % - clean filter - PM 10
>98
Filtration efficiency - % - 1 0 % loaded filter
>99
Specific SiC w eight - m a ssiv e material density kilo/dm3
3.2
Specific DPF w eight - porous wall density - kilo/dm3
1.8
Monolith w eight - bulk density - kilo/dm3
0.85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rutm an 62
BIOGRAPHICAL NOTE
Daniel A. R utm an was bom in M inneapolis, M innesota, on July 23, 1965. He
com pleted his Bachelor o f Science degree in Chem istry in 1989 from The University o f
Texas at Austin. A t present, he is a candidate for a M asters o f Environm ental Studies in
Civil Engineering at Lam ar U niversity in Beaum ont, Texas.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 499 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа