PB31-191116
Selective Catalytic Reduction and
°.Cx Ccnt-ol x-i Japan
-.dlen Cc rp.
Austin,
Prepared for
Industrial Environmental Research Lab.
Kesearcn Triangle Park, NC
Mir 81

-------
ll f
DCN 81-202-001-27-07	EPA-600/7-81-030
March 1981
SELECTIVE CMALYTIC
REDUCTION AND \0,
CONTROL IN JAPAN
EPA Contract 68-02-3171, Task 10
Prepared by:
Gary D. lones
Radian Corporation
8501 Mo-Pac Blvd.
Austin, Texas 78758
EPA Project Officer:
J. David Mobley
Environmental Protection Agency
Industrial Environmental Research Laboratory
Researcli Triangle Park, NC 27711

-------
TECHNICAL REPORT OATA
(PUatc read	cwii on tne tntrst before compieting)
i ncpontno
EPA-600/7-SI-030
4 title and subtitle
Selective Catalytic Reduction and NOx Control in
Japan
la. HEClfUNT-S ACCfciSlOW NO.
PB81—191116
5 RCP-0
68-02-3171, Task 10
12 SPONSORING AGENCV NAME AND AOORESS
EPA, Office of Research ar.d Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13 TYPE O* »t.*ORT A .to »-£RlOCI COVEREO
Task Firial; 3/80-1/81	
14. SPONSOr.'NCi AiSvtf CCOC
EPAAOO/13
,5 supputvENTARv notes	project offlcer is j.	M0blev, Mail Drop 61, 919/
541-2915.
16 ABSTRACT
The report documents the travels of a four-member i.tudy team ill Japan
during March 1980 to assess NOx flue gas treatment (FGT) technology and related
areas. Overall goals of the study were to obtain new information on current issues
concerning application of FGT technology and to update information previously pub-
lished. A total of 28 equipment vendors, process operators, government agencies,
and industry groups were contacted. Substantial recent progress was discovered
with regard to commercial applications of selective catalytic reduction (SCR) tech-
nology to gas- and oil-fired boilers. There are several applications where SCR sys-
tems are operated continuously and are successfully removing 80% of the NOx from
the flue gas stream. Current development and demonstration efforts are aimed at
applying SCR technology to coal-fired boilers since that fraction of Japan's total
electric power generation is expected to increase to 12. 5% in 1995 and since most of
the new coal-fired boilers will use FGT technology for NOx control. Since four SCR
systems on coal-fired boilers are scheduled'to start up in 1980 and 1981, the Japan-
ese activity in the NOx control field should be of considerable interest in the U.S.
for at least the next 4 years.
17.
KEY WORDS AND DOCUMENT ANALYSIS
1 DESCRIPTORS
b IDENTIFIERS/OPEN ENOEO TERMS
c. COSATI Field/Croup
Pollution Boilers
Nitrogen Oxides Fossil Fuels
Flue Gases
Gas Scrubbing
Catalysis
Reduction
Pollution Control
Stationary Sources
Selective Catalytic Re-
duction
13 B 13 A
07B 21D
21B
07A,13H
07D
07C
13 DISTRIBUTION STATEMENT
Release to Public
19 SECURITY CLASS (Thu Report)
Unclassified
21 NO OF PAGES
269
20 SECURITY CLASS fThtl page)
Unclassified
32. PRICE
EPA Form 2220-1 ($-73)

-------
DISCLAIMER
This report has been reviewed by the Industrial Environ-
mental Research Laboratory (RTP), U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the views
and policies of the Agency, nor does mention of trade names
or commercial products constitute endorsement or recommen-
dation for use.
ii

-------
CONTENTS
Sect ion	Page
Figures		v
Tables		ix
ACKNO.vT-EDCEMENTS	xii
ABSTRACT	Xiii
1.	S'JMMARY		1
2.	INTRODUCTION'		13
3.	PROCESS ANT) EQUIPMENT VENDORS		17
U. INSTALLATION'S WITH NOx CONTROL SYSTEMS			29
5. GOVERNMENTAL AGENCIES AND INDUSTRY GROUPS		39
APPENDIX A - PROCESS AND EQUIPMENT VENDORS		43
Babcock-Hitachi K.K		45
Catalyst and Chemicals Industries Co		53
Ebara Corporation		59
Fuji Electric Co	
Gadelius K.K		81
Hitachi Zosen		87
Ishikawajima-Harima Heavy Industries			• 93
JGC Corporation	"f		105
Japan Shell Technology		H3
Kawasaki Heavy Industries		119
Mitsubishi I.2avy Industries		127
NGK Insulators		135
iii

-------
CONTENTS (continued)
Section	Page
APPENDIX B - INSTALLATIONS WITH \0v CONTROI SYSTEMS		147
Chubu Electric	 149
Chugoku Electric		159
Electric Power Development Company	 177
Fuji Oil		195
Hokkaido Electric		201
Toa Nenryo Kogyo K.K		207
Tokyo Electric	 211
APPENDIX C - GOVERNMENTAL AGENCIES AND INDUSTRY GROUPS	 219
Central Research Institute of the Electric Power Industry	 221
Japan Environment Agency		229
Aichi Environmental Research Center		235
Kanagawa Prefectural Government		241
Yokohama City Pollution Control Bureau		247
APPENDIX D - JAPANESE SCR PROCESS VENDOR CONTACTS AND US AFFILIATES...	253
iv

-------
FIGURES
N'unbe r	Page
1	Comparison of conventioml and >^oaified air preheater
d6siK"£>					25
2	Typical construction of catalyst beds in reactor
vessel (IHI)		32
3	Kanagawa prefecturdl government section air pollution
monitoring centei		42
4	AsseEbling and lond'.ng procedure of catalyst		50
5	Honeycomb catalyst! -l.inufactu/ed by Catalyst and Chemicals....	57
6	Flow sheet for EBARA electron beam process		62
7	Diagram of reaction mechanism for EBARA process		63
8	Arrangement of electron accelerator and reactor		64
9	Removal of SO* and N0>- vs. gas rotation ratio		66
10	Continuous operating data for EBARA pilot unit		67
11	NOx and SOx removal levels achievable by the EBARA
electron beam process		68
12	Schematic of Fuji Electric NH 3 Analyzer		74
13	Influence of probe material on N'H ¦> concentration observed		75
14	Sample flow diagram of Fuji Electric NH3 Analyzer		77
15	NDIR differential analyzer for NHa measurement		78
16	Comparison of conventional and modified air preheater
designs				84
17	Operation of air preheater modified for use with low dust
load		84
v

-------
FICl'RES (continued)
Number	Page
18	Test results of Hitachi Zosen Pilot Unit at Albany, Ceorgia... 90
19	Example of IHI steel catalyst holder	 95
20	Fixed bed type reactor for coal-fired applications (vertical
down flow)	 09
21	Fixed bed type reactor for oil-fired applications
(horizontal gas flow)	 100
22	Split economizer arrangement for temperature control on
IHI boiler	 102
23	Construction schedule for Kudamatsu No. 2 ie-NOx Plant
(Retrofit)	 103
24	Construction schedule of Kudamatsu No. 3 de-NOx Plant
(New Boiler)	 104
25	Activity comparison of high and low temperature catalysts
by JCC	 108
26	Activity of low temperature catalyst (JP 501) as a function
of temperature	 109
27	Arrangement of gas channels and catalyst in parallel
passage reactor	 HO
28	KHI catalyst types	 121
29	Arrangement of catalyst cases in reactor	 123
30	NOx removal efficiency and effluent NH3 vs. NH3/NOX mol
ratio	
31	Status of orders received for MHI NOx removal plants	 130
32	Flue gas treatment alternatives for coal-fired .boilers	 131
\
33	Characteristics of honeycomb substrate (coated type)	 133
34	NGK hexagonal honeycomb catalyst substrate (200 mm square x
350 mm, 7 ram hydraulic diameter, 0.8 mm wall thickness)	 139
35	NGK square honeycomb catalyst substrate (150 mm square x
350 mm;	 140
vi

-------
FICURES (continued)
N'unber	Page
36	Conversion of SO; to SO* by catalyst as a function of gas
temperature	 14*!
37	Effect of space velocity on NOx reduction efficiency
(Small scale tests)	 144
38	Effect of gas temperature on NOx reduction efficiency
(Small scale tests)	 145
39	Effect of ^JHt/NOx ratio on NOx reduction efficiency and
NH 3 carryover	 145
40	Catalyst and reactor arrangement on i'4 boiler, Chita
Power Station	.	 155
41	Truck djiivery and unloading of ammonia		155
42	Hoist arrangement for catalyst changing		156
43	Layout of SCR reactor on 05 boiler, Chita Power Station		155
44	SCR reactors at Chugoku Electric, Kudamatsu Power Station		162
45	NH3 Storage and Vaporization System at Chugoku Electric's
Kudamatsu Power Station	 163
46	SCR reactor at Shimonoseki Power Station	 167
47	Reactor, air preheater and ESP at Shimonoseki Power Station... 168
48	Plot view of No. 1 boiler, de-NOx system, ESP and FGD system
at Shimonoseki Power Station	 169
49	Design of gas mixing zone	 171
50	Isometric flow diagram of boiler and SCR reactor at
Shimonoseki Power Station	 172
51	SCR reactor aid catalyst basket at Shimonoseki Power
Station (dimensions in millimeters)	 173
52	Flow diagram of IHI pilot facility at EPDC's Isogo
Power Station	 jg2
53	Flowsheet of Hitachi-Zosen Pilot Plant at Isogo Power
Station	 184
vii

-------
FICURES (continued)
Number	Page
54	Structure of Hitachi Zosen Noxnon 500 and 600 catalyst		185
55	Equipment modifications made at EPDC Takehara Power
Station for SCR systen: installation		186
56	SCR system (under construction) at EPDC Takehara
Power Station		187
5? Comparison of conventional and modified air preheater
designs		189
58	MH1 SCR system applied to industrial boiler at Fuji Oil
Refiner]/		198
59	JGC SCR system applied to CO boiler at Fuji Oil Re/inery		200
60	Plot viev of Hokkaido Electric Coal-Fired Power Station		204
61	Layout of pilot scale flue gas treatment system at Joban
Power Company's Nakoso Station		21<*
62	Schematic of DICA technique for continuous in-duct gas
analysis		226
63	Growth of SCR facilities in Japan		234
64	N'Ox reduction in Aichi prefecture for the period 1974-1978....	239
65	Air pollution monitoring center and information flow
diagram		245
66	Yokohama City pollution monitoring center		250
viii

-------
TABLES
\'ui?bcr	Page
1	Organizations Contacted b" the Study Tear		7
2	Calculated Costs of SCR for a New 700 ftU" Boiler (70 percent
Utilization or A,292,000 MW hr/year, Early 1980 Costs)		ft
3	Summary of NO SCR Installations "Inspected (Existing and
Planned) . . . f		10
4	Subject Areas and Issues Discussed with Japanese Companies		14
5	Vendors Contacted and flieir Products		18
» K
6	Comparison of Homogeneous and Coated Catalyst Elements		21
7	Comparison of Coated Catalyst Substrate Material		k.1
8	Coal-Fired Applications of SCR in Japan		30
9	Investment Costs of Fu] 1-Scale SCR Systems Inspected		35
10	Emission Standards for N0y in Japan as of August 2, 1979
(Source: Japan Environmental Summary, Sept., 1979)		40
11	SCR Installations by Babcock-Hitachi as of March 1980		48
12	Data from Babcock-Hitachi Pilot Plant at EPDC Takehara
Power Station						49
13	Catalyst Requirements as a Function of N0x Removal		52
14	Performance of EBARA Process with High S0x/N0x Concentrations
Using a Simulated Coal-Fired Flue Gas		69
!? Performance Specifications of Fuji Electric NH3 Analyzer		76
16	SCR System Applications by IHI (as of mid-1979)		96
17	Design Data for KHI Catalyst Types			121
18	Effect of Catalyst Formulation on N0X Removal
(380°C, 90% N0X Removal)		122
ix

-------
TABLES (continued)
Number	Page
19	SCR Systems Supplied by KHI	 126
20	Design Data for MHI SCR System at Shiironoseki Power
Station.	 133
21	Properties of NCK Honeycomb Substrate	 141
22	NOx Control at Chita Power Station - Chubu Electric	 152
23	Emission Limitations at Chita Power Station	 153
24	Design Data on MHI SCR System, #4 Boiler, Chita Power
Station	 154
25	Data Summary on Babcock-Hitachi SCR System - #5 and 06
Boiler, Chita Power Station	 157
26	Capital Cost Breakdown of SCR System on No. 3 Boiler
(700 MW) Chugoku Electric, Kudamatsu Power Station	 165
27	Operating Cost Breakdown of SCR System on No. 3 Boiler
(700 MW) Chugoku Electric, Kudamatsu Power Station	 165
28	Design Data for the MHI SCR System Installed at the
Shimonoseki Power Station	 166
29	Cost Breakdown of Selected Modifications at Shiironoseki
Power Station	 175
30	Comparison of Existing and New Fans at Takehara Power
Station	 190
31	Operating Data from Sumitomo Activated Carbon riucebs at
the EPDC Takehara Power Station	 191
32	Conversion of SO2 to SO3 SCR Catalyst	 192
33	NOx Control on Industrial Boilers at Fuji Oil Sodegaura
Refinery	 197
34	Coal-Fired Eoilers Pl.r.'ned by Companies Participating in
Pilot Unit Test at Ndkoso Power Station	 218
35	Targets for Coal Utilization for Electric Power
Generation in Japan (Operating Units)	 223
x

-------
TABLES (continued)
Number	Page
j£ Japjn Environment Agency Standards for Boiler NOx Emissions...	224
37 !!easi rcmenc Methods for HOx		225
3ti Range of Ambient NO' Concentrations in Japan		2"H
39 Emission Standards for N0X in Japan a"? of August ?, 1979		233
fcO N0y Control Terlm imies in Aicbi Prefecture		238
41	Pollutant Measurement Techniques of Ambient Monitors in
Kanagawa Prefecture		244
42	Capital and Operating Costs for Kanagawa Prefectuie Air
Pollution Monitoring Systems		246
43	Major Emission Sources in Yokohama City		249
44	Emission Reduction Requested by Yokohama City to Meet
1985 Target		251
xi

-------
A C KLN OIn'L E DC O ¦ KN T S
The author wishes to acknowledge to the contributions of Jumpei
Ando of Chuo University, David >'obley of EPA-IERL, and Doug Maxwell of TVA
in the planning and execution of this project. The report was reviewed by
these colleagues as well as all of the organizations contacted and represents
the combined efforts of all parties.
xii

-------
ABSTRACT
A four rember study team traveled in Japan during March 1980 to assess
NO^ flue gas treatment technology and related areas. The overall goal of the
study was to obtain new information on current issues concerning application
of this technology and update information previously published. A total of
28 equipment vendors, process operators, governmental agencies, and industry
groups were contacted. It was found that there has been substantial progress
recently with regard to commercial applications of selective catalytic reduc-
tion (SCR) technology to gas- and oil-fired boilers. There are se/eral appli-
cations where SCR systems are operated continuously and are successfully
removing 80% of N"0x from the flue gas stream. Current developments and dem-
onstration efforts are aimed at applying SCR technology to coal-fired boilers
since that fraction of Japan's total electric power generation is expected to
increase to 12.5 percent in 1995 and since most of the new, coal-fired boilers
will use flue gas treatment technology for N"0x control. Four SCR systems on
coal-fircd boilers are scheduled to start up in 1980-81. Thus the Japanese
activity in the N0x control field should provide valuable information to
interested parties in the U.S. in the next four years.
xiii

-------
SECTION 1
SUMMARY
A four member study team consisting of J. David Mobley of EPA/IERL-RTP,
Gary D. Jones of Radian Corporation, J. Douglas Maxwell of TVA, and Dr. Jumpei
Ando of Chuo Universitv traveled in Japan during March 1980 to assess NO* flue
gas treatment technology and related areas. The overall goal of the study
was to obtain new information on current issues concerning application of this
technology and update information previously published. Equipment vendors,
process operators, governmental agencies, and industry groups were contacted
as shown in Table 1.
There has been substantial progress recently with regard to commercial
applications of selective catalytic reduction (SCR) technology to gas- and
oil-fired boilers. Several of these sites were visited where the SCR units
are reliably operating and achieving the design removals. The operators
report that the catalyst life is exceeding the vendors guarantee. Current
development efforts are aimed at applying SCR technology to coal since the
coal-fired fraction of Japan's total electric power generation is expected to
increase from 3.7 percent in 1978 to 12.5 percent in 1995. It is anticipated
that all or most of the new, coal-fired boilers will use flue gas treatment
technology for N0y control in addition to low N0X burners.
A major area of research and development involves minimizing the impacts
of SCR systems on downstream equipment such as air preheaters, particulate
collection devices and SO2 removal equipment. Problems with t-he-air preheater
*
occur when ammonium bisulfate (NHuHSOm) deposits plug and corrode the elements.
NHi«HS0i, is the product of a condensation reaction between NH3, SO3 and H20,
which can occur when the flue gas temperature drops below about 21C°C. Japanese
pilot unit tests have shown that the plugging problem is most severe in units
1

-------
TABLE 1. ORGANIZATIONS CONTACTED BY THE STUDY TEAM
Installations with	Governmental Agencies
Equipment Vendors	de-NOx Systems	and Industry Groups
Babcock-Hitachi
Catalyst and Chemicals
Industries Company
EBARA Corporation
Fuji Ivlcctric Company
Gadelius K. K.
Hitachi Zosen
Is>h Ikaw.iJ tma-Harima Heavy
Industries
JCC Corporation
Japan Shell Technology
Kawasaki Heavy Industries
Kimoto Electric
Mitsubishi Heavy Industries
Hiroshima Research Station
Tokyo Headquarters
NCIK Insulators
Chubu Electric Power Company
CMta Station
Chugoku Electric Power Company
Kudamatsu Power Station
Shimonoseki Power Station
Electric Power Development Company
Tokyo Headquarters
Isogo Power Station
Ta':chara Power Station
Fuji Oil Company
Hokkaido Electric Power Company
Toa Venryo Kogyo Company
Tokyo Electric Power Company
Central Research Institute
of Electric Power Industry
Federation of Electric Power
Companies
Japanese Environment Agency
Japanese Ministry of Inter-
national Trade & Industry
Aichi Prefecture
Kanagawa Prefecture
U.S. Embassy
Yokohama City

-------
vhich fire- coal or high sulfur oil and also remove particulates upstream of
the NO* rcnctor. When particulates art; renoved dowr-Jt rea:n, plugging problems
are slgni f i cant lv reduced. It is felt that Lite par t. uulates produce .i sand-
blasting effect that cleans tne air preheatcr elerrcnts and also that sone of
the NHuHSOi. condenses on the fly ish particles rather than the elements.
Plugging problcrs are reduced or ellcin.itcd by installing soot blowers on
both sides of the air preheater and increasing both the frequency and pressure
of the soot bloving operation. In sone cases, special air preheater designs
will be used in which the inter—.edlate and low temperature zones are manu-
factured as a single eleirc-nc. These have been tested on pilot unit equipment.
A full scale installation is scheduled at EPDC's Takehnra Power Station.
NHj fron an SCR reactor apparently does not impair FGD system performance
although, in sone cases, the wastewater rust be treated to remove nitrogen
coapounds. It is not known if SCR s\stems will affect dry SO2 removal systems
(e.c., spray drying) since these techniques are not used in Japan. The one
apparent adverse impact that nav occu»»»is NTUHSOu affecting the performance
of the downstream baghouse. The effect of an SCR svstem on baghouses is under
investigation. Pilot unit tests are underway; however, data are not yet
available.
Several of the coal-fired SCR applications that -xre under construction
utilize hot-side ESP's for upstreazj particulate removal and there are a variety
of reasons for selecting hot-side particulate removal. These include 1) to
eliminate fly ash from entering the N0< system and potentially causing plugging
or erosion problems; 2} to obtain the capability to remove particulates from
a wide ranee of coals	'-aryir-.g character! Slit_s, and 3) to avoid ammonia
corpounds in the ash that can result when a cold-side ESP is used and affect
fly ash utilization as a by-product and its suitability as a landfill material.
However, cold-side ESP also have unique advantages such ."3 1) lower capital
and operating costs; and 2) allowing the fly ash to reduce or eliminate NH^HSO,,
deposits on air preheaters.
3

-------
With SCR, the flue gas temperature entering the reactor is usually
maintained between about 300 to A00°C. Continuous exposure to higher tem-
peratures can cause sintering of the catalyst material and consequent loss
of activity. Operating at Jess than 300cC can allow NH^HSOi, to deposit in
the catalyst pores, theretv blinding th2 catalyst. Exposure of the catalyst
to flue gas at less than lOO^C can allow water deposition and permanent
darrage. In Japan, the temperature is controlled by either load regulation
or bypassing a percentage of the hot flue gas around the economizer and mix-
ing it with the cooler gas. Another technique that may be applied involves
installing a split economizer and controlling the flow rate through each side
to maintain the desired temperature.
Research and development on catalyst formulations and shapes during
recent years has resulted in sone standardization among the catalyst types
offered. A catalyst formula consisting primarily of oxides of titanium and
vadatiium appears to be universally used. Current research involves adjusting
the mix of these components and adding small amounts of other compounds in
order to meet the specific requirements of each application. The most
important area is that of SO2 oxidation to SC>3. SO3 is undesirable since
it can enhrnce NlUHSCk formation and increase particulate emissions. SO2
oxidation can be a problem with coal-fired boilers where SO2 concentrations
are relatively high. Snail amounts of proprietary additives can be included
in the catalyst formulation or the vanadium content can be reduced in order
to control the anount of SO2 oxidation that occurs.
There has been concern that the catalyst and reactor may plug with ash
when applied to coal-fired boilers. Pilot unit tests have indicated that
plugging is not a problem when honeycomb or pipe shape catalyst is used in a
vertical, downflow arrangement. However, soot blowers will be.in-stalled in
the reactors of current full scale applications as a conservative design
I
measure. Another concern in the US has been catalyst poisoning by tlue gas
components. While it is true that certain alkali metals, such as sodium
and potassium, will slowly poison the catalyst, the concentrations are low
A

-------
enough that catalyst activity vill not be affected during the guarantee
period.
While a pellet shaped catalyst is preferred for gas-fired applications,
the "honeycomb" shape is used for oil- and coal-fired applications. Honey-
comb is a general term for a rectangular parallelepiped element with parallel
channels passing through it in the direction of the gas flow. The catalyst
elements can be supplied in three forms: composed solely of catalyst material
(homogeneous) or composed of either a ceramic or metallic substrate that is
coated with catalyst material. There is no agreement on which type is best
and each offers unique advantages. For example, the homogeneous element will
not lose activity even if the outer layer is lost due to erosion by flyash;
however, eventually the physical strength can be affected. The opposite is
true with the coated variety. Coated catalyst elements also have the poten-
tial for erosion; the element will not be structurally weakened, but catalytic
activity can be lost. The substrate used for the coated catalyst elements is
either a metallic or ceramic material. The metallic material is resistant
to breakage and causes less of a pressure drop than a ceramic substrate.
Alternatively, the ceramic material has better catalyst bonding properties,
is acid resistant and is lightweight. Catalyst life guarantees are usually
one year for coal, one to two years for oil, and two to t'iree years for gas
although the experience on gas- and oil-fired boilers has been that actual
catalyst life exceeds the guarantee.
The labor requirements of the operating, full-scale systems are small.
No additional operating personnel are required and maintenance labor consists
primarily of N~ri3 ana catalyst loading and cleaning the air preheater during
the annual outage. Since there have been no catalyst changes to date, the
labor estimates for this work vary widely. Operators indicate that the SCR
processes themselves are very reliable, essentially 100%. However, in some
cases, a boiler shutdown has been necessary where air preheater plugging has
occurred. In most cases, steps have been undertaken to reduce the plugging
rate to the extent that cleaning is only required during normal boiler out-
ages .
5

-------
Ttte automatic control systems are similar for all of the processes.
Nil 3 injection is controlled by a signal consisting of the product of boiler
load (related to gas flow rate) and the reactor inlet NOx concentration.
For gas and oil applications the outlet Nil concentration is also used as
a trim signal for NHj injection. An NH3 emissions monitor for coal-fired
flue gas has not been perfected although several techniques are under develop-
ment. Usually the NH 3 monitor is a cher.ii luminescent NO monitor with a cata-
lytic converter which converts SH3 to NO. Sulfur oxides apparently interfere
with the monitoring equip,r>ent and there have been problems with the converters
giving incomplete \'H3 conversions.
NO* emissions from boilers utilizing catalytic de-NOx are generally
reduced by 80 percent. Higher reductions are possible, but costs will be
greater for these units. Nrl 3 emissions for oil- and gas-fired applications
are reported to be about 3 to 10 ppm. Emissions from full-scale, coal-fired
facilities will not be known until after the start-up of several units
currently under construction. In addition to NH3 emissions, there is some
concern that other compounds - «?uch as cyanides, nitrosoamines and nitrates -
may also be emitted. SCR svsten vendors and operators, however, were not
aware of any instances where compounds such as these were emitted. The
possibility of a visible ammonium sulfite plume resulting from NH3 emissions
entering a downstream, wet FCD system may be a problem if the slip NH3 is
high (>50 ppm). Plumes of ammonium sulfite are known to occur during certain
atmospheric conditions when NH3 based FGD systems are used. However, the
Japanese have experience with situations in which 10 ppm of NH3 enters the
scrubber and, based on this experience, do not feel that SCR systems will
cause visible plume formation.
Other potential environmental impacts include nitrogen compounds in
the wastewater and catalyst disposal. NHj can enter the wastewater through
the FGD system or from air preheater washwater. In locations where discharge
of this water will cause problems, an activated sludge technique can be used
6

-------
to treat the wastewater. In other locations it may be possible to blend
wastewater containing NH3 with other water discharges. The catalyst disposal
issue has not been fully addressed, l.tiile tha process vendors indicate that
they will dispose of spent catalyst, the specific method of disposal has not
been identified. This is partially due to the fact that, so far, none of the
full scale systems has required a catalyst change.
The costs of SCR systems are dependent upon many factors. In general,
the capital costs for 80 percent N0y control on a low-sulfur-oil-fired boiler
in Japar will range from 3000 to 4000 yen/kW ($12 to 16/kW). Complicated
retrofits will cost more. The costs for coal-fired boilers are estimated to
range from 6000 to 7000 yen/kW ($24 to 28/kW) when installed on a new boiler.
To increase N0X removal from 80 to 90 percent will raise the cost of a system
20 to 25 percent. 90 percent control with low NH3 emissions will increase
the cost by 40 to 50 percent. Catalyst costs often amount to half of the
total capital cost of the system. The cost of catalyst does not change much
from vendor to vendor and a typical value is about $340 per cubic foot (based
on 250 yen per dollar). Estimated annualized costs of SCR systems prepared
by Dr. Jumpei Ando, are shown in Table 2. Ten cases are considered in order
to determine annualized costs as a function of fuel and removal level.
Several SCR installations were inspected by the study team. Data from
these units are summarized in Table 3. As the table indicates, most of the
installations are applied to utility boilers and operate at 80% N0X removal.
Gas- ar>d oil-fired applications are already operative and several coal-fired
applications are scheduled for startup in the near future. In addition to
these facilities, two applications of the Thermal de-NOx process were inspec-
ted. T!ns process involves injecting the ammonia into the high temperature
zone and achieves selective NO reduction without the use of- a" catalyst.' One '
*
application was to a utility boiler at Chuba Electricj's Chita Power Station.
7

-------
TABLE 2. CALCULATED COSTS FOR SCR FOR A NEW 700 MW BOILER
ftoltvr ImIi
00
Cotti
ImuiI Cmi
Coal llfteilffwu
Flo* Pal Plm
1000 Pa'Shr
W( Conri«ri((lMa
Sft, Concml ral|m, ff*
r«rtlro)at» CitntrAtfl
IIm, |rmm/H»'
Irpifi)
CatllTM Trp#
t(i«, mt
«!» taliil»«i, pf*
Vtl) NO. It I > Lltl*
Cataljat lit#, r*"»
Ifirlnr fiaaaar* 0n»f(
tm ft ,0*
Fll»ra)
10*
rr<|)A
woo
*0
IK>
r»i in
% o
J-10
i« n
101
0 1>
J
in
Q It
0	« <1 1A>
1.19 (4 40)
1	)» n v>
i w <» n>
o » (1 n»>
0 ID (0 40)
o m (0 vt)
o to {0 on)
fl tt (7 M)
o n (0 M)
t to (0 «M
7000
to
90
utt*t
> a
1-10
n *
in
l o*
0 If
0	41 <1 R4|
1	JO <4 80)
I *1 r« «i)
i *> (* rt>)
o X (1 »)
o i« (ft II)
0 OA (0 ?»>
o J*	(« **>
o m 3 <« an
l m (* m)
I II <10 !¦))
i « <11 id
o yi 
fl I? (fl 4*)
1	a i* %n
*
n n 
I II <> 00)
J V (1* 7»>
1 10 (70 40]
0 II (7 4«|
o 'i n m>
0	II (0 44)
n 7" (0 *i>
1	m (r *|
0 *) (I 72)
¦i I* 9 nn
rooo
im
IVM
OC«(l ?fl
0 M
I
0 t«
7 71 (ft W)
I ® If W)
I 01 (U 04)
i n 
1	II (4 44)
0 14 <* *~)
r ii io m>
J »Ml \»>
ovtn itn)

-------
TABLE 2 (continued)
Inllic laala
Flu* Ut
1000
i* flew I
IiwiImii Casta
 Cun«M ill Inn,
'tilUdiia CeMfMtc
(Ion, im/h1
K).
C«l«l|«l Tjf*
C«l«ly«t Kilt, ¦
Wli 4*laal.«ta, h*
•r«.« Vblwliy.
1000 hr"'
Uiilfd ItliWt •'
VHIIHU, Hula bl lo
Cilil^l Ufa, yaara
Olbat
(10
fulil.
«• *,U*
Puvar (Mu«f(laa,
piiiiii of bollir
C«t¦)*•!, 10*
<10* 4ullara)*
¦I. »«»" T*»
(10* 4olU««)
IUJU yatiw*
(4sllir«/k«)
fl««4	10* f*o'
(10* feliata)
Cltllflt. 10* fr«'
OO1
teil«i 10* y*o*
(10* ^lUut
OtlHl. 10* Jim1
(I0l fellaial
Tumi, to* r*t
(|04 tollna)
twdl 1t«4 Cm| ,
(at 11 •/!*«>)
louo frt/ita' «f no.
rt*ir*4 4uI1iii/
M aarf I /
1 •
mi
0 14
0 24
I 44 (I) HI
I 10 (I 40)
> >4 (11 It)
I 91 (II 44)
0	II () 41)
1	11 (i M)
0 I* (0 $4)
0 24 <1 04)
J 01 (11 04)
0	70 (J 10)
1	11 (I M)
*Tr»«llo| llw pi fu«lB| ikimj> * but Ul
*Tr««tln| 11m |ii kit! (all ft) ub load
(Ciiialc tMMjroah, aquara I if* (c4wmI dlaartar ~ wall ikl(Lxia)
'AklMll;, 901 f«i»*«l la 4lfllr«l| wlthea* ImhwIri amtaalaa Omi
• l*t|« |n ««|«i Iiu» *11 nl coil ilnct tba R0a
cimmi141lo* aal |U valocllr ai« awl «ll«ra at r«arl»« lalal
*CiK'ral coat (allllaa )(*•/>') • I I fa« |«i, ) I lor oil, id ) ) (a( i
'ImMIii civil 4B|laaarl*t «! •fiiillso
•4j raactoi i wmli| 4wcll
1 no
100
M0
lkn«7vM4
M ai4 l)(
1 I
1)2
0 •)
0 14
2 II (II 24)
} 00 (I 00)
4 II (I* 24)
* ir (ii 4i)
0 71 <) 12)
2 II (II 24)
0 21 <1 01)
0	20 (0 *0)
1	04 -0 20
«4
i J
im
0 41
i
i»
o n
4 a  40)
I >1 <1 *0)
•t|k I Cm!"
2100
XX)
2000
11-2*
I ) 1*4 I 1*
via
2 I
1)2
o to
l
0 24
2 II <11 24)
1	00 (ft U0)
4 II <>• 24)
* II (2? 41)
0 21 (1 12)
2	II (1) 24f
0 21 (I.01)
0 20 (0 00)
« 01 (II 24)
0	41 O W)
1	20 (1 04)
i mo
mo
2000
11-21
1	I Mi I |r
1-10
I I
ID)
0 *4
I
1/0
o n
4 44	(II 04)
2	»	(f 20)
I 14	(21 04)
4 44	(M 44)
I 02 (4 "•)
4 a <12 to)
0	10 (1 20)
9 24 (0 M)
4 02 (24 01)
1	40 O 40)
I SI (I 40)
S«kmiI al famt pMrilU 4y (4* kollir lUlil) fat OMftMilIm «f
»raaa«ta 4raf
'lataraat <101) far laltlal (kii|i af catslfat aaJ 211 (Im1«4I«i I r*irt
fc^rxluIn) af IwwiIwm raat aulbdlag rililyu,
>Cittl)it (Mt	4y III* (fill)
'n 100 000 im/ika
'lUlaly f«r »«"' (20 |t«A« hi fa* |m mmi oil, (1 fra/lw 4f fai eaal)
*Im«4 oa i 210/1
•11 af 40. aa «0|
laalai lady 1*40 C*M«| 4,ltt 000 « kt/pi 70S MlllaMla

-------
TA13LE 3. SUMMARY OF NO* SCK INSTALLATIONS INSPhCTl.D
(EXISTING AND PLANNED)
Coop my SC«ir 'on
TJc nc it lc.it ii>n
tu. 1
site
(V-'>
Sl2<
(HU)
SO K< odv 11
U)
Outl«t sob
(pi«)
V» fvlor
• osl
ll/W)
V w/h. t r«'l 1I
St *r r up
Chubu t LcCtrl Chic-*
#5

700
'00
80
a-io
9-H
1800
V >*v
H-irch 1978

16
I No
700
710
SO
8-10
3-11
1800
\« v.
April 14?8

#«4
on
;oo
700
AO
20
'ill
JfiUO
H< t f of 11
tirch I'JaO
Chuguku tlottrlc kaiiionutsu
V2
Oil
375
)7S
•o
12
IMI
1SO0
Rc WJtlt
April 1979

fl
(HI
700
700
tto
2b
mi
uoo
w
S« pt«i»b«*r
Sh i oyjn«>s4 k {
(\
Coil
1 7*>
17}
r»0
:r,o
Mill
7000
H» Clot It
April 19^
toj I Oi 1 »ia
97
mi
'.0
'.0
87
iS
SUII
-
.
J m'i us 19 /H

"
(O S 0' I
to
10
9;
IS
ire
"
¦»« IUMH
1 u 1 s 1 9 < r 1980
i.J')C Tjn-hirn
91
LO»l
250
l?5
no
60
l-H
% 7000
K
-------
This system is operated continuously and achieves 30% N0x removal. The other
application is to a pipe still heater at the Toa Nenryo Kawasaki Refinery.
This unit is operated only when there is a local smog al,-rt (about 20% of
the time) and achieves a NOx reduction of 30 to 50%.
The catalytic de-NOx units were installed to meet local governmental NOx
emission limitations and are not necessary to comply with the Japanese national
standard. Compliance with local and national environmental agreements is
achieved through a sincere cooperative spirit that exists between industry and
government to solve the country's pollution problems. Local governments main-
tain a very comprehensive and extensive instrumentation system to monitor
ambient pollutant concentrations and emission rates from major emitters.
Although the local governments have no enforcement authority, violations are
rarely, if ever, detected since the companies do not want to jeopardize the
"good will" that exists between the industrial, governmental, and the public
sectors.
11

-------
SECTION 2
INTRODUCTION'
The Environmental Protection Agency is investigating methods of control-
ling S0y enissions from fossil fuel-fired utility and industrial boilers since
these sources conprise a major source of N0x emissions. Coal-fired boilers
are of particular interest since they emit the highest concentrations of N0x.
Currently, NO^ emissions are reduced by the use of special ' low N0x" burners
and other conbustion controls. These techniques are usually capable of 30
to 50 percent redu'-tion of N'0x. Another technique, developed primarily in
J'.pan, treats the flue gas to achieve 80 to 90 percent N0x removal. This
tecnnique is usually termed selective catalytic reduction (SCR) or catalytic
de-N0x since it involves injecting ammonia (NH3) i.ito the flue gas prior to
a catalytic reactor which reduces N0x to N2 according to the following reac-
tions :
4N0 + ANHj + 02 4N2 + 6H20
2N02 + ANH3 + 02 3N2 + 6H20
SCR has been applied commercially in Japan to several oil- and gas-fired
boilers; coal-fired applications will soon be on-line. However, because this
technology has not been applied in the US and since there are several unanswer-
ed questions regarding the application of SCR technology to coal-fired boilers,
J. David Mobley of EPA's Industrial Environmental Research Laboratory, Gary D.
Jones of Radian Corporation, J. Douglas Maxwell of the Tennessee Valley Auth-
ority, and Dr. Jumpei Ando of Chuo University, traveled in Japan during'March','"
1980 to assess the status of flue gas treatment technology for N0x control.
Of partic^ar interest was the selective catalytic reduction technology and its
impact on downstream equipment. The issues of interest for this study are
shown in Table h.
Preceding page blank

-------
TABLE 4. SUBJECT AREAS AND ISSUE 3 DISCUSSED WITH JAPANESE COMPANIES
Design/Operation of Process and Equipment
Hot-side versus Cold-side ESP
Air Preheater Soot Blowing and V'ater Washing
Catalyst Erosion
Catalyst Life
Catalyst Poisoning
Cat.-lyst Composition and Configura' irn
Catalyst Loading Procedure
Reliability
Maintenance Requirements
Labor Requirements
Control Systems
Flue Gas Temperature Control
Effects of Temperature Extremes on Catalyst
80 versus 90 percent NO Removal
SO2 to SOj Conversion
Thermal de-NOx Process Applications
Impacts on Other Equipment
Economizer
Air Preheater
Baghouse
Electrostatic Precipitators (ESP)
Flue Gas Desulfurization (FGD)
Dry S02 Scrubbing
Environmental Impacts
FGD Wastewater
Air Preheater Washwater
Fuming at the Stack
Emissions of NO^, NH3, Cyanides, Nitrosoaniines, Nitrates
Catalyst Disposal
Fly Ash Disposal
Other
Catalyst Cost
System Cost
Fly Ash Utilization
Analytical Techniques for N0X, NH3 and SO3
Indus trial/Civilian/Governmental Relationships
14

-------
The applicability of SCR systems to coal-fired boilers is of particular
importance in the US, since the national emphasis on coal-firing will increase
ambient NOx concentrations. In areas that are noncompliance ir> NOx or hydro-
carbons, both new and retrofit application's of SCR are being considered for
utility and. industrial boilers. Boiler operators have expressed a great deal
of concern over the impact of a:—noma emissions from SCR systens on downstream
equipment. It is known that air preheaters can plug when NHj, SOj and water
combine to form ammonium bisulfate	There is also concern over the
effects of ICH 3 on ESP's, baghouses and FCD systems.
Another factor is the effect of removal level on cost and system design.
Most SCR applications being considered in the US are anticipated to achieve
90 percent NOx removal, whereas SO percent is typical in Japan. In order to
adequately assess US applications, it is necessary to quantify these effects.
The objective of the trip was to develop information on all new issues
and update existing information in areas such as status of development, costs,
N™-33 emissions, CAtalyst and reactor design, and the regulatory framework
driving the development of this technology. To accomplish this objective,
representatives were contacted in three areas:
process and equipment vendors,
• installations with NOx control systems, and
governmental agencies and industry groups.
This report is divided into these same three categories. An overall
sutsaiary is presented in Section 2. iti Sections 3, A, and 5, the results are
briefly summarized for each of the three categories. These sections deal with
15

-------
NOx only, SCR processes and related areas. Detailed meeting reports tor
these and other process typos are contained in the Appendices. Also, in
Appendix D, the principle Japanese SCR process vendor contacts are listed
along with the US licensee.
16

-------
SECTION 3
PROCESS AND EQUIPMENT VENDORS
Process vendors were contacted to determine both the state-of-the-art
technical developirents regarding application of SCR to coal-fired boilers and
recent improvements for oil- and gas-fired applications. Vendors in five
subject areas were contacted as shown in fable 5. These Include all of the
major vendors of NOx-only SCR technology as well as two alternative tech-
niques. Also, two of the four independent manufacturers of SCR catalyst were
contacted.
Instruments for the continuous measurement of NH3 are still in the
development stage. Consequently, meetings with these instrument vendors were
conducted. SCR process vendors were also questioned about instrumentation.
A regenerative air heater vendor whs contacted to learn about state-of-the-art
designs for use after SCR processes since plugging of convent!~ral air heaters
is a significant problem in some applications.
SELECTIVE CATALYTIC REDUCTION - PROCESS AND CATALYST VENDORS
This section presents the results of several meetings with SCR process
vendors and catalyst manufacturers. Since NOx-only SCR systems are the focus
of this repori., only these are discussed here. Individual meeting reports
for all vendors contacted appear in Appendix A.
All of the major suppliers of SCR processes were contacted as were two
of the four catalyst manufacturers. The two catalyst vendors not contacted,
due to time constraints, are Sakai Chemical and Nihon Shokubai. Some of the
SCR process vendors produce their own catalyst including Babcock-Hitachi and
Hitachi-Zosen, while others buy catalyst from one or more catalyst manufactures.
17

-------
TABLE 5. VENDORS CONTACTED AND THEIR PRODUCTS
Product	Vendors Contacted
Selective Catalytic Reduction
NOx Control Equipnent
(SOx only)
Babcock-Hitachi
Hitachi Zosen
Ishikawajima-Harima Heavy Industries
JCC Corporation
Kawasaki Heavy Industries
Mitsubishi Heavy Industries
Alternative NOx Control
Techniques
EBARA Corporation
(Electron Beam, SOx/NOx)
Japan Shell Technology
(SOv/NOx)
Catalyst Vendors
Catalyst and Chemicals Industries Co.
NGK Insulators
Instrument Vendors
Fuji Electric Co.
Kiraoto Electronics
Air Heaters
Cadelius K. K.
18

-------
None of the catalyse manufacturers are associated solely with one process
vendor.
Catalyst
All of the vendors use a catalyst to facilitate the reaction between
injected NH 3 and NOn. Both the catalyst manufacturers and process vendors
conduct their own research and development programs. However, when the
catalyst is sold it is prepared according to the specifications of the pro-
cess vendor.
The catalyst formulations are based on Ti02 and V2O5 as the major com-
ponents. The V2O5 has the capability of also catalyzing the oxidation of SO2
to SO3 and is used extensively in the sulfuric acid industry. With a V2O5
rich catalyst, the conversion can be as high as 5 percent which at high SO2
concentrations results in a significant amount of SO3 exiting the SCR reactcr.
High SO3 concentrations can cause problems due to NHi,HSOr, pluggage of air
preheaters downstream of the reactor. For applications where SO2 is present,
the catalyst formulation is modified to include small quantities of other
proprietary compounds that suppress the SO2 oxidation reaction. This suppres-
sion is accomplished at the expense of NOx removal; for example, reducing SO2
oxidation from 2.5 to 0.5 percent will reduce the NOx removal efficiency by
20 percent for the same volume of catalyst.
The honeycomb, plate, and pipe shapes are preferred for use with oil-
and coal-fired applications, while for gas-fired applications the pellet
shape is preferred over these more complex arrangements.. Most of the reseaich
emphasis is on the oil- and coal-fired applications since these applications
are more difficult to deal with. The most popular shape among the catalyst
vendors seems to be the"honeycomb for reasons of strength and ease of handling.
The honeycomb shape consists of a square block with parallel channels passing
through it. The Dlock is typically 150 to 500 mm on a side and up to 1000 mm
in length. The channels can be either square, hexagonal or triangular in
19

-------
shape. The honeycomb elements can be produced as a ceramic or metal substrate
coated with catalyst material or as a homogeneous form composed purely of
catalyst material. The honeycomb shape is easier to assemble into modules
since consideiably fewer pieces have to be assembled ard they can be easily
stacked within the module.
There is no agreement as to which catalyst type, homogeneous or coated,
is best since each offers distinct advantages, as shown in Tables 6 and 7. The
homogeneous catalyst is softer and can be eroded by fly ash; however, the newly
exposed material is still catalytically active. Continued erosion will eventu-
ally structurally weaken the catalyst element. To overcome erosion degradation,
the first few centimeters of the inlet face can be impregnated with a hardening
material. Coated catalyst elements also have the potential for erosion. The
element will not be structurally weakened, but catalytic activity can be lost.
Ceramic is claimed to be superior to metal as a substrate material since the
surface is rougher and the thermal expansion is similar to that of the catalyst.
These two properties give a better bond between the catalyst and substrate
which minimizes exfoliation and erosion of the catalyst. However, the bietallic
substrate has the advantage of thinner walls which allow a smaller volume of
catalyst to be used.
Since erosion is experienced only at the inlet face, one method of pre-
vention is to install a "dummy" layer at the reactor inlet. This layer
consists of hard, inert material in the shape of catalyst elements and absorbs
the erosive force of the fly ash. This technique should work with both homoge-
neous and coated catalyst elements.
For coal-fired applications, there is no definite agreement as to which
type of catalyst, coaled or homogeneous, is be«:t. Of the systems currently
under construction on «.oal-fired boilers, two are using a homogeneous catalyst
and two are using a coatea ratalyst.
The catalyst manufacturer will guarantee the catalyst performance to
the process vendor who in turn, guarantees the process performance to the
customer. The actual guarantee period depends on the characteristics of each
20

-------
TABLE 6. COMPARISON OF HOMOCENEOUS AND COATED CATALYST ELEMENTS
Coated	Homogeneous
Luwer cost (less catalyst	Less affected (at least
material used)	initially) by erosion
Lighter weight
Creater physica1 strength
TABLE 7. COMPARISON OF COATED CATALYST SUBSTRATE MATERIAL
Metallic	Ceramic
Resistant to breakage	Lower potential for exfoli-
ation
Lower pressure drop due to	Acid Resistant
greater open frontal
area
Lightweight
21

-------
application; however, tlie guarantees generally seem to be two tc three years
for gas-fired application, one to two years for oil- and one year for coal-
fired units. It appears that the guarantee perioc can be affected by NOx
removal level, particulate loadings, sulfur concentration, gas velocity,
and temperature range.
It is generally assumed that the catalyst manufacturer will dispose of
spent catalyst and the cost for this is apparently included in the cost of
the replacement catalyst. This issue Is still somewhat unclear. A definite
disposal method has not yet been established primarily because none of the
comnercial SCR units has required a catalyst change. The manufacturer may
insist on this method in order to protect the confidentialty of the formula-
tion; however, transporting spent catalyst from the US to Japan for disposal
may be prohibitively expensive. The vendors indicated that the valuable
components may be recovered for reuse, but not as catalyst material. Some of
the metals used may present waste disposal problems due to Japanese disposal
regulations. They may also be disposal problems in the US due to regulatory
constraints depending on the catalyst conposition. Since the complete compo-
sition of the catalyst is proprietary, it is impossible to determine if, in
the US, the ipent catalyst can be landfilled or must be treated.
With coated catalysts, it may be possible for	the substrate to be cleaned
and reused. Since the volume of catalyst material	is much less than with
homogeneous elements, disposal costs may be less.	The catalyst costs vary
somewhat depending on the type, but in general the	cost is about ¥ 3 x 106/m3
($340/ft3) for oil-fired applications.
Reactor Design
In general, there are two types of reactors that are used for oil- and
coal-fired applications, horizontal flow and vertical flow. While either is
applicable to oil-fired units, the vertical downflow reactor is preferred
for coal-fired applications. The catalyst in most cases is assembled into
22

-------
¦modules which are steel containexs containing several catalyst elements.
The modules are of a size and weight that can be conveniently handled and
loaded into the reactor. The modules are loaded either from above using a
hoist, or from the side using a special dolly that rolls oti rails or struc-
tural steel within the reactor, ""^e modules are supported within the reactor
by structural beams at each leve of catalyst. Usually three to four levels
of catalyst are used.
Specific design values such as space velocity, gas velocity, catalyst
pitch, NHj:NOx mole ratio, location of the reactor relative to the boiler,
etc., are unique for each application and are specified by the vendors on a
case by case basis.
N'Ov: Removal
NOx removals of 90 percent and higher are potentially obtainable with
SCR systeirs. The systems installed in Japan usually operate at 80 percent
removal or lower (as low as 50 percent). The vendors indicate that 90 per-
cent removal can be achieved, but will cost more since more catalyst and a
higher NHaiNOx mole ratio will be necessary. Keeping the NHj emissions low
at 90 percent NOx removal levels will also require either additional catalyst
or some other technique for NHj control.
Temperature
Temperature is important in the operation of SCR systems as there are
both upper and leer limits. If the ueraperature is allowed to get too high,
the catalyst particles will undergo sintering and the catalyst activity will
be reduced. Generally, temperatures in the range of 450°C will not cause
damage if the exposure time is short, 1 to 2 hours. Low temperatures, <300°C,
can allow NfUHSOi. to form on the catalyst if sufficient SO3 is present. This
can cause blinding of the catalyst surface, but the deposit can be removed by
raising the flue gas temperature to >300°C. Temperatures <100<>C can allow
water deposition and permanently reduce the catalyst activity.
23

-------
The desired operating temperature range is 350 to 400°C and for some
applications a temperature control system is recommended by the vendor in
order to maintain a minimum temperature. Control is achieved by using an
economizer bypass, load regulation, or a split economizer. The split econo-
mizer can either be a large and small economizer arranged in parallel with
variable gas flow through each, or two separate economizers, one upstream and
one downstream of the SCR reactor.
Costs
The costs quoted for SCR systems by all of the vendors are similar.
Investment costs for a typical 80 percent control system applied to an oil-
fired boiler are about ¥3000 to 4000/kW ($12 to $16/kW). A similar system
applied to a coal-fired boiler would be about ¥6000 to 7000/kW ($24 to $28/kW).
These figures are for new units. The cost of SCR systems installed on existing
boilers can be substantially higher. No general figures are available for
retrofit systems due to the high variability between individual installations.
One vendor indicated that investment costs for a 90 percent NOx removal
system are 20 to 25 percent higher than those for an 80 percent system applied
to an oil-fired boiler if increased NH3 emissions are allowable. For 90 per-
cent NOx control with low **H3 emissions the costs will be AO to 50 percent
greater, rather than 20 to 25 percent.
AIR PREHEATER VENDORS
There are two air preheater manufacturers in Japan: Gadelius K. K.
(Ljungstrom type) and Ishikawajima-Harima Heavy Industries (Rothemuhle type).
Gadelius is preparing an air preheater design for EPDC that will resist
plugging by NHuHSOu, which is a common problem when an SCR system is located
upstream. This is especially true if particulate removal also occurs upstream
of the air preheater since particulates tend to clean the elements. Gadelius
has performed pilot unit tests which indicate that with coal-fired units,
2<»

-------
NH^HSO!, plugging is not a problem when a cold ESP is used, but is a severe
problem if a hot ESP is used. The plugging occurs at the interface between
the elements of the intermediate and cold teiiperature zones.
Both situations will be tested soon on full-scale equipment. At Chugoku
Electric's coal-fired Shiir.onoseki Power Station, an SCR will be installed up-
stream of the air preheatcr and cold-side ESP. The SCR system is designed
for 50 percent removal and, therefore, slip NHj (unreacted NH3) is likely to
be low. An SCR system will be installed downstream cf a hotside ESP at EPDC's
Takehara Power Station where a modified air prcheater design will be used.
Two air preheaters of this design will be installed and other electric com-
panies indicate that this design will be used when SCR is applied in conjunc-
tion with a hot-side ESP.
The modified design consists of combining the intermediate and cold
sections into a single element. This permits more effective soot blowi'ig
which, for the new design, is done from both the hot and cold sides. ihe
large element is made of corrosion resistant, coated material similar to typi-
cal cold end elements and a spacing of 3.5 mm issued. The design is shown
conceptually in Figure 1.
Intcrmedi ate
(2.5 nn)
Cold
(6 mm)
Soot
Blower
Conventional Arrangement of Heat Transfer Elements
Soot _
Blower
Combined
Intcrnediate and
(3.5 ran)
C?Dld
Soot
Blower
Modified Design by Gadelius K. K.
Figure 1. Comparison of conventional and modified air preheater designs.
(Numbers indicate clement spacing)
25

-------
Soot blowing procedures are also modified to include higher frequency and
higher steam pressure. The life of this type of air preheater element may
be about half of that when SCR is not applied. Operating costs are, there-
fore, likely to be higher since a larger element will be changed more
frequently.
INSTRUMENT VENDORS
The most popular method for continuous stack monitoring of NOx is
chemiluminescence which is employed at 75 percent of the locations using
SCR. There are several manufacturers of chemiluminescent monitors including
Shimadzu, Horiba and Yunaco. This method is also used for NH3 measurement
by addition of special NH3 converters; however, the validity of the method
is in question when SOx is present in addition to NH3 and NOx. In these
situations, NT!3 concentrations lower than actual are indicated by the instru-
ment. EPA pilot plant tests in the US have indicated that SO3 definitely
causes problems and that SO2 may also. These tests also indicated that only
about 70 percent of the NH3 is converted to NO.
Other techniques for NOx analysis are nondispersive infrared (NDIR) and
nondispersive ultraviolet (NDUV). These two can give erroneous readings when
used to measure NH3 in the presence of SO2; however, it was indicated that
they could be modified to prevent these problems.
Fuji Electric has developed an NDIR instrument and sample conditioning
system that reportedly does not have problems with SO2 interference. With
the Fuji Electric analyzer, the sample is passed through two converters to
convert NH3 and NO2 to NO. An additional converter can be added to reduce
SO3 to SO2 in order to avoid NVUHSOi, deposition in the sample line if neces-
sary. This sample is passed through a differential NDIR analyzer and compared
with an unconverted, reference stream to produce an NH3 signal.
26

-------
Another technique that is being investigated by Kimoto Electric and
others is second derivative spectroscophy. This technique has not yet been
applied to measure NHj under flue gas conditions.
The Central Research Institute for the Electric Power Industry (CRIEPI)
has examined all NHj and N0x analysis uethods and feels that direct-inview-
gas-anal>sis (DICA) coupled with NDIR «?nd NDUV has high potential for this
application. The US representative of tnis instrunentation is Environmental
Data Corporation, a subsidiary of Therao Electron Corporation.
Overall, instrument vendors in Japan have developed chemilunmescent
analyzers that can successfully monitor N0y and NH3 in flue gas streans with
low SO2 and particulate concentrations. Instruments for use with high con-
centrations arc being developed, but have not yet been applied and demon-
strated commercially.
27

-------
SECTION 4
INSTALLATIONS WITH \Os CONTROL SYSTEMS
SCR s>stens have been installed on nany gas- and oil-fired boilers and
are operating successfully. There is one application to a coal-fired boiler
and five others, scheduled to start-up in the near future as shown in Table 8.
Coal use for power production is expected to increase substantially over the
next ten years and it is likely that many of the new boilers will require
SCR to limit NOx emissions. In the US, there are many unresolved issues
concerning SCR technolopy as were listed in Section 1. The site visits were
designed co answer these questions.
NOx CONTROL EQUIPMENT
Per fomance
Most of the SCR systems in Japan operate at 80 percent NOx removal since
this is sufficient to meet the local emission standards. The operators report
that the systecs are very reliable and that the performance meets or exceeds
the guarantees. This is especially true for catalyst life guarantees. None
of the installations visited has required a catalyst change. Current guaran-
tees of catalyst life are two to three years for gas-fired applications, one
to two >ears for oil-fired applications, and one year for coal-fired applic-
ations. However, the actual life of che catalyst has exceeded the guarantees
at the oil- and gas-fired sources that have been operating for extended periods.
Catalyst
For oil- and coal-fired applications, the honeycomb, plate and pipe
catalyst shapes are all used. The catal>st formulation is typically Ti02
and V2O5 combined with small amounts of other materials to improve the
29 Preceding page blank

-------
TABLE 8. COAL-FIRED APPLICATIONS OF SCR IN JAPAN
Size SCR
Power Company Power Station Unit MW Vendor
Percent NOx
Removal	Start-up
EPDC
EPDC
Chugoku
Hokkaido
Joban
Job an
Takehara
Takehara
Shimonoseki
Tomakomai
Nakoso
Nakoso
ill
ill
01
111
if 8
<19
125
125
175
3f'0
600
600
B-H
kh:
MKI
B-H
MHI
IHI
80
80
50
>80
80
mid 1981
mid 1981
early 1980
late 1980
mid 1983
mid 1983
30

-------
overall performance and keep the SO; oxidation low. SO2 oxidation will
typically be about 0.5 to 2.5 percent of the inlet depending on the temper-
ature, catalyst form.ilation and SO2 concentration. For gas-fired applica-
tions, the pellet shaped c italyst is satisfactory and there are no problems
with SO3.
For the coal-fired applications, the catalyst and reactor are designed
in a vertical, downflow arrangement to minimize the potential fcr plugging
by flyash. Soot blowers are installed on somo units above tiie beds to control
ash buildup; however, these are installed as a conservative design measure in
case plugging develops and are not expected to be necessary.
There are up to four beds within the reactor consisting of one or two
layers of catalyst modules, each of which contain many catalyst elements. An
example is shown in Figure 2. In some cases a dummy layer of catalyst modules
is installed at the reactor inlet to prevent abrasion of the catalyst surface.
The homogeneous catalvst shape, while more prone to erosion, does not lose its
activity after the surface layer is lost and may be preferred to coated
-atalyst structures for use with coal-fired boilers. The catalyst manu-
facturers claim that the coated catalyst will hold up equally well and the
use of homogeneous catalyst appears to be a conservative design decision from
a performance standpoint in this new application.
Temperature Control
Temperature control is necessary on scne units during low load operation
when the flue gas temperature falls below 340°C with high SO2 concentrations
or 320°C with medium SO2 concentrations. On most units, however, the flue
gas temperature does not fluctuate widely enough to require control. Where
control is necessary, it can be accomplished by using an economi-eer bypass;
*
auxiliary heaters are not used. Where an economizer bypass is used, a mixing
zone may be necessary upstream of the reactor to allow mixing of the two gas
streams. Another technique for temperature control is load regulation. This
31

-------
Flue gas
inlet
honeycomb incsh
Unit catalyst
Catalyst
layer
Flue gas
outlet
Figure 2. Typical construction of catalyst beds
in reactor vessel (IHI).
32

-------
Involves establishing a minimum load for the boiler which corresponds to a
satisfactory clnimirn temperature, On one Industrial boiler, flue gas at the
desired tetrtDerature was obtained bv locating the reactor upstream of the
economizer entirely.
Labor Requirements
All of the inSTillanionn contacted Indicated that no operating labor is
required for the d-e~XOx system. The maintenance labor is also minimal, requir-
ing only two nen for a few hours every three to five days to unload from
delivery trucks. More significant labor requirements will occur when a
catalyst change is necessary which will be scheduled for the annual outage
if possible. KTiile none of the SCR system operators have experienced a
catalyst change, one operator estimated the labor requirement for a 175 MW
boiler to be about 300 man-days at seven hours per day, or about 20 people
for 15 days. Since this operation has not yet been performed, the labor esti-
mates vary widely.
Gas Analysis
At almost all of the installations NO* and NH-j are treasured by continuous
monitors using the chemiluroinescent technique. The Shimadzu MiA-302 was the
most conmonly observed instrument. Other brands are available and alternative
techniques are under development. Th« accuracy of the instrxxaents is checked
using wet methods. This type of unit will probably also be applied to the
coal-fired SCR systems, although its ability to accurately measure NH3 is
questionable. An improved NH3 instrument has been tested by Hitachi Zosen at
E?DC's Isrigo Fover Station; however, it h^s nc»t been applied commercially.
Other companies are developing Iffl3 analyzers for this application and some of
these are discussed in Section 3.
33

-------
Cost
The costs of the installed units vary widely with the most significant
factors being the type of fuel and whether the SCR system is new or a retro-
fit. Costs for most of the installations contacted are shown in Table 9.
These costs are for the SCR equipment only and the total costs for retrofit
installations may be significantly different.
IMPACTS OF NOx CONTROL EQUIPMENT ON AND BY OTHER FLUE CAS
TREATMENT EQUIPMENT
Particulate Control Fquipment
The NH3 emitted by SCR systems is not expected to adversely impact the
operation of a cold-side ESP since NH3 is often injected upstream of these
units on oil-fired boilers for ash conditioning. However, the NH3 can affect
the fly ash utilization by depositing on the surface. The impact of NHt.HSO^
on baghouse efficiency is not known in Japan and is under investigation. Bag-
houses do not appear to be applied commercially to much extent in Japan and,
hence, this impact is not of great concern to the Japanese.
The location of the ESP relative to the SCR reactor can significantly
impact the potential for plugging of the catalyst bed and air preheater,
especially on coal-fired units. Pilot plant data indicate that adverse effects
Tiay occur with a hot-side ESP which removes particulates upstream of the SCR
reactor. The ash that remains in the flue gas tends to be fine and sticky,
and can eventually build up and plug the catalyst. The opposite occurs when
a cold-side ESP is used because the full ash load has a sandblasting effect
which cleans the catalyst surface. This increases the catalyst erosion
although there are countermeasures for this effect. This sandblasting effect
also tends to keep the air preheater free of NHi,HSOi« deposits, although a mod-
ified air preheater or soot blower may also be necessary. The full ash load
allows some of the NHi,HSOi, to deposit on the ash particles rather than down-
stream equipment.
34

-------
TABLE 9. IKVESTMENT COSTS OF FULL-SCALE SCR SYSTEMS INSPECTED
Coapiny

OiuftMku Electric
Fuji Oil
Hokkaido Flcttric
EPDC
Suik'fprJ
Tooakoant
Takelisri
Fuel
t(>
n
n
i:
*/
u
ti
41
(MW>
1M,
rsr
on
on
Oil
Coal
rtl I
CO 4 Oil
Coal
CoaI
Coal
700
/on
700
375
700
'.0
10
no
250
?}0
Mie
CM
7nn
?no
700
m
7 00
u>
40
10
90
12*
125
rocfllvtlc do-'-Ox Svtt«a
Pero fit *.o"« Out I it NO*
3i%ovn1	fppoi)	Vendor
B0
ao
BO
8;
92
"~HO
30
30
H- tO
a- to
20
12
2 ft
:'„q
15
li
to
*0
eo
b-h
B-H
:he
IH1
*Wl
MM I
I( C
B-H
B-H
rof I r
Hi t rofit
Ritrgtit
\«»v
EUtruilt
V v
k-t tul U
«t«w
Ri t rof IC
R.*t rot IC
Gj»i
SlJirl »ip	(V/kW)
H»rrh 1978 |ft>0
A.rll 1976	1300
KjnJi )<>H0 U00
A[i r 11 19 iN 5900
Vpt.efccr 1979 1700
Apill |t»1 1/OOb
J m«i trv 1*)7S
jiily l'77n
October 19BO MO.OOO
June 1981 % /OOO
June 1981 * 7000
°»2i0 - 51
^Inf lwdlns* boiler modi He.u lorvi for cccnuni-tr bypdri

-------
Since a cold-side ESP is cheaper than a hot-side ESP, a cold-side ESP (or	per-
haps a baghoui>e) nay be preferable Co a hot-side ESP when SCR is used for NO^
control. The resistance to plugging of the air preheater or catalyst bed	has
not been proven on full-scale, coal-fircd equipment either in Japan or in	the
US for either the hot- or cold-side ESP.
In Japan, hot-side ESP's will be used with most of the SCR systems cur-
rently uncier construction on coal-fired boilers. There are several reasons
for this, one being that when these systems were designed several years ago,
the SCR catalysts had not been adequately proven with high ash loadings. Other
equally important reasons are that cold ESP's are not effective with the low
sulfur coal sometimes received in Japan and because there is concern that NH3
compounds in the collected fly ash will affect the by-product utilization.
One possible problem is that an ammonia odor occurs when ash containing utamonia
compounds is moistened. In situations where ash is disposed of by landfilling
into the inland sea, water pollution problems can occur. If the ash is used
in concrete manufacturing, the customer may not want to buy ash contaminated
with ammonia compounds. The problem may be eliminated by maintaining slip
ammonia at very low concentrations.
Air Preheaters
Several plants have reported problems with air preheater plugging by
KHt.HSOi, which is a product of a condensation reaction between NH3, SO3, and
H2O. In some cases, more rigorous soot blowing procedures have controlled
the pluggage to the extent that water washing the air preheater is only
required once per year during the scheduled outage. The plugging occurs at
the interface between the intermediate and cold temperature elements where the
temperature is low and soot blowing is least effective.
To prevent air preheater plugging, most plants will modify standard soot
blowing procedures to include blowers at both the cold- and hot-sides, more
frequent blowing and higher pressure steam or air. Some installations will
36

-------
use a modified air preheater design which combines the intermediate and cola
temperature sections into one continuous element.
It should be noted that these air preheater plugging problems are occur-
ring on units with low ash loadings; no units with high ash loadings have been
operated. The effects of NH^HSOi, on air preheaters in full-scale, coal-fired
boilers are not known. Data from full-scale facilities in Japan should be
available by 1982. As mentioned earlier, the high ash load of a coal-fired
boiler can have a beneficial cleaning effect on the air preheater.
SO? Control Equipment
Systems in wh*ch a limestone FGD unit exists downstream of an SCR unit
will be operational in the near future in Japan. There are no data available;
however, potential problems havr been considered.
There are no adverse effects of SCR systems on the performance on down-
stream limestone/gypsum FGD systems. The FCD system will absorb NHj; however,
reliability and SO2 removal are not affected. Some potential problems have
been identified including fuming and NH3 contamination. An ammonium bisulfite
(MUHSO3) plume can occur which is a function of ambient temperatures and
NH3, SO2 and H2O concentrations. Such a plune is not expected, however,
since no plume occurs at boilers where NH3 is used for ash conditioning. In
these situations the NH3 concentration at the scrubber inlet is about 10 ppm,
which is very analogous to what would occur if an SCR system was located
upstream. It is speculated by some that a plume might occur if NH3 emissions
>50 ppm occur.
SCR systems can impact wastewater treatment requirements .since they can
add nitrogen compounds to the wastewater by both air prteheater washwater and
t
FGD system blowdown. Where this can cause problems such as eutrophication,
an activated sludge or ammonia stripping process is used to remove the nitrogen
37

-------
compounds. Where this is not a problem, the air preheater washwater can be
blended with Che repular wastewater. There are also potential solid waste
impacts.
Typically in Japan, the gypsum generated by FGD systems is washed If it
is to be used commercially. Ammonium salts in the gypsum may cause waste-
water problems and require treatment such as activated sludge. In Japan,
this problem is location specific since the water regulations for ocean areas
are different than those for inland sea areas. In situations where the gypsum
is used as landfill material, no ammonia leaching problems are anticipated.
This is based on the fact that in Japan there are three plants which use an
N"H3/CaO dual alkali FGD system and experience no problems with 'inmonia in the
gypsum by-product.
38

-------
SECTION 5
GOVERNMENTAL AGENCIES AND INDUSTRY GROUPS
GOVERNMENTAL AGENCIES
Environmental regulation in Japan is similar to the US in that a federal
/
agency establishes the minimum requirements for the country as a whole. Local
governments have the option of adopting the federal standards or establishing
their own, more stringent standards. The local governments utilize sophisti-
cated air quality monitoring systems to assure that the public is protected
and that the regu1ations are complied with.
Federal Regulations
The Japan Environment Agency has recently modified the standards for
ambient air quality and established thr»ie concentration ranges based on a
daily average sample. Above 0.06 ppm NO2, countermeasures must be undertaken
to reduce the level to 0.06. Within the range 0.04 to 0.06 ppm NO2, no sig-
nificant change is allowed; areas below 0.04 ppm NO2 are allowed to deterio-
rate .
The Environmental Agency also sets emission standards foe point sources.
The standards, shown in Table 10, are such that they can be met through use
of combustion modifications alone.
Local Regulations
It is the local governments, city and prefectural, that have required
many of the SCR applications. This occurs primarily when the plant owners
negotiate with the local citizens over a plant siting or expansion. A
39

-------
TABLE 10. EMISSION STANDARDS FOR NOy I-N JAPAN AS OF AUGUST 2, 1979
(Source: Japan Environmental Summary, Sept., 1979)
Type of
facility


Standard
value (ppo)


Unit:
1,000 NsJ/h
\Dat« of
\lnstal1-
\In8
OjX N,
be fore
Aug 9. *73
after
Aug. 10, '73
before
Sept 9, ' 75
after
Sept. 10, '75
before
June 17, '77
after
June 18, '77
before
Aug 9, '79
after
Aug. 10, 179

Gas
500 ft aho.c


,00
60

f1ring
- 5'JO



100


o - ioo


130

100


LG - 40

150
130



5-10
less t**an 5

150

Solid
taate-
r 1 a L
firing
100 & above

til Apr 30,*B0
600
f rom May 1,'80 .
CoO
460





40 - 100
6
til Aug. 9,*B2
600
from Aug.10,'92
4 00


400


10 - 40

til Aug 9,'62 froa Aug.10,*82
630 4 80
480


C

5 - .0

fron H
,iv 1, ' P0 460





less : " a n 5

'-en Aug 10.'84 480



Liquid
firing
500 & above

til Apr. 30,*e0
2 30
from May 1,'80:
160

150
130


100 - 500
4
til Apr 30.'80
230
f rom May 1,183
190
ISO
150



i0 - 100

190






13 - 40

f ro=i May 1,
•80 2 30





5-10

(- Sept 9/77)
froa Oct. 1
,'80 250

(Sept. 10,'77 -)


less than 5

<- Sept 9,'77)
froa Aug. 10,*84 250
180
40

-------
cooperative spirit which exists between the local citizens and the plant
provides the driving force for the plant's compliance with the negotiated
emission limits.
Ambient Air Quality Monitor?ng and Man.-igement
Each prefectural government and some city governments maintain a system
of ambient air and stack monitors which measure pollutant concentrations at
strategic points throughout the prefecture. A typical example is shown in
Figure 3. These data are relayed to a central data processing center where
It is compiled and reduced. If an air pollution problem occurs, such as
photochemical smog, the center notifies key industries to cut back emission."
and traffic is controlled in the prcblem area. The center also has the
responsibility of notifying the local citizens and advising them. There is
good cooperation between the plants and the center, and the system seems to
work well.
INDUSTRY GROUPS
The two industry groups contacted, the Central Research Institute of the
Electric Power Industry (CRIEPI) and the Federation of Electric Power Companies
(FEPC), are both looking to the future when coal will be a much more signifi-
cant energy source. Coal-fired boiler capacity is expected to increase from
4410 MW in 1978 to about 34,500 MU by 1995 and NOx control will probably be
required on all of these. They feel that combustion modifications and SCR
have been successfully demonstrated on gas- and oil-fired boilers and that
these techniques will also be useful on the coal-fired units.
SCR still has some problems to overcome in order to be used successfully
with coal. The most significant of rhese at the present time are dealing with
flyash or the lack of it, solving the air preheater plugging problems and
developing a continuous NH3 analyzer. Programs to meet these objectives are
already underway. EPDC is installing full-scale SCR units at Takehara while
continuing pilot unit testing at Isogo. The full-scale units will utilize the
new design air preheater built specifically to avoid plugging and corrosion.
41

-------
Figure 3. Kanagawa prefectural government
section air pollution monitoring center.
42

-------
APPENDIX A
PROCESS AND EQUIPMENT VENDORS
A3

-------
BABCOCK-HITACHI K.K.
Kure Works
Kure, Hiroshima Prefect:'.i,_
March 13, 1980
Hiroshi Kuroda, Department Manager
T. Hiruta, Senior Engineer
Takashi Akivama, Senior Engineer
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank

-------
BAB COCK-H J TA CHI
Babcock-Hltachi is a manufacturer of boilers and pollution control
equipment as well as several other types of industrial equipment. They
have developed both pellet and plate type catalysts for N0X control and have
supplied several SCR s>stems as shown in Table 11. Two of the units will be
installed on coal-fired boilers; one at Hokkaido (Plant A-6) and another at
Takehara (Plant A-12). Of the utility installations listed, plants A-l, 2,
3, 4, 5, 7, 8, and 9 are currently in operation.
Babcock-Hitachi has conducted several pilot unit tests in the develop-
ment of their process including coal-fired testing at EPDC's Takehara power
station. At this pilot unit flue gas was taken from upstream of the air
preheater and ESP, passed through a small scale hot-side ESP, the SCR
reactor and a small scale air preheater. Some data fiom this unit are shown
in Table 12. Tests with low ash loadings were conducted for 10,000 hours.
A bypass around the ESP was then installed to perform tests with high ash
lead ings.
The catalyst used for high dust applications is a plate type of
catalyst with a configuration similar to that of air preheater elements, as
shown in Figure 4. The plates are metal with catalyst coated onto the
surface. The individual plates are held together in a metal basket
(0.5 meter cube) and several of these baskets are combined to make a module.
The modules are stacked onto structural steel supports within the reactor
by using a special dolly with jacks on it. The dolly travels on additional
struciural steel within the reactor. Usually, three separate stages are used
in the reactor. The catalyst and the loading procedure are also depicted
in the Figure.
Preceding page blank

-------
TABLE 11. SCR INSTALLATIONS J?Y BABCOCK-HITACHI AS OF MARCH 1980
//
Plant
Cas
flow rate
(Nw3/h)
Fuel
Percent
NOy
Removal
Operating
date
Remarks
1
A-l
300,000
Crude oil
80
Jun., 1977
25%
of 450 MW
2
A-2
2,000,000
LN'G
30
Mar., 1978
700
MW
3
A-3
2,000,000
LNC
80
Mar., 1978
700
MW
4
A-4
483,000
Hea\rv oil
50
Jul., 1978
175
MW
5
A-5
466,000
Heavy oil
30
Jul., 1978
156
MW
6
A-6
280,000
Coal
90
Sept.t 1980
25%
of 350 MW
7
A-7
980,000
Kerosene
80
Mar., 1981
90 MW (Stag)
8
A-8
466,000
Heavy oil
30
Jan., 1980
156
MW
9
A-9
450,000
Heavy oil
30
Feb., 1980
156
MW
10
A-10
470,000
Heavy oil
30
May, 1980
156
MW
11
A-11
740,000
Heavy oil
80
Jul., 1980
250
MW
12
A-12
399,500
Coal
80
Jul., 1981
50%
of 250 MW
13
A-13
970,000
Heavy oil
80
Jan., 1981
375
MW
14
A-14
1,330,000
Heavy oil
80
Jun., 1983
500
MW
15
A-15
1,565,000
Heavy oil
80
May, 1981
60 0
MW
16
A-16
1,645,000
Heavy oil
80
Jul., 1981
600
MW
17
A-l 7
470,000
Heavy oil
30
Aug., 1981
156
MW
18
B-l
15,000
Heavy oil
70
Sept., 1976


19
B-2
20.000
Heavy oil
90
Sept., 1977


20
B-3
16,000
Kerosene
90
Oct., 1977


21
B-4
20,000
Heavy oil
90
Apr., 1978


22
B-5
30,000
Kerosene
90
Sept., 19 78


23
B-6
50,000
Tar Sand Oil
80
Jan., 1980


1 J
C-l
500,000
C0C/BFG -
95
Oct., 1976


25
C-2
3,600
Electric
Furnace
84
Jan., 1977


26
C-3
10,000
Acid Pickling
Gas
90
Dec., 1979


Note 1: A-utility boiler; B-industrial boiler, C-others
Note 2: In A-4, 5, 8, 9, 10, and 17 plate type catalysts are installed in
the restricted space of the flue gas duct between economizer and
the air preheater.
48

-------
TA5LE 12.
NO< Removal
N"H3:N0X Mole Ratio
Gas Temperature
Raactoi £P
Flue Gas Flow Rate
SO2 Conversion to SO3
Inlet S02
Test Period
DATA FROM BABCOCK-HITACHI PILOT PLANT
AT EPDC TAKEHARA POWER STATION
80%
0.83
350°C
20 mm H2O
1250 Nm3/hr
<0.2%
1500 ppm
6000 hours
49

-------
Gas
Catalyst Element
Crane
Catalyst Unit
Catalyst Block

Transportation
Loading
Figure Assembling and Loading Procedure of Catalyst

-------
The cacalyst and reactor are capable of handling particulate concentra-
tions of 15-20 g/Nm3 and achieving 82 percent N0x removal with a pressure drop
of 12 to 13 mm H2O. Abrasion testing has been performed in laboratory equip-
ment where the flue gas flow rate varied from 5 to 25 m/s and the dust load
was also varied. From these accelerated erosion tests, it is calculated that,
after 20,000 hours of normal conditions, the catalyst weight loss would be
<1.0 percent. This loss did not affect the performance of the catalyst.
Reportedly, the cacalyst performs equally well with either a hot- or cold-side
ESP and, therefore, must not require the abrasive effect of fly ash for
cleaning.
The recommenced operating temperature is 300 to 420°C with the SO3
content determining the actual low temperature limit. A few minutes of high
temperature will not harm the catalyst; however, the implication is that
extended high temperature periods can deactivate it. On some Industrial
boilers, a gas heating system is used to prevent low temperature excursions.
This arrangement is used at plants B-2, 3, and 5. No heating or economizer
modifications are necessary on the utility installations. Babcock-Hitachi
also lias a low temperature catalyst that can be used under low temperature,
low sulfur conditions.
NH3 emissions are reportedly <10 ppm and sometimes <5 ppm; however,
these are not measured directly as NH3. Babcock-Hitachi determines NH3
emission by a calculation assuming a 1:1 reaction between NH3 and N0X and
measuring the N0X concentrations. NHa emissions have caused no adverse
effects on the performance of downstream FGD equipment.
Problems can occur with NHuHSOi. pluggage in the intermediate section
of conventional air preheaters. A low SO2 conversion catalyst has been
developed to minimize these problems. To eliminate this problem, Babcock-
Hitachi recommend a modified air preheater design in which the intermediate
ar.d cold sections are manufactured as one continuous piece. Hot-side soot
blowing is recommended in addition to cold-side soot blowing.
51

-------
The costs for a 600 MW SCR facility with 90 percent N0X control are
estimated to be 50% greater than those for 80 percent N0x removal. Catalyst
costs are the primary difference between the 80 and 90 percent removal cases
since catalyst volume requirements arc significantly higher as shown in
Table 13.
TABLE 13. CATALYST REQUIREMENTS AS A FUNCTION OF NO REMOVAL
x
Percent N0x removal efficiency
Relative catalyst volume
80
100
90
130*
aThe specific amount will depend on the NK3:N0y mole ratio and the NH3
emission specification.
The guauntee ranges from one to two years depending on the
particulate loading of the gas. However, liabcock-Hitachi is planning to
make a two-year guarantee in an offer for one high particulate level
application.
52

-------
CATALYST AND CHEMICALS INDUSTRIES CO.
Headquarters
Tokyo, Japan
March 24, 1980
Royozo Aoi
Yoichi Nishimura
Hisao Nishikawa
David Mobley
53

-------
CATALYSTS AND CHEMICALS INDUSTRIES COMPANY
Catalyses and Chemicals Industries Company (CCIC) has supplied 1500m3
of catalyst to SCR process vendors for use in electric power plants. Their
primary product in this area is homogeneous, honeycomb catalyst elements
which are preferred to coated types because they maintain activity even if
eroded. This area is one of the current areas of SCR catalyst R&D
which include:
•	resistance to erosion
•	improving de-NO^ activity
•	suppressing SO? oxidation activity
•	minimizing NH3 emissions
Particulates from coal-fired boilers can erode the inlet face of the
honeycomb and, although N0X removal is apparently not seriously affected,
the erosion can cause structural problems over a long period of time. They
are also developing a new manufacturing method for the Ti02 base material
which will provide erosion resistance with coating.
It is desired to keep the SO3 concentration in the outlet gas low in
order to avoj.d problems with NiUHSOu formation downstream; however, the
V2OS catalyst tends to oxidize SO2. In applications where the SO2 concen-
tration is low (<100 ppm) SO2 oxidations of 5 percent are allowable. For
medium and high S0Z concentrations, SO2 oxidation is suppressed to <0.5
percent by the use of complex compounds of vanadium oxide with metals such
as tungsten and molybdenum. The activity of the catalyst can be reduced by
these compounds in proportion to their concentration.
CCIC can make a catalyst in a tubular or pipe shape, but prefers
the honeycomb for several reasons. The tubular catalyst, they claim, is
difficult to assemble, has a low surface area and requires a large volume
Preceding page blank
55

-------
for high NOx removal. For example, a 600 MW plant will require 300 Co
400 m3 of honeycoab catalyst, but 3 to 5 times that amount if a tubular
catalyst is used. The honeycomb catalyst is illustrated in Figure 5,
The catalyst dimensions depend on the flue gas composition. For low
and medium sulfur oil, they supply a catalyst with a 5.7 to 6.6 tnm channel
diameter and a 1.3 to 1.4 ram wall thickness. Catalysts with larger channel
diameters and wall thicknesses are necessary for coal. Either can be manu-
factured in lengths of up to one meter. Some rules of thumb for coal vs oil
fired applications are that for coal, the space velocity will be 65 percent
of that for oil, the catalyst volume will be 50 to 60 percent greater and
the catalyst weight will be 20 percent greater.
High temperatures for short periods will not harm the catalyst;
however, exposure to high humidity can be detrimental. The 10 to 20 percent
H20 typically found ir. flue gas is not harmful at 350°C. Problems can occur
if the temperature drops to <100°C where condensation takes place. Two
things can happen. Condensed water ir> the pores is difficult to evaporate
due to the capillary pressure resistance to evaporation. Secondly, water
can make the metal crystals aggregate. Both will reduce the catalyst activity
and the latter effect will cause permanent damage.
The catalyst guarantee will depend on the specific flue gas conditions
of the application. One year after start-up is typical; however, for low
sulfur applications, two years may be offered. The company will take re-
sponsibility for disposing of spent catalyst although so far they have not
received any. They plan to recover the useful metal and discard the rest.
Vanadium, tungsten, and molybdenum may pose waste disposal problems and
require treatment. For US applications, it may be economically prohibitive
to return the catalyst to Japan for disposal when compared to the cost of
local disposal.
56

-------
-	•» :-V.' -..'r
Figure 5. Honjy.ib catalysts manufactured by Catalyst and Chemicals.
L">w and medium sulfur oil application.
Channel diameter = 5.7 to 6.6 mm.
Wall thickness = 1.3 to 1.4 mm.
57

-------
Soot blow ng the catalyst may
applications. On:e per day should
for coal. They have no preference
be necessary on medium and high sulfur
be sufficient for oil; more frequently
for air or steam.
58

-------
F.BARA CORPORATION
Headquarters
Tokyo
March 25, 1980
Koichi Yamada, Manager, Plant Engineering Dept.
Tohru Agarida, Chief, Plant Business Development Sales
Keita Kawamura, System Engineering Dept.
Shinji Aoki, Svstem Engineering Dept.
Charles 01 son, Consultant
David Mobley
Cary Jones
Doug Maxwell
59

-------
EBARA CORPORATION
The EBARA Corporation has been developing tneir process for simul-
taneous removal of N0X and S0X fo- several years. The process utilizes
electron beam radiation and NH3 in produce a react ion between both S0X and
N0x. One potential arrangement of the equipment is shown in Figure 6. One
new aspect that makes the current design different tnan previous designs is
the addition of NH3 to the flue gas stream. With NK-, present, the reaction
products are (KHOzSO^ and NHi,NO in solid form, vh&, without NH3, a
chemically uncharacter iz<-.d powder is formed. NHj is injected in the
following mole ratios: 1 to 1 for N0X and 2 to 1 foi S0X.
Although the equipment is located downstream ol the air preheater, it
is necessary to cool the 150°C gas further to 70°C v ith additional heat ex-
change equipment. The reactions that take place ar
-------
Air Electrostatic
Boiler heater precipitator
Stack
V
Electron
acceleroror
Reactor room _i
Water
Electron
accelerator
—Q
Flue gas
induction fan
¦HReactor
Reactor room

- - J
Heat exchanger
(Indirect type) "
Reactor
Figuie 6. Flow sheet for EBARA electron beam process.

-------
Reactor
Combustion e~
flue gas |||
Electrostatic
precipitator
Stack
,SOx+ NOx
/ NH3
Particles
f(NH<02SO4	I
\(NH4)2S04-2NH4N03j
ON
U)
0H.0.H02
x H2SO<, HNOs
SO? + NOx
10"5occ
h2so«-!^:!J-(nh<))2S04
H2S04
NOx^-~r-HN03
(NHa)2S0<» ^.f' -*•(NJH4)2S04'2NH4N03
HNOa
N2,02,H20
e~
0H,0, HO2
Figure 7. Diagram of reaction mechanism for EBARA process.

-------
Wall of Reactor
(Stainle-s Steel Ductwork)
Electron
Accelerator
Flue Gas + NH3
1 v
-Titanium foil window
Figure 8. Arrangement of electron accelerator and reactor.
64

-------
The process has been tested on pilot scale equipment using gas from
a sintering furnace at a steel plant. The electron distribution in the re-
actor was not uniform, and an impeller was used to rotate the gas to achieve
a more uniform integrated dose for the gas flow. As shown in Figure 9, the
removal of S0X and N0x was more efficient when the ratio between the im-
peller rotation speed and the gas flow rate was ahojt 1/60 rpm per Nm3/h.
The pilot unit treated 3,000 - 10,000 Nm3/h of gas for one year and
during the test period a 30-day continuous run was performed. The data from
this run are presented in Figure 10. Removal of N0X remained at about 80
or 90 percent depending on the N0y analysis method used. Two NOx analysis
methods, PDS and NEDA, are shown sjace EBARA has done tests which indicate
that the PDS method produces"low measurements of N0X concentration when gas
is irradiated and contains NH3. EBARA used a modified NEDA method in which
NaOH was added to the sample flask and O2, rather than Os, was used as the
oxidant.
The removal capabilities of the EBARA process are summarized in
Figure 11 where removal is shown as a function of electron dosage. A
dosage of 5 wh/kg is typical.
Tests with a simulated coal combustion gas are shown in Table 14.
Compared with the results shown in Figure 10 and Figure 11, S0X and N0X
removals in Table 14 are slightly lower. However, by using results such as
those shown in Figure 9, a simple interpolation suggests that simultaneous
S0x and N0X removal more than 90 percent could be obtained by operating the
equipment at the optimum rotation ratio of 1/60 rpm per Nr>3/h.
The ash present in coal-fired flue gas will absorb some energy from
the accelerator, but this los* is low. For example, with 30 g/Nm3 of fly
ash, about 3 percent of the energy will be lost to absorption by fly ash.
65

-------
100
SOx
NOx
•o
c
<3
60
x
o
z
U-J
40
o
8 9 (x 1/60)
6
7
4 5
3
2
0
1
/rpm*h
Gas Rotation Ratio t ^3
Figure 9. Removal of SOx and N0X vs. gas rotation ratio.
66

-------
April
May
27
28
©
®
©
2

4
©
6
©
8
9
©
It
12
13
©
15
16
17
18
19
20
©
22
23
2C
25
'Nj
-p
pi
go
orin
(/)
c
So
"sS-fe
ZU ° -
I
c o .
Q) — ~ h
OcOgS
° 9s° &
OT g~«H
Q~
100
90
80
70 -
^	S0x 	 		*" >• ^ ,.-» ,—v ^	
W3pcPcA/\^	c>^ooQX)cocp^bcb
NOx(Simplified NEDA method)	"/> ~
220
200.
180
160
1 ~ /	N *
]0) CLl
Ouj a
xl
^ o
V-
UJ o
70
¦ 50-
N0x(PDS Conversion volue) ^
A
# a*a
t h ft	-a
vw1 V* ^
A* SOx .	4 i
w -vcs.^ .
Sintering	• Sintering
machine stopped	machine stopped
"• _®°
,0-VV"*0*,
V
«- io
E 8
\ 4
c £ "No adjustment
Ohof flowmeter Adjustment
for NH3 addition j

E JVT /V*
- 30-ii*r*
~ 10LL___I	I	
Treated gas volume 3000 Nm3/h
Absorbed energy 4 2 Wh/!
-------
100
•a
c
CO
o
in

CD-
(PDS Conversion Value)
(Simplified NEDA method)
NO : ca. 180 ppm
SO^: ca. 200 ppm
Treated gas volume: 9000 NmVh
Reaction Temperature: 70°C
2	3	4
Flue gas absorbed energy (Wh/kg)
Figure 11. N0x and S0X removal levels achievable
by the EBARA electron beam process.
68

-------
TABLE 14. PERFORMANCE OF EBARA PROCESS WITH HIGH SOx/NO
CONCENTRATIONS USING A SIMULATED COAL-FIRED X
FLUE GAS
Conditions
Gas Flow = 1500 Nm3/h (900 SCFM)
Temperature = 75°C
S0X = 1900 ppm
N0x = 620 ppm
Energy Absorbed = 1.8 mrad (5.0 watt-h/kg)
Results
SO Removal = 80%
N0x Removal = 82%
Exhaust NH3 = 50 ppm
69

-------
The costs of the EBARA process have been estimated for a system
treating 1.5 to 3.0 x 106 Nm3/h of flue gas from an oil- or coal-fired power
plant. The gas composition used in the cost calculation is 200 ppm N0X,
1500 ppm S0X and 0.1 to 0.2 g/Nm3 of particulates. Removal efficiencies are
90 percent for SO2, 80 percent for N0x and 99.8 percent for particulate
products. For this situation, the capital costs are ¥25,000/kw ($100/kw)
and operating costs are ¥l.85/kw (7.4 mill/kwh). This calculation does not
include any by-product credit for the ammonium sulfate and ammonium nitrate
solids.
70

-------
FUJI ELECTRIC CO.
Tokyo Works
Tokyo
March 24, 1980
S. Ueda, Assistant Manager of R & D Department
Atsuo Watanabe, Manager of Analyzer Development Group
Eiji Tsuji, Manager of Administrative Department
Masahiro Murasee, Chief of Administrative Department
Takashi Kimoto, (Kimoto Electric Co.)
David Mob ley
Gary Jones
Doug Maxvell
71

-------
FUJI ELECTRIC CO.
Fuji Electric manufactures a variety of instruments, one of which is an
ammonia analyzer. The overall system utilizes a series of converters fol-
lowed by a nondispersive Infared (NDIR) analyzer. The output is compared
to a reference signal and NH3 is calculated by difference. The basic arrange-
ment is shown in Figure 12. The sample first passes through a filter on the
probe tip to re"ove particulates. The sample probe material of construction
is very important. For example, with a stainless steel probe, the. NH3
will react with NO and NO2 in a typical d^-NOx reaction as shown in Figjre 13.
Alternatively, a titanium probe will r^ot affect the NH3 concentration.
In the first converter, NH3 is oxidized to NO at 700°C. The converter
also oxidizes some NO to NO2 via a side reaction. The overall reaction in
the first converter is
This converter has also an undesirable side reaction in which several percent
of the SO2 is oy.idized to S03. The sample stream is then passed through a
second converter also at 700°C to convert NO2 to NO. This converter con-
sists of specially treated activated carbon which reacts with NO2 according
to the following reaction
At low NO^ concentrations, activated carbon is claimed to be superior to
Hn02, which is used as an NO2 converter in some chemiluminescent
instruments.
Following the conversion steps, the sample passes through the pump
and a permapure dryer in order to reduce H2O interference to below 0.5 ppm
NH3
*¦ 0.97 NO + 0.03 NOz + 1.5 H20
N02 + C 	to-
HO + CO
73
Preceding page blank

-------
Infrared
NO Annlyjcr
NHj/NO
' Converter
Probe with
V 11tcr
NO>/NO
Converter
Figure 12, Schematic of Fuji Electric NH3 Analyzer.

-------
ppm NHj
40
30 -
• A——__aX-		
^			-A—
	i
\
\
A titani on
© Stainless Steel
u
n	1	1	7-
2no	400	°C
Temperature of Probe
Figure 13. Influence of probe material on NH3 concentration observed.
75

-------
NO equivalent. The sample flow diagram is shown in Figure 1A. A separate
"reference" stream is taken at the same flow rate and dried, but not passed
through any converters. Each stream is then passed tnrough the NDIR differ-
ential typo N'O analyzer shown in Figure 15. The NHj concentration is pro-
portional Lo the difference in light absorption.
The instruments performance specifications are shown in Table 15.
TABLE 15. PERFORMANCE SPECIFICATIONS OF FUJI ELECTRIC NH3 ANALYZER
Item	Specification
Sample Gas Temperature Range
Sample Gas Pressure Range
Sample Gas Composition
Noise
Minimum Detectable Concentration
Drift
Interference
15% C02
SOi
CO
HC
200 to 400° C
-150 to 350 mm HzO
> 1% 02
<	1000 ppm CO
v 500 ppm HC
0.25 ppm
0.5 ppm
+ 1 ppm/week
<	0.5 ppm
Negligible
76

-------
0 S<7otn
Pressure
r mire
III < n
Kn« rr
liU,
V.Oi/'JO	Hl*t j- «A
fonvrr H* r CotV
P^fFTUlpltr''
|»rv*»r
H S»/rsln
a
Nllt/S'l O'tivpffr
FILt»r
Si "ril
vol V«"
C i */bin
I) 11 f * > >»11 •)
tW|>,
So Ana I\tfr
-u

rr^Tq
I	ct)	 ¥
1
0 3f/oln
Figure \U. Sample flow diagram of Fuji Electric NH3 Analyzer*

-------
Inter'erence
compensating
detector
Sanp'e &3S •
outlet
Sample cell II
Sonple gas —

-------
lour similar instruments are currently applied on oil-fired boilers of
the Tokyo Electric Company at Yokosuka, but have not been tested on flue gas
from coal or high SO^ concentrations (>200 ppm). Also, the instrument appears
to haes are cooled. The company representatives indicated that this
problem could be averted through the use of a third converter which reduces
SO3 to S<'2 ar.d that they have already developed such a converter.
Fuji Electric is intfrestad in marketing the instrument in the US;
however, servicing will be a problem unless it could be provided by their
US reprctentative, Fuji Electric Corporation of America, or another US firm
which would contract with Fuji Electric to provide servicing. The costs for
the insLium<'nt, with no service guarantee is $18K including the analyzer and
$13K without it.
79

-------
GADELIUS K.K.
Headquarters
Kobe
March 11, 1980
Fumichika Yoshikawa, Deputy Department Manager
Eiji Okamoto, Group Manager
Susumu Sakashita, Manager of Proposal Section
Y. rukaya
Nils Hornmark, General Manager
David Mobley
Gary Jones
Doug Maxwell
si Preceding page blank

-------
GADELIUS K.K.
The air preheater engineering group of Cadelius is involved in a
joint development program with EPDC to develop a specialized air preheater
design for application downstream of SCR processes. Tt has been well es-
tablished that air prehcaters located after an SCR system are likely to
experience plugging problems due to NHitHSOu deposits. According to Gadelius,
these deposits form in the middle, or intermediate temperature, portion of
the air preheater where the temperature is <250°C and plug the interface
between the intermediate and low temperature sections. They are difficult
to remove by soot blowing since this technique is effective in cleaning
only the elements adjacent to the soot blower. The NHi,HSOi, is corrosive
and reacts with the air preheater elements to form NHi»FE(S0O2 anc*
NH^A1(S0^)2-
The company's program with EPDC began in 1976 with a small (10,000 Nm3/hr),
conventional Ljungstrom air preheater at EPDC's Takasago power station. Tests
were conducted in which NH3 was injected into flue gas with a full particulate
load (12-20g/Ntn3) . NH3 was maintained at 50 ppm while SO3 varied between
1 and 20 ppm. After 4000 hours of operation, the AP increased 40 percent over
normal. Gadelius felt that one water washing per year would be sufficient
to control this amount of plugging. Deposits were less than expected due to
the abrasive effect of the ash.
Subsequently, testing began with a low dust gas (100 mg/Nra3) from a
hot-side ESF and NH3 and SO3 concentrations maintained at 10 and 20 ppm,
respectively. With the conventional air preheater, a four element design
designated DU/DL'/DU/N'F6, severe plugging occurred aft..r 400 hours. A modi-
fied two element design, desigrated DU/NF3.5, was then substituted. In this
design, a large (900 mm) hot section is used in conjunction with a "large
(1150 ran) cold section which is enameled. The modificatiotj is shown
schematically m Figure 16.
Preceding page blank

-------
Intermediate
Cold
Soot
Blower
foot —i
R lower
Conventional Arrangement of Meat Transfer Elements
Combined
Intermediate and Cold
Soot
Blower
Modified Design by EPDC and Gadelius K.K.
Figure 16. Comparison of conventional and modified
air preheater designs.
A closer plate spacing was used on the enameled elements in order to maintain
the same neat transfer as before. Also, the soot blowing technique was
modified so that both sides of the air preheater were blown. Following these
modifications the unit has been operating successfully for 1500 hours.
Specific data are shown in Figure 17.
Flue Gas 330°C
Air 280°C
150°C
30 °C
i»AP=145mm l^Oj1
Figure 17. Operation of air preheater modififed for use
with low dust load.
84

-------
There are soize other design modifications to this new air preheater.
Corten steel, a corrosion resistant material, is used on the cold side. The
frequency and pressure of soot blowing is increased to 1.5 tons/hour, six times
per day on the cold side and one ton/hour, twice per day on the hot side. On
a high dust application, Gadelius expects that soot blowing can be reduced to
once per day.
With respect to water washing, Gadelius recommended that higher pressure
water (7 to 10 kg/cm3) be used and that once per year during a scheduled boiler
outage would be sufficient. For a 500 MW unit with two air preh'aaters, a water
usage of 2000 tons was estimated.
The capital costs o£ the modified air preheater are estimated at 1.5
times nornal. Installation will also cost about 10 percent more. Operating
costs for a 'ow fly ash system will be about three times normal, due
primarily to the increased steam usage. The life expectancy of t.he cold end
for a low fly ash system may be reduced by about one-half of that for the
situation without SCR, thereby increasing maintenance costs. Belter results
are expected for high fly ash applications.
A full-scale preheater has been ordered by EPDC for installation at
their Takehara pover station and should be operational in 1981.
85

-------
HITACHI ZOSEN
Tokyo Office
Tokyo
March 19, 1980
Masa> Kinoshita, President
Masuichi Yasuda, Managing Director
HIdeya Inaba, General Manager
Sliingo Tanaka
Mat»ayoshx Ichiki
Mt .ao Takeda
Joji Saruwutari
David Mobley
Cary Jones
Doug Maxwell
Preceding page blank

-------
HITACHI ZOSEN
Hitac'1.1 Zosen, in conjunction with their 1j censee Cliemico Air Pollution
Control Corporation, is operating a pilot scale (0.5 >IW) SCR unit in the US
and treating gas from a coal-fired boiler. The project is sponsored by the
EPA and is located at Georgia Power Company's Plant Mitchell in Albany,
Georgia. The discussions with Hitachi Zosen focused on tests being conducted
at the power plant.
The pilot plant utilizes Hitachi Zosen's NOXNON 500 catalyst which
consists of catalyst coated on a metallic substrate arranged in triangular
honeycomb configuration. In March, the pilot plant experienced a decrease
in catalys activity which appears to be due to a combination of surface
blinding bj a fly ash and catalyst delamination by the soot bloving steam.
In order to avojd the dust plugging problem, a new catalyst will be in-
stalled. This will be the NCXNON 600 catal>st which uses stainless steel
gauze as the substrate rather than a metallic plate. The pitch of the new
catalyst will be larger, 14mra and this is expected <.o eliminate the plugging
problem.
Soot blowing of the catalyst occurs once per shift with steam and
inspection of the catalyst element, showed that the active material had been
removed from the metal at the inlet face. It is suspected that the soot
blowing ste.un has some condensate present which may promote abrasion at the
inlet face. Future soot blowing will be done with air to avoid this
problem.
Prior to this problem, ihe pilot unit experienced smooth operation
during the test series. A set of operating data for two weeks in February
is shown in Figure 18. As can be seen, NOx removals are generally >90
percent and appear to be affected somewhat by flue gas temperature which
varied between 327 to 382°C (620 to 720°F>. Unreacted NH3 in the outlet
gas is consistently below 4 ppm.
so Preceding page blank

-------

O
O

•i1.fl
';' | 1 '? -
n,lJl
M
i:i i-i H- f r-i! M i. i ;
ill ';;l. I t'| T1 -! 1 I
n«*5t ».
jplJ3^
90
*>
to
ji'
;. .1:
,i : i
, m w
...... N.J> | |i U -¦
Cfc«ned{jiJi»flrj;.»«A7\ i . j i ; :
•¦Nt i*#pjrj(	It! .' ' 1 '
" 'I'M L; L: it!'
! i • • J » ! • I I ' I
[ f.rri " 1 * ; ;
hlT.r r-|Tr.riYj-
I .Li.' I ' UJj-i. >\L;
1 UjinliJ
w £TA_J!ilft J"li.nt.
3
t feWSEJZ_L09_Che»t	
1 2-H-'80~ t-;&- HO

*1
U
rF,iJ5
» J*

l*
r»:«e
»s«

I<~'
• #.«*
CJ
J-'l
I*<
P &
#3

)•*

p»r


• 1 :1*
•v'l I
M»
- «<

P> »•
i#
Ml
£
n rr
•% .»>
•
»fv.
Ml
J«*
3*
ever
»i •*
~» >•
fit.
ftp
* ?V

i"
l«<
J*4
|U
11 #v
)l »p
»>v
n _
	
-> Oo
K
.'¦«
I'f
J'*
|f*_
Ml©
Tp
'•*.
2-15
£
rj-1*
iJV
Jit
2>l
»«*
3*
MJ •
o »:
lit.
r.r
Figure 18. Test results of Hitachi Zosen Pilot Unit at Albany, Georgia.

-------
The current plan is to install the new catalyst and complete the test
plan. A continuous demonstration run will then begin and last for 3 tr 6
months to show the stability of the process and to allow independent veri-
fication testing to be conducted.
The cost of the 600 series catalyst is anticipated to be about
20 percent less than that of the 500 series, after mass production is
achieved. This translates to about ¥2.6 x 10G/m3 ($10,612/m3 or $300/ft3).
To dispose of this catalyst it will be necessary to crush the catalyst
and separate the metal and chemical components by acid treatment. Then,
chemical treatment of the recovered catalyst will be necessary to remove the
vanadium which can be recycled The costs associated with this disposal
method have not been determined.
91

-------
ISHIKAWAJIMA-HARIMA HEAVY INDUSTRIES
Toyo Office
Tokyo
March 25, 1980
Nobuo Aoki, Department Manager
Hidetaka Aoki, Mecha ileal Engineer
David I lobley
Gary Jones
Doug Maxwell
Preceding page blank
93

-------
ISHIKAWAJIMA-HARIMA HEAVY INDUSTRIES
Istixkawajima-Harima Heavy Industries (IHI) is a major supplier of
utility boilers in Japan and has developed SCR technology that can be ap-
plied to these boilers. They have supplied seven systems to utility boilers
and have orders for eight more as shown in Table 16. All except one are
oil-fired boilers. One application of particular interest will be to a
600 MW coal/heavy oil fired boiler at Joban Electric's Nakoso Power Station.
This system will start up m 1983.
For both oil- and coal-fired applications, IHI recommends honeycomb
shaped catalyst elements. Three catalyst manufacturers make the catalyst
to IHI's specifications of chemical composition, dimensions and physical
properties. The catalyst is manufactured in individual elements 150 mm x
150 mm on the face and up to 1G00 in length. These are assembled by IHI
into modules by using steel catalyst holders. An example of the steel holder
is shown in Figure 19.
Figure 19. Example of IHI steel catalyst holder.
Preceding page blank
95

-------
TABLE 16. SCR SYSTEM APPLICATIONS BY IHI (AS OF MID-1979)
Operation
No.	Deliver to	Fuel	Type Capacity Comnence- Remarks
(Nm3/hr) merit Time
1 The Kawasaki Factory C heavy oil Moving 180,000
Ajinomoto Co., Ltd.	bed	(60 MWe)
Jan. 1978
Utility Steam
Generator
Low sulfur
crude oi]
Fixed
bed
1,020,000
(350 MWe)
Apr. 1978
Combined with
non-catalyst
dentrifica-
tion
Himeji No. 1 P/S
No. 5 boiler,
Kansai Electric
Power Co.
Low sulfur Fixed
heavy oil bed
490,000
(156 MWe)
June 1978
Kudamatsu P/S
No. 2 boiler,
Chugaku Electric
Power Co.
Naphtha,
NGL, Low
sulfur
heavy and
crude oil
Fixed
bed
1,050,000
(375 MWe)
Apr. 1979
Kudamatsu P/S
No. 3 boiler,
Chugoku Electric
Power Co.
Fixed
bed
1,900,000
(700 MWe)
Sep. 1979
Utility Steam
Generator
Low Sulf ur
heavy and
crude oil
Fixed
bed
534,COO
(156 MWe)
Feb. 1980
Nishi Nagoya P/S
No. 6 boiler,
Chubu Electric
Power Co.
Low sulfur
heavy and
crude oil,
Naphtha
Fixed
bed
1,316,400
(500 MUe)
June 1980
Utility Steam
Generator
Low sulfur
heavy and
crude oil,
Naphtha,
LNG
Fixed
bed
1,700,970
(600 MWe)
Feb. 1981
96

-------
TABI.E 16. SCR SYSTEM APPLICATIONS BY IHI (AS OF MID-1979)
(Cont inued)
Utility Steam
Generator
Low sulfur
heavy and
crude oil
Fixed
bed
534,000
(156 llWe)
Aug. 1981
10 Toyaraa Shinko P/S	Low sulfur
No. 2 boiler,	heavy and
Hokurj.ku Electric	crude oil
Power Co.
Fixed
bad
1,370,000
(500 MWe)
Nov. 1981
11 Higashi Niigata	P/S Low sulfur
No. 2 boiler,	heavy and
Tohoku Electric	crude oil,
Power Co.	LNG
F ixed
bed
1,660,000
(600 MWe)
Dec. 1981
12 Utility Steam
Generator
13	Utility Steam
Generator
14	Utility Steam
Generator
15	Nakoso P/S
No . 9 bo iler,
Joban Kyodo
Thermal Power Co.
Low sulfur
heavy and
crude oil,
Naphtha
Coal and
heavy oil
Fixed 1,020,000	Mar. 1982
bed	(375 MWe)
F ixed
bed
Fixed
bed
Fixed
bed
1,020,000
(375 MWe)
1,020,000
(375 MWe)
1,700,000
(600 MWe)
May 1982
Jul- 1982
Apr. 1983
97

-------
The catalyst modules are lowered into the reactor with a hoist and stacked
using structural steel within the reactor as support. As an example, the
catalyst bed at Chugoku Electric's Kudamatsu Station is 12a x 12ra, contains
144 modules, each weighing 1.5 tons. Five to eight minutes are necessary to
replace each catalyst module.
Two shapes of reactors are used. For coal-fired application, a vertical
downflow reactor at> shown in Figure 20 is recommended to avoid particulate
deposition. For oil-fired applications a horizontal flow reactor is recom-
mended (Figure 21).
IHI has tested this process on coal-fired flue gas at EPDC's Isogo
Power Station as part of a cooperative agreement. (See EPDC meeting report)
Early in the testing air preheater pluggage was experienced; however, modi-
fications to the elements and soot blowing procedure have alleviated this
problem. IHI markets the Rothemiihle, rotating duct air preheater and these
are installed at Isogo. In full-scale applications to coal-fired units
they feel that some water washing will be necessary.
IHI prefers the use of a cold-side ESP unless a hot-side ESP is
necessary to prevent fly ash contamination by NH3 deposition or for use with
low sulfur coals. With a cold-side ESP, the full dust load will clean the
air preheater and eliminate plugging problems.
In order to avoid problems with dust plugging the catalyst, honeycomb
channels 30 percent larger than normal are used for coal-fired applications.
There have been no problems wirh catalyst pluggage at Isogo where a hot-side
ESP is used. Deposition of fine particulates on the catalyst-has been ex-
*
perienced at other pilot plants. If a hot-side ESP i£ necessary, the
reactor can be designed with a higher flue gas velocity since the erosive
effects of large particles will not bo a factor. The higher velocity may
prevent adhesion of the small particulates.
98

-------
Flue gas
inlet
honeycomb mesh
Unit catalyst
Catalyst
layer
Flue gas
outlet
Figure 20. Fixed bed type reactor for coal-fired
applications (vertical downflow).
99

-------
Catalyst layer
Plue"
gas
inlet
3o _>
OCP
Unit module
Unit catalyst
Hoieyconb mesh
Figure 21. Fixed bed type reactor for oil-fired
applications (horizontal gas flow).
100

-------
Temperature control may be necessary to keep the flue gas temperature
within the 300 to 400°C range. Higher temperatures, e.g., 450C, can be
tolerated for short periods; however, extended periods at high temperatures
can cause catalyst sintering. Temperature control on one IHI boiler will
be accomplished by a split economizer as shown in Figure 22. During low
load operation, gas flow will be biased through the smaller economizer
section to maintain temperature. Tiiis is originally a Foster Wheeler design
for other purposes that has been adapted by IHI for SCR temperature control.
It is expocted to add on the order of 10 percent to the cost of the new boiler.
IHI indicated that 90 percent N0x control is possible, but difficult
due to gas velocity and NO^, concentration gradients across the duct. These
gradients make it difficult to control the	mole ratio at 1.0 at all
points in the duct, especially on large units. As a result, it will be
difficult for IHI to guarantee 90 percent N0y removal and low NH3 emissions
(<10 ppm). Ninety (90) percent removal with unrestricted NH3 emissions can
be guaranteed; however, air preheater pluggage problems may be increased.
In terms of cost, a coal-fired application is expected to cost
50 percent more than an oil-fired application. A 90 percent N0X removal
system will cost about 20-25 percent more than 80 percent control; however,
a 90 percent system with low NH3 emission requirements will probably cost
40-50 percent more than an 80 percent N0X control system.
The construction periods necessary for construction of SCR facilities
are also of ifttciusL. Construction schedules for IHI S
-------
SMALL
ECONOMIZER
////////
LARGE
¦ECONOMIZER
70-1988-1
GAS FLOW.%
HIGH LOAD
LOW LOAD
50
>50
50
<50
Figure 22. Split economizer arrangement for temperature
control on 1H1 boiler.
102

-------
11 k-Mb
" I Juno July
August
jy
Oi t
Nov.
houndjCIon
itructu-il Works

-------

1978
1979

Aug sept. Oct. Nov, Dec.
Jan. }Lb. Kunh April Ha> June Jul)
btru«.iual Works
KuUclor
Kuawior Insulation
h Imd
Insulation for Flues
Inj11 unit. 11L mJ Controls
Kjk
loses tor Auxiliaries
Inbcil Catalyst
t-ormal upcnitton




















Ihbl HlfV OK JJUIUR






		
	
	
	
- — ~
-H

Figure 24. Construction schedule of Kudamntsu No. 3 de-NOx Plant (New Boiler).

-------
JGC CORPORATION
Tokyo Office
Tokyo
March 25, 1980
Senjl Takenaka, Senior Deputy Manager
Mitsuaki Shiba
Shuichi Ohno
David Mobley
Cary Jones
Doug Maxwell
105

-------
JGC CORPORATION
JCC has been marketing a pellet-shaped catalyst and the parallel passage
reactor (Paranox process) for N0X control of ash-laden flue gas. They offer
two catalysts, designated JP ]02 (high temperature) and JP 501 (low
temperature). The activity of each of these catalysts are dhown in Figure 25.
The high temperature catalyst is designed for flue gas temperature greater
than 350°C and, therefore, is similar to other de-NOx catalysts. The low
temperature catalyst is designed to operate in the 200 to 300°C range with
periodic thermal regenerations. At these lower temperatures, NHi«HS0t,
deposits on the catalyst and progressively lowers its activity as shown
in Figure 26. The catalytic activity can be restored by heating to above
350°C. The frequency of regeneration depends on the S0X concentration in
the flue gas and the low temperature catalyst may not be suitable for boiler
applications after the air preheater with high S0x concentrations since the
period between regenerations would be very short. For example, with 500 ppm
of S0x the optimal operating temperature is 230-250°C. The outlet temperature
of the air preheater is only 150°C; hovevcr, upstream of the air preheater,
the high temperature catalyst can be used. The low temperature catalyst has
been tested on sintering furnace and coke oven gases and has been applied
commercially to a coke oven. This application has a low sulfur concentration
(50 ppm) requiring regeneration for 2 to A hours every five days.
The low temperature catalyst can be used fror 350 to 40C°C, but
higher temperatures can be detrimental. For example, with the 501 catalyst
at 380 to 390°C, SO2 oxidation increases and at 430°C, sintering takes place.
The parallel passage reactor consists of vertical gas channels bounded
by vertical catalyst beds. Two gas channels within a unit cell are des-
cribed in Figure 27. The catalyst is retained in the unit cells by stein-
less steel gauze which also allows passage of gaseous components to the
catalyst. A unit cell is 500 mm x 500 mm x 1 m and has a weight of about
250 kg. The catalyst can be changed by removing the covers on the catalyst
Preceding page blank

-------
100
H
eo
6
o

V
>
I
u
JI'501 High Tonp Cut
x
§
JP102 Lnv Tcnp. Cat.
200 ppm
pptn lt/0 10X
Nj! Dolnneo
250
300
no
200
Figure 25. Activity comparison of high and low
temperature catalysts by JGC.

-------
100
NO: 200 ppm
NHj: 220 ppm
SO2: 220 ppm
10%
200
150
100
PERIOD (Hour*})
Figure 26. Activity of low temperature catalyst
(JP 501) as a function of temperature

-------
Catalyst
Cas Flow
i I
, J P
$ ty
W '<& W
•; W
W, '0, YA
' '%
' ' v2
Figure 27. Arrangement of gas channels and catalyst
in parallel passage reactor.
110

-------
chambers. These covers are used to block the gas channels when fresh
catalyst is loaded from the top and vibrated into the catalyst containers.
This technology is owned by Shell. Studies by JGC have indicated that the
investment cost of this arrangement for relatively clean gas is not competi-
tive with other N0x control systems and, JGC has added the honeycomb monolith
to its product line. The monolithic catalyst technology belongs to JGC and
is preferred to the parallel passage reactor for de-NOx applications with
gases with low particulate concentrations because it has lower catalyst and
electrical costs. The vertical screens on the catalyst beds create a
resistance to gas phase mass transfer that is higher than with the monolithic
catalyst. Apparently, higher gas velocities are used to overcome this re-
sistance with consequent increased electrical usage by the fan. Catalyst
cost is also a factor and JGC indicated that the cost of the monolithic
catalyst was about 10 percent lower. Prior to the de-N0x catalyst and
reactor combination, JGC marketed the Shell (SFGD) process for S0x removal
which utilizes the parallel passage reactor. N0X can also be reduced
simultaneously by addition cf NH3.
JGC has not marketed their monolithic catalyst for application to
boilers since they do not sell boilers. In Japan, the boiler manufacturer
also supplies the SCR system if one is necessary.
Ill

-------
JAPAN SHELL TECHNOLOGY
Tokyo Office
Tokyo
March 10, 1980
K. H. Lang, Director/Manager, Technical Department
M. Miyauchi
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank
113

-------
JAPAN SHELL TECHNOLOGY
Japan Shell Technology, which is associated with the JGC Corporation,
markets the Shell Flue Cas Desulfurization (SFGD) process and the Paranox
process. Shell was involved in the development of both the SFGD and the
Paranox processes, and has installed the SFGD process at its Showa
Yokkaichi Sekiyu (SYS) refinery.
The SFCD process utilizes a regenerable acceptor material to remove
SO2 from flue gas. The SO2 is released in a concentrated stream when the
acceptor is regenerated with H2. N'0X can also be reduced simultaneously in
the process by addition of Nil3 since the sulfated acceptor material acts as
an X0x reduction catalyst. It should be noted that the SFGD and the Paranox
processes are completely different processes.
The SFGD acceptor material is a special alumina carrier impregnated
with copper and consists of extrudates (approximately 1.5 ram in diameter by
4 mm in length). The acceptor (contained within individual cells) is
loaded into the parallel passage reactor (a proprietary design). Catalyst
loading involves careful inspection of the cell elements, filling the cells
with catalyst and reassembling the unit cells. These steps require ex-
perienced supervision to prevent rough handling which could cause weld
failure and catalyst attrition. A weld failure can allow the acceptor to
migrate from within the cells.
A third batch of acceptor is being installed at the SYS refinery
due to somewhat premature loss in activity of the second batch. Strict
quality control is necessary when manufacturing the acceptor.
*
Currently, the only application of the SFGD as designed by Shell is
at the SYS refinery on a boiler firing residual fuel oil. The fuel is from
a Middle Eastern crude and contains approximately 2.5 percent sulfur and a mod-
erate amount of vanadium. Early on, there were problems with fouling; however,
Preceding page blank
115

-------
soot blowing has eliminated this. Soot blowing is performed every day on
a sequential basis to each o£ the soot blowing jets in order to reduce the
instantaneous steam demand. The steam requirement is about 2-3 tons of
18 atm. steam per day. At SYS, steam blowing is preferred to sand cleaning
since the latter fouled the acceptor and was not as effective.
The SYS facility achieves 90 percent SO2 and 50-60 percent N0X removal
which is sufficient to comply with local regulations. The N0X removal could
be improved if necessary. SO2 is converted to elemental sulfur via the
Claus/SCOT processes which together achieve 99.9 percent recovery, give a
very pure product and no undesirable streams.
If both SO2 and N0X removal is the goal, it is theoretically beneficial
to operate two separate reactors, one for each pollutant, rather than a single
dual-purpose reactor, due to the fact that action that improves efficiency
for one pollutant tends to lower the efficiency -with respect to the other.
However, the SFCD process provides a potentially advantageous
compromise that achieves simultaneous reduction of SO2 and N0X adequate for
many applications.
The parallel passage reaction not only displays the advantage of a
low pressure drop configuration in which fouling can be adequately con-
trolled, but the relatively large amount of metal internals involved also
provides for good heat transfer in the exothermic SO2 acceptance reaction
and during the regeneration cycle.
There are three commercial Paranox processes of JGC design at
Kashima Oil, >"uj i Oil and Nippon Steel. These plants, which control only
N0x and use the JGC catalyst, have experienced no problems. These plants,
except Nippon Steel, are no longer operational since N0X removal at these
locations is no longer required. When they were in operation, the two
116

-------
plants at the oil refineries achieved 90 to 96 percent N0x removal. The
unit at Nippon Steel is treating 200°C gas from a coke oven and removes
95 - 98 percent of the NO .
r	A.
117

-------
KAWASAKI HEAVY INDUSTRIES
Kobe Works
Kobe
March 11, 1980
Senji Niwa, Manager of Plant Engineering
S. Nireki, Plant Engineering Division
David Mobley
Cary Jones
Doug Maxwell
119
Preceding page blank

-------
KAWASAKI HEAVY INDUSTRIES
Kawasaki Heavy Industries (KHI) has developed several catalyst shapes,
but the primary ones are shown in Figure 28.
Passages can be made
from 4mm to 13ram
Extruded Honeycomb
Lengths can be
made from
50 cm to
lm.

J 30mm
r 4
20
mm

Figure 28.
Tubular or Pipe
KHI catalyst types.
The honeycomb is recommended for oil-fired applications whereas the tube type
is recommended for coal. With both types the catalyst is V2O5 on a Ti02
carrier. Typical spac^ velocity and specific area data are shown in Table 17.
TABLE 17. DESIGN DATA FOR KHI CATALYST TYPES
Catalyst Type
Specific Area
(m2/m3)
Space Velocity*
(hr*1)
Honeycomb
300
5000
Tube
170
3000
*80 percent removal
Preceding page blank
121

-------
Catalyst volume for the honeycomb is defined as:
L x H x W x N
N is the number of catalyst units. For the tube shape the volume is defined
as:
(outside diameter)2 x L x N
Currently, KHI is developing a catalyst formulation that will minimize
conversion cf ^02 to SO3 in order to reduce formation of NHiiHSOi* downstream
of the reactor. Their effort has been successful, but N0X removal efficiency
is lower for the new formulation as shown in Table 18.
TABLE 13. EFFECT OF CATALYST F0R>rULATI0N ON NOx REMOVAL
(330°C, 90% NOx Removal)

Conversion of SO2 to SO3
Relative NOx Removal
Catalyst Type
(% of inlet concentration)
for same Catalyst Volume
Normal
2-3
1.0
Low SO2 Conversion
0.5
0.8
The KHI catalyst to be installed at the SCE Huntington Beach power plant,
the EPR1 pilot unit at the Arapahoe power plant and at the EPDC Takehara
power station will be of a formulation intermediate to those shown in the
table.
Catalyst deactivation has also been studied and KHI has found that
potassium compounds such as KC1 and K2SO1, will poison the catalyst.
It is necessary to keep the flue gas temperature in the range of.300 to
400°C to prevent-damage..	-gas&s that do not contain SO2, this range can
be extended to 250 to 450°C. Lower temperatures will allow deposits to form
and activity to drop. Higher temperatures will damage the Ti02 probably by
causing sintering.
122

-------
In application, the catalyst is manufactured iuto steel frauu- cases to
allow quick stacking in the reactor and to \o,d 
-------
100
UJ
o
UJ
Od
X
O
3C
"HXiox [10L 1110
Figure 30. N0X Removal Efficienc" and Effluent
XH3 Vs. NH^J0x Kol Ratio.
124

-------
one in which the NH3 is converted to N0x by a converter containing stainless
6teel pellets and one with no NH3 conversion. NHj concentration is deter-
mined by difference.
In terms of overall system arrangement, KHI prefers to have particulate
removal downstream of the SCR reactor because the fly ash tends to keep the
catalyst surface free of deposits. Erosion of the catalyst at the flui: gas
inlet was previously a significant problem, but this has been _>olved by either
dipping the tips of these catalyst elements into a hardening solution or making
a harder catalyst. The N'Ox removal capability is not affected.
The process is controlled by using two signals: a feedforward signal of
N0X flow rate which is the product of gas flow (boiler load) and N0X concen-
tration, and a feedback N0X removal signal calculated by measuring the outlet
N0X concentration. The feedforward signal is controlling the feedback
signal is used as a trim on the primary control. Unstable operation occurs
when only the feedback signal is used.
KHI either has supplied or has orders for seven SCR systems as shown in
Table 19. There are two large units; one at EPDC's coal-fired boiler at
Takehara and one at SCE's Huntington Beach Station in the US.
125

-------
TABLE 19, SCR SYSTEMS SUPPLIED BY KHI
Client
!.int of ?lani
Fuel
Loral Ion
Cap«cUy(Kss3/h)
(juinilty Type
1). 1 Ivri y Date
Hcdvy
Jlti'uBl rlrB, 1 Id
Marl«a fcurka
UPC

I tpan
7, COO
I f«t Myl 1r IK\
Rinnval 5*»lia
V. f 197A
Tnkyo StilKaura
EUctrlc Co., Ltd.
YnVoKima Works
11C
kunviyrtwa, Japan
1?,900
) ditto
Harrh 1478
Myogo Prefecture
Hultugava <>i.wnge
C<-nt er
Ho ivy
Oil
Hyogo,
Jupan
4,700
] ditto
H-rch 1978
fUkOiakl Heavy
Induct r i«¦
Akabhl VorVa
Jk4vy
Otl
Hyogo,
Japan
2>.00Q
1 ditto
Nov<*&b«T 19J8
F1 ¦: trie Power
Ar ipali'^ C< rura' inj»
SI n t l«*n
Con I
n« uvrr
USA
, CoWtnJo
ft,ooo
I Cut *lyt If W\
Ri O..". »l n I Ul
T«*st l.tdtile
f>t j>t» mhrf 19/V
Electric Povur
Di vi lopuinr Co
TokiYinra Fowrr
r>t ich
C'nrraiIng 5t«t loD
on
Run "s^ad,
Ca 1 I f o i r» 1 a
J4V.W0
1 ditto
*>i pic t>*"* 1980
ro
a*

-------
MITSUBISHI HEAVY INDUSTRIES
Hiroshima Research Station
Hiroshima
'•sarch 13, 1980
Masumi Atsukava, Deputy General Manager
Yoshihiko Nishimoto, Asst. Chief Research Engineer
Toru Seto
David Hobley
Jumpei Ando
Gary Jones
Doug Maxwell
Power System Headquarters
Tokyo
March 26, 1980
Taaio Fujita, General Manager
Tadamasa Sengoku, Group Manager
Kiyoraitsu Twata, Project Group Manager
Atsushi Morii, Project Group Member
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
127

-------
MITS JBISHI HEAVY INDUSTRIES
Mitsubishi Heavy Industries (MHI) is actively developing selective
catalytic reduction (SCR) technology and. as of March 1980, had 21 orders
for NOx removal plants including one for a coal-fired boiler (Figure 31).
Mill anticipates more coal-fired applications in the future when boilers that
were c aigned for coal use, but currently burn oil, switch back to coal.
For cor 1-fired applications in Japan there are two types of arrange-
ments as shown in Figure 32. Each has advantages and disadvantages. In the
Type "A" system, with a hot-side ESP, the main problem to overcome in deposi-
tion of N'Hi,HSOu on the air preheater elements and catalyst surface. In the
Type "B" system, the particulates Lend to clean the catalyst and preheater
surfaces, but also tend to erode the catalyst. MHI uses erosion resistant
catalyst and a dummy layer to prevent erosion in the Type "B" systsm.
MHI and catalyst manufactucrs, under joint-development agreements, have
developed many types of catalysts which meet various requirements for SCR
applied to boilers with different fuels. The major features of catalysts are:
(1)	Long life and less deterioration.
(2)	High strength for temperature fluctuation for boilers.
(3)	Erosion resistance.
(A) Less oxidation of SO2 to SO3.
MHI and catalyst manufacturers take care of the disposal of spent catalyst.
MHI prefers the V205/Ti02 catalyst and will modify the concentration, of
*
other components and the manufacturing process based op the flue gas composi-
I
tion. Tlie catalyst is guaranteed for one year, but a two or three year life
is expected.
Preceding page Wank

-------
6000
5000
4000
E
4-1
'JJ
>1
•J
"3
O
c
ctf
3
cr
3000
2000
1000
BOILER
NO, OF PLANTS
Heavy Oil
Firing
Coal Firing
20
12
6
12
6
12
6
1979
1980
1981
1982
Figure 31. Status of orders received for MHI NOx removal plants.
130

-------
Mot
ESP
V
FGD and
Stack
Boiler
Air Preheater
Arrangement "A"
Example: EPDC Takehara Power Station
Cold
ESP
\/
-&• FGD and
Stack
SCR
Boiler
Air Preheater
Arrangement "B"
Example: Chugoku Electric
Shimonoseki Power Station
Figure 32. flue gas treatment alternatives for coal-fired boilers.
131

-------
NOx removal Is adjusted by changing the NHj:NOx mole ratio as follows:
reole ratio
removal
0.61
0.81
60%
80%
At 80 percent removal, NHa in the outlet gas is <5 ppm. Both an ESP and an
FCD unit will reduce \'H3 emissions below this amount if located downstream.
MH1 has not found a suitable MI3 analysis technique although several
have been tested. A major analysis problvra is obtaining a representative
sample since there are concentration gradients in large ducts.
NH3 that has deposited on the fly ash is removed from the collected ash
by spraying with a small amount of water..- The overall reaction is:
The liberated NH3 is collected in a hood and sent to the stack.
The SCR system supplied by MHT for coal-fired applications is located
at Chugoku Electric's Shimonoseki Power Station. Start up i«; scheduled for
April 1980 and all of the flue gas from the 175 MW boiler will be treated.
nitially, the NOx removal will be only 50 percent which is sufficient to
satisfy Chugoku Electric's agreements, however, the reactor will be sized for
80 percent removal. Design data for the plant are listed in Table 20.
Tive stages of homogeneous catalyst will be installed. Each stage is
approximately 2 meters in height and contain 0.65 reefer3•of'catalyst. A
dummy layer of catalyst is installed at the reactor entrance to prevent
catalyst erosion. Experience has shown that only the-entrance face of the
catalyst suffers from erosion. Provisions are included to allow NOx removal
to be increased to 80 percent by increasing the catalyst bed height to 0.8
meter and the NHsiNOy mole ratio to 0.81.
(NH i») 2 SOi, + Ca(0H) 2 —*• CaS0„*2H20 + 2^3
132

-------
TABLE 20. DESIGN' DATA FOR MHI SCR SYSTEM AT
SHTuO\'OSEKI POWER STATION
Item
Design Value
Flue gas flovrate
Inlet temperature
Inlet NOx
550,000 NmVhr
NO removal
x
NH3*N0 mole ratio
NHj emissions from reactor
Inlet SOo
Reactor dimensions
370 °C
500ppm
50%
<0.51
<5ppm
1600ppm
11.8mW x 8.5mD x 13.3mH
Temperature control is achieved by an economizer bypass. At full load
the flue gas temperature is 370°0, but at 25 percent load the temperature
drops to 300°C. Tne minimum acceptable gas temperature is above 300°C,
thereby necessitating the bypass for temperature control.
Soot blowers are installed at each stage as a precautionary measure.
No new fan is necessary due to the FGD fan being overdesigned. The cost of
the retrofit is approximately ¥2xl09 ($8.2xl06) or $47/kW including boiler
modifications.
MHI has also been involved with Joban Electric, a cooperative of three
power companies, on testing SCR applied to coal and coal/oil mix boilers. The
tests are reported to be favorable and will be announced by Tokyo Electric in
early 1980. (See meeting report with Tokyo Electric.)
133

-------
NCK INSULATORS
Headquarters
Nagoya
torch 17, 1980
Masavoshi Asai, Ceneral Manager
Takeshi Kawada, Assistant Manager
Takashi Uno
T. Fujimoto
K. Sawada
David Mob ley
Jumpei Ando
Gary Jones
Doug Maxvell
135
Preceding page blank

-------
NCK INSULATORS
NGK is one of four major catalyst manufacturers in Japan that supply NOx
reduction catalysts to SCR process vendors. The catalyst is a coated honey-
comb type using a ceramic substrate although NCK can produce homogeneous
types.
Wall thickness of the coated type can be considerably thinner than that
of homogeneous type. Therefore, coated type gives higher N0X removal effi-
ciency with lower pressure drop as shown in Figure 33. Catalyst material is
crated on the substrate with a thickness of 0.15 mm.
Hexagonal, square and triangular channels are available in various
rhannel sizes. Ceramic substrates are reportedly superior to metal substrate
for several reasons including:
•	low thermal expansivity
•	high strength
good catalyst substrate bond
acid resistance
The difference in the coefficients of thermal expansion between the
catalyst and the ceramic substrate are only 1_ percent, whereas between the
catalyst and metal substrate the difference is 150 percent. It is felt that,
with the metal substrate, temperature cycling will cause catalyst exfoliation
and aggravate catalyst erosion problems. The rough surface of the ceramic
creates a good bond between catalyst and substrate which also inhibits
catalyst exfoliation.
The hexagonal channel shape was common but the trend in Japan, which is
supported by NGK, is towards the square channel shape because of its high
strength. Examples of each type are shown in Figures 34 and 35.
Preceding page blank
137

-------
/' '(solid type)
Figure 33. Characteristics of honeycomb substrate (coated type).
138

-------
Figure 34. NCK hexagonal honeycomb catalyst substrate (200 mm square x
350 mm, 7 mm hydraulic diameter, 0.8 mm wall thickness).
139

-------
Stars
fWw/ Fillrr
Figure 35. NGK square honeycomb catalyst substrate (150 mm square x 350 mm).
140

-------
Table 21 shows the properties of the honeycomb in comparasion with
those of a pellet.
TABLE 21. PROPERTIES OF NCK HONEYCOMB SUBSTRATE


Wall
to Wall
Channel Width


Property
Unit
7 mm
Hexagonal
7 mm
Square
10 mm
Hexagonal
10 mm
Square
Pellet Diameter
7 mm
Specific
Surface
Area
m2/n>3
437
469
277
364
543
Bulk
Density
kg/m3
480
425
630
320
800
Superficial
Gas
Velocity
ra/g
8
8
8
8
1
Pressure
Drop at
Design
Velocity
mm HjO/m
48
37
45
22
350
Wall
Thickness
nun
1
0.8
2
0.8
-
The ceramic material is mullite and the catalyst consists primarily of
oxides of titanium and vanadium. A third component is added to suppress the
SO2 oxidation reaction. The reaction can be suppressed as much as desired,
but at the expense of NOx reduction efficiency. This suppression is necessary
especially for high temperature gases in which a considerable fraction of the
inlet SO? is converted to SO3 as shown in Figure 36.
NGK has experienced some erosion of t^e catalyst at the inlet face and
are currently improving this characteristic. They feel that ceramics are
superior to metal substrates for the reasons mentioned previously. After this
face-erosion, the ceramic substrate will remain and prevent further erosion.
Continuous erosion of newly exposed catalyst can occur with homogeneous
catalyst types.
141

-------
~
300	350	400
Gas temperature (°C)
Figure 36. Conversion of SO2 to SO3 by catalyst as a function of gas temperature.
142

-------
The catalyst appears to be capable of achieving >90 percent NOx redac-
tion at typical SCR operating conditions. Cataly t performance is shown in
Figure 37 through 39. The first two figures also compare the performance of
the honeycomb catalyst to that of a pellet type. NGK hs not sold any of
this catalyst for coal-fired applications, although they are conducting tests.
There are two oil-fired applications; one is at Fugi Oil's refinery on an
industrial boiler. The catalyst life guarantee depends on each application.
In any case, NCK will guarantee the catalyst to the process vendor who will
guarantee the catalyst to the user.
143

-------
100.
8-t
>
U
3
-o
70 -
60 "
7 cam hydraulic dl;
pellet-shaped cat;
(4 - 6 nrra dia)
10 mm hvdraulic
-T
10
T
12
lyst
dia
14
16
Space velocity * 10 3 (H )
NGK honeycomb-
shaped catalyst
Pellet-shaped
catalyst
Gas velocity (m/s)
Gas temperature (°C)
NH3/N0x (mol/ntol)
Gas composition (ppm) :
N0x
S02
SO 3
7.0
360
1.0
100
200
8-10
0.6
380
1.0
150
80
3
Figure 37. Effect of space velocity on N0X reduction efficiency.
(Small scale tests)
144

-------
¦J
c
V
3
¦o
V
o
z
100-
90-
r: 80-
70-
60-
7 mm hydrauli
pellet-shaped ca
(4-6 mm dia)
10 mm hydraulic di
250
—I—
300
—I—
350
dia
:alyst
4 00
450
Gas temperature (°C)
NGK honeycomb-	Pellet-shaped
shaped catalyst	catalyst
Space velocity (hr x)

6,000
6,000
Gas velocity (m/s)

7
0.6
NH3/NO (mol/mol)

1.0
1.0
Gas composition (ppm):
NOx
100
150

S02
200
80

S03
8-10
3
Figure 38. Effect of gas temperature on N0X reduction efficiency.
(Small scale tests)
145

-------
50
40
Q 0
30
nj
u
20 z
10
0
0.6 0.7 0.8 0.9 1.0 1.1 1.2
NH3/NOx ratio (mol/mol)
NGK Honeycomb Catalyst
Hydraulic dia. of channels
(mm)
Space velocity
(hr~ 1)
Gas velocity
(m/s)
Cas temperature
(°C)
Gas composition: NOx
(ppm)	SO2
SO 3
6,500
6.8
380
100
200
8-10
Figure 39. Effect of NHa/NO^ ratio on N0X reduction
efficiency and NH3 carryover.
146

-------
APPENDIX B
INSTALLATION'S WITH NOx CONTROL SYSTEMS
147

-------
CHUBU ELECTRIC
Chita Power Station
Chita, Aichi Prefecture
March 18, 1980
Suewo Takeuchi, Superintendent
Tomoaki Shinada, Assistant Superintendent
Mr. Tehira, Engineering Assistant
David Mob joy
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page
149

-------
CHITBU CLECTPIC
The Chita Power Station consists of six boilers of which three use SCR
systems and one uses thermal de-NOx. The boilers and associated data arc
shown in Table 22. NOx control equipment is installed to meet the regula-
tions of the local prefectural government and these are summarized in Table 23.
NO. 4 Boiler - SCR Unit
The SCR system on this boiler was supplied by Mill and began continuous
operation in March 1980. Descriptive data on this unit are described in
Table 24. The reactor and catalyst arrangement are shown in Figure 40. The
catalyst is stacked in four stages with each stage consisting of two layers
of catalyst packages which, in turn, consists of 50 catalyst elements.
The cost of this installation it: 2.5 x 109¥ ($10 x 106) which is rela-
tively expensive. This is due primarily to the fact that it is a retrofit.
Ductwork proved to be a significant cost item. The catalyst life guarantee
is for one year and the power station is hoping for two.
No additional operators are required to operate the de-NOx unit. Some
labor is required to unload the periodic truck deliveries of NH3 (Figure 41).
When necessary, the catalyst will be changed by using a small hoist (Figure
42). This operation is expected to take a month or more and may require a
50 to 100 percent increase in the annual outage, which can range from 30 to
50 days.
N'o. 5 and 6 Boiler - SCR Unit
These units were supplied by Babcock-Hitachi and have been operated for
two years. Descriptive data are shown in Table 25. The catalyst in each is
arranged into four stages each 110 mm thick and 116.6 n2 in surface area. In
total, each boiler uses 125 tons or 95.4 m3 of catalyst. The layout of a
single reactor is shown in Figure 43.
151
Preceding page blank

-------
TABLE 22. NOx CONTROL AT C1IITA POWER STATION - CliOBU ELECTRIC
k/i
ro
Boiler
MW
375
375
500
700
700
700
Fuel
ResId or
Crude Oil
Res id or
Crude Oil
Resid or
Crude Oil
Resid or
LNC
LNG
LNC
• i_s f s s r_-sva m •
NOx Controls

Combustion Modifications
Combustion Modificiatons,
MI!I Thermal de-NOx
Combustion Modifications
Combustion Modifications,
Combustion Modifications,
Babcock-Hitachi SCR
Combustion Modifications,
B.ibcock-llitachi SCR
Outlet NOy ppm
(3Z 02)
<	148
<	106
<	106
<	21
1 11
<	11

-------

TABLE 23. EMISSION LIMITATIONS
AT CHITA
POWER STATION
Item
Regulation

Compliance Technique
Water
COD £ 10 mg/1
•
Chemical treatment

TSS £ 15 mg/1
•
Rainwater to oil separator

Oil £ 1 mg/1



pH 5.8 - 8.6


NO
As shown in

Combustion modifications

Table 22

(controlled mixing and OFA]



SCR with NH3


•
Thermal De-N0x with NH3
S02
-
•
Low sulfur fuel
Ash
< 20 mg/Nm3
•
ESP's on boilers 1, 2,



3 and It
Noise
< 50 phones
•
walls around transformer


•
FD fan insulated


•
valves covered
153

-------
TABLE 24.
DESIGN DATA OM MHI SCR SYSTEM, 04 BOILER,
CHITA POWER STATION
General
Gas Volume
NOx Removal
NOx Inlet
N0X Outlet
Reactor
Temperature
NH
NHj: NOx Mole Ratio
NH3 Outlet
Catalyst
Space Velocity
Material
Shape
Pitch
Manufacturer
1,960,000 NmJ/hr
80%
100 ppm
<	20 ppm
388°C
0.9
<	10 ppm
5430 hr"1
Transition Metal Oxide
Hexagonal Honeycomb
8 TTTTT1
NGK
154

-------
De-NO Reactor
ammonia
catalyst Layer
Boiler
to stack
forced
draft
fan
air
heater
steam
air heater
Reactor
NH3 Injection
gas
catalyst
package
each blank represents
50 elements
5x5x2
Catalyst Element
6.8 W1.2 mm wall
pitch 8 mm
Figure AO. Catalyst and reactor arrangement on
Ob boiler, Chita Power Station.
155

-------
mmwtzrmu
Figure 41. Truck delivery and unloading of ammonia.
Figure 42. Hoist arrangement for catalyst changing.
156

-------
TABLE 25. DATA SUMMARY ON BABCOCK-HITAC11I SCR SYSTEM -
if5 AND 06 BOILER, CHITA POWER STATION
Reactors per unxC
Catalyst Type
Catalyst Size
NH3:NOx Molt Ratio
NO Removal, design
N0x Inlet
NO Outlet
x
Catalyst Manufacturer
80%
30-40 ppm
8-9 ppm
Hitachi
Pellet
0.9
5 mm x 5 mm
2
No. 2 Boiler - Thermal de-NOx lipit
This is an MHI supplied system which achieves about a 30 percent NOx
reduction. A NH3:N0x mole ratio of 1.0 is used and NHs breakthrough is main-
tained at <10 ppm by biasing the NH3 flow rate through the various NH3
injection nozzles. The nozzles are arranged in two banks with 8 and 7 noz-
zles, respectively. Flue gas temperature determines the NH3 flow rate to
each bank of nozzles. NOx gradients in the duct determine the hole spacing
in the nozzle. This is checked and modified once per year. Total NH3 flow
rate is controlled based on the fuel flow rate. The dual injection grid
arrangement was preferred to the alternative, H2 injection, for safety and
expense.
The cost of the total system uas ¥660 x 106 ($2.64 x 106) which includes
equipment, NH3 storage and vaporization, and the license fee.
The station hfiS experienced some problems with air preheater plugging,
especially on the 112 boiler after the thermal de-NOx- Addition of a hot-
side soot blower did not help ivjch and water washing is required once or
twice a year. A water wash is initiated when the AP is > 140 mm H2O and about
200 tona of water are used per .ash. No special wastewater treatment is
performed.
157

-------
air
to
boile r
to stack
Figure A3. Layout of SCR reactor on 95 boiler, Chita Power Station.
158

-------
CHUGOKU ELECT?IC
Xudamatsu Power Station
Okayama Prefecture
March 12, 1980
M. Kikkawa, Ceneral Manager
David Mobley
Jurapei Ando
Gary Jones
Doug Maxwell
Shimonoseki Power Station
Shimonoseki, Yamaguchi Prefecture
March 28, 1980
Takuro Kisse, General Manager
C. Yada, Vice Manager
. Seo, Chief of intenance Division
Izumo, Chief of Environment Division
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
159

-------
CHUCOKU ELECTRIC
At Chugoku Electric's Kudnmatsu facility, the flue gas from boiler 62
(375 >W) and 113 (700 Mis') is treated for NOx by SCR systems supplied by THI.
The plant has an agreement with the city to keep total NOx emissions below
154 NmVhr. The de-NOx system on the "2 boiler is a retrofit that consists
of a single reactor which reduces NOx from 160 ppm to <32 ppm. The 03 boiler
included SCR in its initial design and consists of two parallel reactors which
reduce the N'Ox from 130 ppm to <26 ppm. Both boilers are fired with low
sulfur oil. The reactors are shown in Figure 44 while the NH3 storage/vapori-
zation system is shown in Figure 45.
The IHI de-N'Ox systems are guaranteed for 80 percent removal at an
S"H3-.>\0x rsole ratio of 0.90. Initially, the NOx removal was 85 percent, but
recently this has been reduced to 82 percent by using a mole ratio of 0.85.
The i12 unit has operated for one year with no degradation and the plant
aanager is expecting at least one more year before catalyst replacement is
necessary. Pressure drop through the reactor is 50 to 60 nroHjO.
There have been problems with air preheater pluggage by NJUHSOi,. On
the £2 boiler, the air preheater has required washing every two months and
the air prehea!.ers cn the (t3 boiler have also required washing. The Plant
is trying to improve the soot blowing technique to minimize air preheater
plugging. Normally the pressure drop is 100 mmHzO, but washing is required
when this increases to 300 mmHjO. 5000 tons of water are required for one
washing. Wash water containing the ammonium sulfates is stored and neutral-
ized at a rate of 50 T/h and stored for reuse.
The flue gas conditions that lead to preheater plugging are <4 ppm NK3,
4 ppm SO3, and 100 to 110 ppm SO2 at the outlet of the reactor. Gas exits
the preheater at about 140°C. The deposits are experienced in the inter-
mediate temperature zone.
Preceding page blank
161

-------
H2 Boiler (375 MU)
#3 Boiler (700 MW)
One of two SCR reactors
Figure 44. SCR reactors at Chugoku Electric, Kudamatsu Power Station.
162

-------
Figure 45. NH3 Storage and Vaporization System at Chugoku Electric's
Kudanatsu Power Station.
163

-------
Currently, chev soot blow with air, but plan to switch to steam in order
to avoid blowing the hot-side with cold gas and potentially causing deposits
to form on the hot end elements. The low temperature side is blown once per
shift and the high temperature side once per day. A soot blowing cycle
consists of blowing for two hours with 14 kg/cm2 air. Previously, the high
cenperature side was blown only once per three days. Further improvements
including going to higher diameter nozzles are planned in order to reduce
required water washings to once every three months.
v>'ith SCR, NH3 emissions are lower than previously when NH3 was injected
for ash conditioning. NH3 in the reactor outlet is monitored continuously
by a Shimadzu instrument which utilizes a chemiluminescence NOx analyzer. A
sample stream in which NH3 has been oxidized to NO is compared to an un-
oxtdized stream and the NH3 concentration is calculated by difference.
The de-NOx units are controlled by using the product of gas flow and
NOx concentration. Removal is used as a trim signal. There is no tempera-
ture control, but NH3 injection is stopped if the temperature is <275°C. On
the ?3 boiler, this corresponds to about 1/3 load.
The capital costs of the SCR system applied to the 03 boiler are shown
in Table 26 and the operating costs in Table 27. These costs translate to
$14.3/kW and 1.84 mills /kWh assuming 70 percent power utilization. However,
since the catalyst is expected to last much longer than one year, the actual
operating costs will be less than these estimates. The capital cost of the
SCR system on the "2 boiler was ¥2.2xl09 ($9xl06) or about $24/kW. This
system was more expensive since it required an ID fan and more complicated
ductwork.
The labor requirements of the SCR systems are reported to be small.
Station personnel indicated that no operational labor was required and,
except for water vashing the air preheater, no regular maintenance labor is
necessary. Catalyst replacement is estimated to require six people for two
weeks once every two years.
164

-------
TABLE 26. CAPITAL COST BREAKDOWN OF SCR SYSTEM ON NO. 3 BOILER (700 MW)
CHUCOKU ELECTRIC, KUDAMATSU POWER STATION


Capital
Cost

Component
¥10*

$10°
Percent
Catalyst
1.1

A.5
44
Reactor
0.7

2.9
28
NH3 System
0.24
• «
0.98
10
Testing and Start Up
0.16

0.65
6
Other
0.30

1.2
12
TOTAL
2.5

10.2.
100
TABLE 27. OPERATING COST BREAKDOWN OF
CHUCOKU ELECTRIC, KUDAMATSU
SCR SYSTEM ON NO. 3 BOILER (700 MW)
POWER STATION


Cost/Year


Component
¥10b

$10b
Percent
Interest
250

1.02
13
Depreciation
321

1.28
16
Tax
35

0.14
2
Catalyst
1131

4.52
58
Power
83

0.34
4
nh3
76

.30
4
Others
62

.25
3
TOTAL
1958

3.22
100
Basis: 7 years depreciation
10% interest
1 year catalyst life
Note: Japanese utility power company economics include the catalyst both
in the investment cost, which is depreciated, and in the operating
cost.
165

-------
Chugoku Electric's Shimoneski Power Station consists of a 175 MW coal/
oil-fired boiler (No. 1) and a 400 oil-fired boiler (No. 2). The coal/
oil-fired boiler will soon switch to 100 percent coal-firing. The coal-fired
boiler has recently been fitted with a retrofit SCR system that was scheduled
to begin operation in April 1980 and, as such, is the first commercial appli-
cation of SCR technology to a coal-fired boiler. Some design data on the
SCR system are shown in Table 28. Construction on the retrofit began in
October 1979 and was essentially complete by the end of March 1980. The
typical annual outage period of one month was extended to two months to allow
connection of the modified ductwork to the boiler. When operational, the
entire flue gas stream will be treated by an SCR reactor, an air preheater,
a cold ESP and a limestone/gypsum TCD system, Figures 46 and 47. A plot
view of the boiler and SCR system is shown in Figure 48.
TABLE 28. DESICN H\TA FOR THE MHI SCR SYSTEM INSTALLED
AT THE SHIMONOSEKI POWER STATION
Item
Design Value
Flue gas flow rate
Inlet temperature
Inlet N0x
N'0x removal
NH3 N0x mole ratio
NH3 emissions from reactor
Inlet SO2
Inlet SO3
SO? oxidation by SCR catalyst
Reactor dimensions
Space velocity
550,000 Nm3/hr
370°C
500 ppm
50% (Reactor sized for 80%)
<0.52
<5 ppm
1600 ppm
20-23 ppm
<1.0%
11.8m W x 8.5m D * 13.3m H
3000 h"1
166

-------
Figure 46. SCR reactor at Shiraonoseki Power Station.
167

-------

Figure 47. Reactor, air preheater and ESP at Shimonoseki Power Station.
168

-------
Boiler
~
-de-NO Reactor
ESP
FGD
NH3 storage
and vaporization
stack
Figure 48. P.'.ot view of No. 1 boiler, de-NOx system, ESP and
FGD system at Shinonoseki Power Station.
169

-------
The SCR system is supplied by MHI and designed for 50 percent NOx
removal since this is sufficient to meet the local emission standards. In
practice, an NT^NOx mole ratio of 0.56 will be used to get 55 percent
removal with low KHj emissions. Higher removal levels may be required in
the future if the emission standard becomes more stringent or if the nitrogen
content of the coal increases, so the system was designed for these antici-
pated levels. To modify the now existing SCR system for 80 percent NOx
control will involve installing a larger amount of catalyst.
The square honeycomb catalyst has an 8 mm channel diameter, a 1.8 mm wall
thickness, a 150 mm x 150 mm face and is 450 mm in length. The formulation
is based on the popular Ti02/V?0$ combination including a third component
and is manufactured by Sakai Chemical. The catalyst elements are assembled
into reusable steel basket? by the catalyst manufacturer. The catalyst is
loaded by hoisting the basket to the appropriate level and setting it on a
cart. The cart runs on rails that are integral with the exterior walkways
of the reactor and is rolled past a removable wall, removed from the cart
and set in place. It is estimated that a catalyst change will require 20
people 15 days to complete, working seven hours per day. Disposal of spent
catalyst is provided by the manufacturer.
Die reactor is a vertical downflow unit with a layer of dummy catalyst
elements at the inlet face to prevent catalyst erosion by fly ash. The
operating temperature for the SCR process is 360°C and will be controlled
so as not to drop below 340°C, although 320° may be an acceptable lower
limit according to MHI. Temperature control is provided by an economizer
bypass and includes a flue gas mixing zone to insure an even temperature
distribution across the reactor. The design of gas mixing zone is shown
in Figure 49. KHa is injected upstream of the mixer to take advantage of
the mixing effect. An isometric flow diagram of the whole system is shown
in Figure 50 and the layout of the reactor is shown in Figure 51. The drawing
of the reactor shows space at the bottom that can be ined to install a low
temperature economizer. It can also be used to install additional catalyst
if higher NOx removal levels are required.
170

-------
NH,
fat
Flue
Cs s
J
Duct
/

Figure 49. Design of gas mixing zdne.
171

-------
Boiler
Economizer
Bypass
Cas
Mixing
Eco
/\
Air Prehcatcr
Reactor
NH3
Storage
and
Vaporization
Figure 50. Isometric flow diagram of boiler and
SCR reactor at Shimonoseki Power Station.

-------
u>
16700
Top View
m
Catalyst
Containing
144 Catalyst
Elewcnts
11800
Figure 51. SCR reactor and catalyst basket at Shimonoseki Power Station
(dimensions in millimeters).

-------
Some ratifications to the air preheater were made. The intermediate
temperature zone element was changed to a cor*->sion resistant material
similar to th
-------
slipstream from both boilers prescrubbers that is treated will amount to
100 T/d total.
The total cost of the retrofit SCR system is V2 * 109 ($8 x 106) which
includes *1.7 * 109 ($6.8 x 106) paid to MH1 for the complete SCR system
including the economizer bypass and ¥0.3 -< 109 ($1.2 x 106) of the plants
own expense. This is roughly one-third of the cost of the TGD system.
The plant personnel indicated that the ductwork was a significant
portion of the cost. The costs of seme of the modifications are listed
in Table 29.
TABLE 29. COST BREAKDOWN OF SELECTED MODIFICATIONS
AT SHIM0X0SEKI POWER STATION
Item
Cost
¥
$
NH3 Injection
Facilities
40-50 x 106
1.6-2.0 x 10s
s
Air Preheater
Modi fications
30 x 106
1.2 x 10
175

-------
ELECTRIC POWER DEVELOPMENT COMPANY
Headquarters
Tokyo
March 10, 1980
Toraoio Kimura, Executive Director
Yasuyuki Nakabayashl, Assistant Ceneral Manager
Kunihiko Mouri, Thermal Power Engineer
David Mobley
Jumpel Ando
Cary Jones
Doug Maxwell
Isogo Power Station
Kanugawa Prefecture
March 26, 1980
Y. Shimizu, Assistant Ceneral Manager
T. Yamaki, Vice Chief of technical Division
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank
177

-------
Takehara Power Station
March 14, ]980
K. Fujiyama, Assistant General Manager
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank
179

-------
ELECTRIC POWER DEVELOPMENT COMPANY
The Electric Power Development Company (EPDC) is a semigovemmental
company that has evaluated several SCR pilot units on coal-fired boilers at
its Isogo, Takasago and Takehara power stations. Two of the systems tested
will be applied full-scale at EPDC's Takehara power station. Three separate
meetings were held; at the Isogo power station and at the Takehara power
station, and at the Tokyo headquarters. The results of .ill three meetings
are summarized here.
When EPDC began resting SCR processes at its Isogo, Takasago and
Takehara power stations, catalysts from five processes were tested. Following
the tests, EPDC selected the catalysts from Kawasaki Heavy Industries (KHI)
and Babcock-Hitachi (BH) as the best two as of late 1978. As a result of
these tests, the KHI and BH processes were selected for full-scale installa-
tions at EPDC's Takehara power station.
The EPDC Isogo power station consists of two, 265 MW, coal-fired utility
boilers. Currently at Isogo, there are two operating pilot units, IHI and HZ.
The IHI unit (Figure 52) treats 1000 Nm3/hr in each of four reactors - two
applied to particulate laden flue gas and two treating flue gas after a pilot-
scale, hot ESP. The reactors treating the ash laden gas are followed by
an air preheater and a baghouse. The reactors treating clean flue gas are
followed by an air preheater and a 1000 Nm3/hr limestone/gypsum FGD unit
capable of treating one-half of the flow from these reactors. The FGD unit
includes a Rotherruhle type regenerative air heater for gas/gas heat exchange,
a mist elininator and a wet ESP. This is a comprehensive arrangement of
pollution control equipment that EPDC and IHI hope will demonstrate the
effects of SCR on downstream equipment as well as screen catalysts and reac-
tors for use with both high and low dust flue gases. The low dust system has
operated for two years, however, the baghouse was only recently installed
(December 1979).
Preceding page blank
181

-------
00
hO
> 1 UC (. IH
(¦mm Unit
* f i «»11c* 11 /< r
Air
Pr«.hi ater
Hoc
fbp
\/
HIH
-HZ]-1
\ M»
V
Q~
1L» I jn
Bir
Hou*»c
*	 Ai*
Ale
Reactors Trtlicaters
-S"
|W'0« I i '
Km
H ))
f.is/C.as
Hent Exchanger
-To
M m k
-HZ]
[~Vct 1
—Hmp |—1
Hist
Fl lfnin.it or
Figure 52.
Flow diagram of IHI pilot facility at EPDC's Isogo Power Station.

-------
A Hitachi Zosen pilot unit is also operating at Isogo and a flow sheet
of this facility is shown in Tigure 53. There are three reactors; one treating
a low dust flue gas and two treating high dust flue gas. The unit has been
in operation for 4 vesrs and is currently testing the 600 series catalyst. At
an NHi:N0x mole ratio of 0.9, they experience 80 percent removal and NH3 emis-
sions of 5 ppm. With both catalysts (500 series and 600 series) they have had
no erosion or plugging problems and have not required soot blowing. In both
cases the catalysts consisted of a triangular honeycomb arrangement (Figure 54)
with the parallel plates constructed at 12 mm intervals. Test results are not
yet available for this unit.
During the facility tour the gas/gas heat exchanger was being disassembled
in order to clean corrosion frcm the elements. Plant personnel telt that the
corrosion was due to mist from the FGD when the wet ESP was not operating;
however, it appears to be possible for the corrosion to be caused by ammonium
sulfates froi' the SCR system. The heat exchanger has been in operation for
two years and is cleaned once every six -lonths. It is blown with steam for
m-situ cleaning.
Outlet NC^ and NH3 are monitorsd by an instrument manufactured by Anntsu
Electric. N'H 3 is analyzed by UV absorption and NH^HSOi, deposits are avoided
by maintaining the lines at temperatures above 300°C. The instrument apparently
gives good agreement with wet methods.
The pilot facilities at Isogo are two of the five joint ventures between
EPDC and process vendors that ha
-------
To
ESP
From
Economizer
Heaters
Soot
Blower
Reactors
Figure 53. Flowsheet of Hitachi-Zosen Pilot Plant
at Isogo Power Station.

-------
Figure 54. Structure cf Hitachi Zosen Noxnon 500 and 600 catalyst.
185

-------
Cold
ESP
Boiler
FCD
ID
Fan
Pre-
Heatcr
Air
Stack
Existing Arrangement
•Existing
	—Now
oo
p — —	1
\ F° I
1 Fan ,
I	J
Air
I	1
i Hot i
ESP \
I	I
I-
To
Stack
SCR
i	I
I	I
FGD
Boiler
Modified Arrangement
Figure 55. Equipment modifications made at EPDC Takehara Power Station
for SCR system installation.

-------
(1)
(1)	Boiler
(2)	Hot ESP
(3)	SCR Reactor
(&) Air Preheater
(5)	Limestone/Gypsum FCD
(6)	Stack
(7)	KH3 Storage/
Vaporization
(6)
(7)
100m
	I
Scale - 1
1000
Plant Layout
Figure 56.
Construction Area
SCR system (under construction; at EPDC Takehara power station.
187

-------
The hot ESP was added to prevent contamination of the ash by ammonium
sulfates. In Japan, ash is either sold as an aggregate or disposed of by
landfill in the sea. Tly ash contaminated with ammonium sulfates that is
used as landfill in the sea can cause a red tide and kill fish. On one
occasion, wash water from the />2 boiler ESP, which utilizes NH3 injection
for ash conditioning, was discharged to the fly ash pond at the beach and
killed the fish in the pond. The Takehara station is particularly sensitive
to this problem since it is located on the inland sea which experiences a
much lower water turnover rate than coastal seas. EPDC also feels that
ammonium sulfate contamination may affect the saleability of the fly ash and
has conducted tests to remove NH3 from the ash.
The new air preheater will be installed at the Takehara SCR facility to
reduce the potential for NFUHSOi* fouling. Since the particulates will be
removed upstream, the air preheater will not receive the benefit of sanci blast-
ing effect of these particulates. As a result, it is felt that there may be
problems with air preheater pluggage if the: existing, conventional air pre-
heater is used.
The new air preheater is a new design developed by Gadelius K.K. The
new design consists of two sets of heat transfer elements as shown in Figure 57.
This air preheater is discussed in more detail in the meeting report on Gadelius.
fwo new fans are also necessary at Takehara. The forced draft fan is
necessary because with the new equipment layout, the air preheater is located
farther from the boiler and, consequently, there is a larger draft loss in
the air duct. The induced draft fan is necessary to overcome the draft loss
created by the SCR reactor and associated ducting. The existing and new fans
are compared in Table 30.
About 50 kW are used by the NH3 vaporization and delivery system. This
makes the net change in energy consumption required by the SCR system equal
188

-------
Intermediate
2.5 mm
fold
6 mm
Conventional Arrangement of Heat Transfer Elements
Soot
Blower
Combined
Intermediate and Cold
3.5 mm
Modified Design by Gadelius K. K.
Figure 57. Comparison of conventional and modiTied air preheater design

-------
to 3500 kW or 1.4 percent of the power output of the boiler. The plart
personnel felt that a better equipment layout would reduce energy usage to
1450 kW.
TABLE 30. COMPARISON OF EXISTINC AND NEW FANS AT TAKEHARA POWER STATION"
kW
Fan	Existing	New
Forced Draft (air)	630 x 2	13j0 x 2
Induced Draft (gas)	1350 x 2	2350 x 2
An NHj storage capacity of 75 tons is planned to meet the req.jiremcn-.n
for S0X control of both the //I and 02 boilers. N0X control equipment will b:
installed on the It2 boiler (350 MW) in 3 to 4 years when it switches to coal.
The two units are expected to require a total of 5 tons/day of NH3.
NH3 concentrations in the outlet stream of the SCR reactor are expects
to be very low. NH3 emissions will be further reduced by the downstream FGD
systes; however, this will requiie the installation of an activated sludge
system to control the nitrogen concentration of the wastewater to <10 ppm a.>
required by regulation. Currentiy, the FGD system purges 15 tons/hour of
wastewater which is treated in an existing treatment system. The activated
sludge treatment sysLem will apparently operate in series with the existins
treatment system.
The total cost of applying SCR to the #1 boiler at Takehara is 8 x 109
yen which includes the hot-side ESP's, air preheaters, fans and the waste
treatment system. This cost is relatively high because of rhe additonal
equipment necessitated by retrofitting the SCR system to an existing boiler.
This is the investment costs; annualized rosts are not yet available.
Also at Takehara, a test program is underway on a dry, N0X/S02 removal
process. The system is supplied by Sumitomo Heavy Industries and utilizes
a moving bed of activated carbon. EPDC's primary reason for developing a dry
190

-------
SO2 control process is to save water and to have a process that is more
compatible with SCR.
The test unit removes 90 to 95 percent of the SOz and 30 percent of the
N'Ox fron the flue gas. The NOx removal aspect of the process is similar to
typical SCR processes; however, an St^iN'Ox mole ratio of 0.7 is required to
get the 30 percent removal. l.Tiile the NHsrN'Ov reaction may be equimolar,
the excess NH3 is adsorbed by the thar as a sulfate. This reduces carbon
consumption during thermal regeneration. Some details on the process are
sunma'rized in Table 31.
TA3LE 31. OPERATING DATA FROM SUMITOMO ACTIVATED CARSON PROCESS
AT THE EPDC TAXEHARA POWER STATION
Operating time on test unit
SO2 removal
N0X removal
NHj:N0x mole ratio
Inlet SO2 concentration
Inlet N"Ox concentration
Operating temperature
Carbon consumption, typical
Regeneration
SO2 workup
Catalyst dimensions
7000 hr
90-95Z
30%
0.7
1300 ppni
320 ppr
no-isc c
0.5% of total volume/day
Heatinr. with gas that is low in O2
Resox
About 1/4 cm. dia. x 1 cm. langth
It is necessary to collect particulates (ash and catalyst fines) re-
sulting from the catalyst screening step and both an ESP and baghouse are
being evaluated. With the relatively low NOx removal of this process a
separate SCK system aav be required. The p'3 boiler will have SCR and wet
FGD but the vendors have not yet been selected.
191

-------
In addition to the test units and full-scale applications, EPPC discussed
the general aspects of SCR applied to coal-fired boilers. They feel that air
pre'ieacer pluggage by NHuHSOi* requires changes in air preheater operation,
material and design. Past experience indicates that the pluggage occurs at
the interface between the intermediate and cold temperature zones because soot
blowing is not effective in this area. They hepe that the modified element
design will allow effective soot blowing throughout the air preheater. In
addition, both hot and cold sides use improved, note effective soot blowers.
The ne» design of air preheater is heavier and will cost roughly twice as much
as conventional units. ^f necessary, this air preheater will be washed during
the annual outage and the water treated with the boiler tube washwater.
EPDC feels that SCR is very reliable, 99+ percent, and that redundancy
in reactors is not necessary. The systems will require no operators or
additional maintenance personnel. At Takehara, the catalyst manufacturer will
supply the labor necessary to change out the catalyst.
The investment costs for SCR applied to a new boiler are ¥7000/kW ($28.5/
kK) and operating costs are ¥0.6/kWh (2.5 mills/kWh). These costs are based
on one year catalyst life, seven years depreciation, 8 percent interest and
100 percent financing.
In terms of NOx removal, EPDC feels that 80 percent is the most reasonable
target. Ren-ovals of 90 percent are technically feasible, but the conversion of
SO2 to SO3 will increase as the catalyst volume is increased. Typical values
are shown in Table 32.
TABLE 32. CONVERSION OF S02 TO SO3 BY SCR CATALYST
NOx Removal
Percent
Percent of Inlet SO2
Converted to SOa
80
1
90
2
192

-------
Catalyst manufacturers are working on this problem and it is anticipated
that for 80 percent NOx removal, SO; conversions will be <0.5 percent by the
time that SCR is applied to the Takehara r'3 boiler.
EPDC is examining continuous NH3 measurement techniques, but currently
use a manual method. NH3 analyzers have not worked in the past, but they
are trying a model by Anritsu Electric at their test units. Fxperience has
shown that SO2 concentrations of >500 ppn cause reliability problems. For
NOvr measurement, they use eheffliluminescence.
Control of the SCR system is typically done b-/ using a feedforward sig-
nal consisting of the product of inlet KO^ concentration and flue gas flow
rate. This is do.ie by measuring boiier load. NHj is controlled with a
rotameter arrangement.
19 J

-------
FUJI OIL
Sodegaura Refinery
Chiba Prefecture
March 21, 1980
Yoshihiro Hirotani, General Manager of Refining Department
Y. Miyabuchi, Vice Chief
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank
195

-------
FUJI OIL
Fuji Oil's Sodegaura Refinery is located in the Chiba prefecture and
must comply with local regulations established by the prefectural government.
These regulations have necessitated NOx controls on four boilers; two have
combustion modifications and two h3\/e catalytic de-NOx units. The boilers
are listed in Table 33.
The i?7 boiler utilizes an MHI SCR system located between the boiler and
the economizer as shown in Figure 58. The boiler was built with the SCR
designed as an integral unit. Therefore, the SCR reactor exists as extended
duct work between the boiler and economizer. This location is necessary in
order to treat gas in the 330 to 370°C temperature range. The SCR system
has operated for 2.25 yeais with no catalyst deterioration and 1 to 2 more
years are expected. The catalyst is a coated, ceramic base catalyst manu-
factured by NCK. Several different formulations are Installed in thp
reactor as part of an MHI test program.
TABLE 33. NOx CONTROL ON INDUSTRIAL BOILERS AT FUJI OIL SODEGAURA REFINERY
Boiler No.
Fuel
Steam
Rate, T/h
de-NOx
Equipment
NOx,
Inlet
ppn
Outlet
5
Fuel Oil
160
Flue Gas Recycle
Staged Combustion
150
80
6
Fuel Oil
160
Flue Gas Recycle
Staged Combustion
150
80
7
Fuel Oil
200
SCR (MHI)
115
15
CO Boiler
CO + Fuel Oil
45
SCR {JGC)
180
15
Preceding page blank
197

-------
WJJ*
No. 7 Boiler
oil
180°C
NH
to stack
Forced	SCR
Draft	Boiler	Reactor
Fan
Econo-	ESP
mizer
Figure 58. MH1 SCR system applied to industrial boiler at Fuji Oil Refinery.
198

-------
Some plugging was experiencad in the economizer after six months opera-
tion necessitating a water wash. During this period the economizer AP in-
creased twelvefold. The problem stemmed from some pluggage in the catalyst
bed which caused the concentration of unreacted NH3 in the reactor outlet gas
to increase. This increased the reaction temperature for WUHSOi, formation
and allowed NH^HSOi, to form on the economizer tubes. Additional soot blowers
were installed in the reactors and the soot blowing procedure for the reactor
and economizer was changed. This allows the boiler to operate one year before
the economizer must be cleaned. Soot blowing is done carefully to avoid
contacting the catalyst with water droplets.
Plant personnel feel that the catalyst face plugging may have been due
to a silica combustion additive used to increase the melting point of the ash.
Silica was used since MHI advised that the alternative compound, composed of
magnesium and lime, could harm the catalyst. Catalyst face plugging has not
occurred since the use of the additive was stopped and the additional soot
blower was installed.
MHI has apreed to take back the catalyst for disposal when it is spent.
Water from the economizer wash is disposed of by neutralizing the pH with
NaOH and routing it to the existing plant water treatment system. The treat-
ment system consists of oil separation, neutralization ponds, filtration and
activated carbon adsorption.
The JGC SCR systen applied to the CO biler is shown in Figure 59. This
unit was operated for six months and shutdown when the refinery dropped back
to 70 percent capacity and the NOx offset from the CO boiler was not required.
Plant personnel indicated that this unit was less energy efficient due to the
high reactor AP and the inline heater.
Ihe plant personnel also indicated that both units have been trouble-
free and reliable. One nan is sufficient to operate both the boiler and the
SCR system.
199

-------
CO Boiler
Q
Forced	Boiler
Draft
Fan
mizer	Burner Draft Parallel
Fan	Passage
Reactor
200-230°C
Econo-	Inline
to stack
Figure 59. JGC SCR system applied to CO boiler at Fuji Oil Refinery.
200

-------
HOKKAIDO ELECTRIC
Headquc'r ters
Tokyo
March 21, 1980
Mr. Yoshida, Manager of Technical Division
Hideji Imanishi, Deputy General Manager of Hokkaido Plant
Y. Katsuki
Hideo Kojima, Hitachi, Ltd, Manager of Environmental Equipment Dept.
David Mobley
Jun.oei Ando
Gary Jones
Doug Maxwell
201

-------
HOKKAIDO ELECTRIC
Hokkaido Electric is currently constructing a coal-fired power plant
near the city of Tomakomai on Japan's northern island of Hokkaido. The 350
MW boiler will burn low sulfur (0.3 percent) coal containing 20 percent ash.
One quarter of the 106Nro3/hr flue gas stream will be treated by SCR for 80
percent NOx removal. The flue gas will also be treated by a hot-side ESP and
a limestone/gypsum FCD. The FCD will treat half of the total flue gas. A
plot plan of the plant is shown in Tigure 60.
The hot-side ESP is used because years ago, when the project was planned,
SCR had not been proven with high particulate loadings. Also, the low sulfur
coal makes a hot-side ESP desirable.
The boiler, FCD and SCR will be supplied by Babcock-Hitachi. The boiler
will use dual register burners and flue gas recirculation to limit NOx forma-
tion in the boiler to 200 ppra. The SCR will reduce the total NOy measured at
the stack to <170 ppm which meets the emission regulations.
NH3 emissions are expected to be very low since on]y one-fourth of the
gas is treated by SCR. Consequently, no special provisions for air preheater
dealing or soot blowing are planned. Some NH3 may be removed by the FGD
system but the overall concentration is expscted to be small. NH3 will be
injected at a mole ratio of 0.83 and one weeks storage of NH3 is provided.
Of the 170,000 T/yr of ash produced, 18 to 24 percent will be sold to the
cement industry and the remainder used for land reclaimation. Spent catalyst
will be disposed of by Babcock-Hitachi. The catalyst to be installed at
Hokkaido is guaranc^cj for one year.
An economizer bypass will be installed for temperature control since
it was considered necessary when the plant was designed four yeaTS ago.
However, due to the fact that when the boiler load drops so do both NOx con-
centration and space velocity, the thinking has cnanged. Plant personnel
Preceding page blank
203

-------
Boiler
OO o
Hot ESP
SCR
FGD
ID Fan
Figure 60. Plot view of Hokkaido Electric Coal-Fired Power Station .
204

-------
currently feel that the bypass will be unnecessary to maintain the 300°C
minimum design temperature.
The number of operating personnel has not been decided. Potentially,
there will be a separate control room with 2 to 3 operators to handle envi-
ronmental control (ESP, SCR, PCD), water treatment and c«-al handling. The
main boiler control room will have 5 to 6 operators and a total of 100 people
will be required to staff the whole plant. It is estimated that a catalyst
change will take about one week following cool down during the one month
scheduled boiler outage.
A second 600 MW boiler is planned, but not yet authorized. If autho-
rized, Hokkaido plans to re-evaluate several of the design criteria including
hot vs. cold-side ESP's.
205

-------
TOA NENrtYO KOGYO K.K.
Kawasaki Refinery
Kanagawa Prefecture
March 24. 1980
Yukio Yasumura
Toshihide Mori, Section Head
S. Yamada, Manager
Gary Jones
Doug Maxwell
Preceding page blank
207

-------
TOA NEN'RYO KOCYO
Toa Nenryo (Tonen) utilizes an Exxon Thermal de-NOx system on 1 pipe
still at their Kawasaki refinery. The unit is not ope_ated continuously since
the NOx emissions are under the 0.38g/mcal limit. Howjver, when there is a
smog alert issued by the Kanagawa air monitoring center, the thermal de-NOx
system is operated to effect a 20-50 percent N'Ox reduction. Last >ear, this
resulted in 10 days of operating time.
Injection of NH3 and reducing gas is controlled by the outlet IIH3 concen-
tration which is maintained at <20 ppm. The reducing gas is a refinery stream
consisting of H? and hydrocarbons. Use of the reducing gas is seldom necessary
and is injected only wl.en the gas temperature drops causing the outlet NH3 to
r.^se. When injected, its flow rate is roughly 20 percent of the MH-j flow rate.
Continuous NHj analysis with chemilJtnlnescent analyzers has been a prob-
lem and Tonen has experienced problems similar to those in the US. A maior
problem is that the instrument will indicate NH3 concentrations that are
one-half of the actual concentration. Tonen claims to have solved this prob-
lem and applied for a patent. No details are currently available except that
temperature control is very important. The patent disclosure is expected in
late 1980. A brief visual inspection of the system showed a rather large
sample gas treatment station located adjacent to the outlet duct and upstream
of the Shimadzu NHA-302 analyzer. This system rtporcedly has a measurement
accuracy of ± 10 percent.
Tonen also has installed the thermal de-NOv equipment to two utility
boilers (468,000 Nm3/hr each) 3nd a CO boiler (34,000 Nn3/hr) in other area.,
of the plant. Operation of these units is similar to that installed in the
pipe furnace.
The de-NOx system applied to the pipe still was checked for the oresence
of HCN; however, it was not defected when reducing gas is not irject^d.
Preceding page blank
209

-------
TOKYO ELECTRIC
Headquarters
Tokyo
March 2o, 1980
Tetsao Takeuchi, Deputy Manager of Thermal Power Dept.
Masahara Nagaoka, Deputy Manager of Environmental Protection Dept.
Hitoshi Tominaga
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
211
Preceding page blank

-------
TOKYO ELECTRIC, TOKYO HEADQUARTERS
Tokyo Electric is currently involved in a cooperative effort with
Tohoku Electric Power Company, Joban Joint Power Company and Mitsubishi
Heavy Industries to demonstrate an integrated flue gas treatment system
treating coal-fired flue gas on pilot scale equipment. The participants
feel that demonstrating a system in which particulates, NOx and SOx are re-
moved from the same flue gas stream will provide information more meaningful
than just testing a single step. Accordingly, a set of pilot scale equipment
including both hot and cold-side ESP's, a baghouse, air preheaters, an SCR
system, a FCD system, a gas/gas heater, a wet ESP and three waste water
treatment systems has been assembled at Joban Power Company's Nakoso Power
Station. The equipment is arranged to provide flue gas with both high and
low particulate loadings to the SCR reactor and air preheater as shown in
Figure 61.
As of Spring 1980, 5000 hours of testing had been completed using high
fly ash loadings in addition to 4000 hours of testing with low fly ash load-
ings. Tokyo Electric is prepared to release most of the test results although
some results must wait until the final tests are completed.
The SCR systems operate at a NH3:N0x m^le ratio of 0.85 and can achieve
up to 8b percent NOx removal, although 80 percent is the guarantee. Both
honeycomb and plate type catalysts have been tested in both vertical and
horizontal configurations. Initially the air preheaters were both of conven-
tional design; however, it was necessary to modify the air preheater used
with low ash test unit and this is described in the discussion of the results
of low ash testing.
Tests with High Ash Loadings
Testing with full particulate loadings has shown catalyst erosion and
plugging proolems do not occur. To prevent erosion, a uunmy space in the
Preceding page blank

-------
2000 Sra*/h
Air
Prchcater
Cold
LSP
SCR
Wt t
fcSP
Hot
LSP
SCR
Treotcd
gas
Figure 61. Layout of pilot scale flue gas treatment
system at Joban Power Company's Nakoso Station

-------
shape of catalyst elements is installed at the reactor inlet and has been
very effective. The catalyst was inspected frequently and no catalyst
plugging occurred even though no soot blowing was performed. Consistent NOx
removals of 80 percent were achieved.
Air preheater pluggage by NtUHSOi, did not occur anywhere and this is
attributed to a sand blasting effect of the fly ash. During testing, the air
preheater was soot-blown once per day.
A double effect particulate control system is used with the high ash
load SCR equipment which includes a c&ld-side ESP followed by a new design
MHI prescrubber for particulate removal. Tests in which the ESP outlet
concentration ranged from 100 to 300 mg/N'm3 showed that the particulate
concentration at the outlet of the prescrubber ranged from 15 to 35 mg/Nm3.
This second ash removal step is considered necessary in Japan because it is
likely that, at any coal-fired power station, a wide variety of foreign coal
types will be burned including low sulfur coals. Ash from low sulfur coals
does not respond well to ESP treatment and therefore, the particulate scrubber
allows a consistent level of particulate control without regard to coal type.
Tests with Low Ash Loading
These tests involved use of a hot-side ESP which is an alternative to
the two step technique described above. The performance of a hot ESP is not
affected by the sulfur content of the coal. This is a more difficult situ-
ation for the SCR system since the ash remaining in the flue gas is finer,
more sticky and more prone to cause plugging problems. Tests showed that
catalyst plugging tended to occur at the inlet face. The plugging is worse
with small pitch catalyst shapes, low gas velocities and horizontal gas flow
reactors. The plugging can be controlled by careful section of catalyst
shape, gas velocity, reactor configuration and soot blowing frequency. On
the pilot unit, soot blowing of the catalyst bed is necessary once every four
days, however, specific data on the other design values were not available at
215

-------
the time o£ the meeting. Once the appropriate design and operating param-
eters were established, steady state operation of the SCR system was
achieved.
The low ash condition caused considerable plugging in the intermediate
and low temperature zones of the air preheater. To combat this problem a
specially shaped element has been installed in which the intermediate and
low temperature elements have been combined into a single element. The
new element, combined with soot blowing three times per day, has apparently
minimized the plugging problem.
Other Tests
The FGD system ip a limestone/gypsum process and has operated with
high reliability and achieves 90 percent removal ol SO2. The wastewater treat-
ment facility consists of a clarifier, an activated sludge unit for nitrogen
removal and a COD treatment process; presumably an aerator. These units
have successfully reduced suspended solids, nitrogen and COD such that
Tokyo Electric and Lhe other participants feel that these processes can
be successfully applied to full scale systems. The activated sludge sys-
tem is not necessary yet, but will be in the future if NH3 omissions in
waste water are regulated as is anticipated m japan.
Tne tests with the baghouse are still underway and results are not
available. They have experienced some operational problems which may or
may not be due to NfUHSOu. Further testing should reveal the source of
the problems which involve bag plugging and high ptessure drops. The re-
sults of the baghouse tests will be available by the end of September, 1980.
The personnel at Tokyo Electric now feel that SCR NO^ removal systems
can be successfully applied to coal-fired boilers without seriously impact-
ing other flue gas treatment equipment. It is important that the overall
system be considered duiing the design phase in order to minimize adverse
216

-------
impacts. Technically, the most desirable system is to install the SCR
reactor and air preheatcr upstream of the ESP in order to benefit from the
abrasive effect of the fly ash. Catalyst erosion can be eliminated by in-
stalling a dummy layer of material shaped ljke the catalyst at the inlet
of the reactor. A vertical downflow reactor is preferred with all coal
applications and soot blowing is probably not required, but may be installed
as a safety measure.
In Japan, where the coal type is variable, hot-side ESP's or baghouses
may be necessary. Hot ESP's will require most emphasis on the design of the
SCR system and air preheater in order to avoid plugging problems. Baghouse
cechnology for power plant applications is in an early stage of development
in Japan and it is too early to determine if bagnouset can be used downstream
of SCR systems. Tokyo Electric hopes that further testing will answer the
remaining questions.
The results of this integrated test program are important to the partic-
ipant companies since all are planning to install coal-fired boilers in the
future as shown in Table 34,
For full scale SCR systems on coal-fired units Tokyo Electric expects
the cost to be ¥5000 to 7000/kW ($20-28/kW) for 80 percent removal, half of
which is the catalyst cost. The catalyst life guarantee will be one year.
Although 90 percent removal can be achieved, they prefer 80 percent since
significantly less NH3 is necessary and reactor outlet NH3 emissions are
reduced, thereby minimizing problems with downstream equipment. Their plans
include using combustion modifications in addition to SCR for NOx control.
217

-------
TABLE 34. COAL-FIPED BOILERS PLANNED BY COMPANIES PARTICIPATING
IN PILOT UNIT TEST AT NAKOSO POWER STATION
Company
Boiler Size, >IW
Fuel

Start-up
Tokyo Electric
1000
Coal

1988
Tokyo Electric
1000
Coal

1988
Tohoku Electric
600
Coal

1985
Tohoku Electric
600
Coal

1985
Tohoku Electric
600
Coal

1985
Joban Power
600
Coal/Oil
(10/90)
1983-4
Joban Power
600
Coal/Oil
(10/90)
1983-4
218

-------
APPENDIX C
GOVERNMENTAL AGENCIES AND INDUSTRY GROUPS
219

-------
CENTRAL RESEARCH INSTITUTE OF THE ELECTRIC POW-R INDUSTRY
Headquarters
Tokyo
March 27, 1980
Yoshirai Ishihara, Manager of Environmental Chemistry Dept.
Kenkichi Masui, Research Advisor
Teruo Yatabe, Senior Research Engineer
Tsuneo Nakanishi, Deputy Manager of Siting and Environment Dept.,
Federation of Electric Power Companies
Mamoru Masuda, Chief of Siting and Environment Dept.
Federation of Electric Power Companies
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank
221

-------
CENTRAL RESEARCH INSTITUTE OF THE ELECTRIC POWER INDUSTRY
The Central Research Institute oE the Electric Power Industry (CRIEPI)
is a general research organization representing Japan's electric power indus-
try. Control of nitrogen oxides is an important issue to CRIEPI and TEPC
since increased utilization of coal by the utility industry is anticipated.
By 1995 it is anticipated that 12 to 13 percent of the country's total elec-
tric power will be produced from coal as shown in Table 35.
TABLE 35. TARGETS FOR COAL UTILIZATION FOR ELECTRIC POWER
GENERATION IN JAPAN (OPERATING UNITS)
Year
Coal-Fired Boiler
Capacity, MW
Percent of Total Electric
Power Generated by Coal
1978
4410
3.7
1985
10,000
5.6
1990
22,000 - 23,000
9.5 - 10
1995
33,000 - 36,000
12 - 13
All existing utility boilers have been modified to include combustion
modifications and both FEPC and CRIEPI feel that SCR has been successfully
demonstrated on gas- and oil-fired boilers. With some development they feel
that SCR technology will be equally useful on coal-fired boilers and feel that
most all of the coal-fired boilers will have SCR for NOx control. It is
anticipated that the requirement will be a result of local agreements and not
a result of regulations established by the Japan Environment Agency which, at
the present time, only reqjire combustion modifications as shown in Table 36.
SCR units will operate at 80 percent NOx control rather than 90 percent since
80 percent is felt to be a more economical and practical control level.
The investment cost of SCR is about ¥3000 to 4000/kW ($12 to 16/kW) for
oil-fired applications and about ¥6000 to 7000/kW ($24 to 28/kW) for coal-
fired applications which represents about 1.5 percent of the cost of a new
boiler for the oil-fired case.
Preceding page blank

-------
TABLE 36. JAPAN ENVIRONMENT AGENCY STANDARDS FOR BOILER NOx EMISSIONS


Existing
Newly Installed


Flue Gas
Boilers
Boilers
02
Boilers
(10 3NM3/h)
(ppm)
(ppm)
(%)
Gas Firing
0ver-500
130
60
5

100-500
130
100


40-100
130
100


0
1
o
150
130

Solid Firing
Over-100
480
400
6

40-100
600
400


10-40
600
400

Liquid Firing
Over-1000
180

4

500-1000
180
130


100-500
190
150


40-100
190
150

224

-------
The reliability of SCR systems applied to coal-fired boilers is expected
to be similar to that for oil-fired boilers; however, more catalyst may be
required and catalyst life may be lower. NH3 emissions are not a major con-
cern since NH3 is currently used on oil-fired boilers for ash conditioning
with no r*"oblems. High concentrations of NH3 in waste water may require
activated kludge treatment or stripping if eutropliication is a local prob-
lem. Catalyst disposal will be provided by the catalyst manufacturers and
all current SCR users in the utility sector have contracts to this effect.
CRIEFI has also done research on stpek gas analysis methods including
those for N0x and NH3. The principle methods used in Japan for N0X measure-
ment are shown in Table 37. In power plants, chemiluminescense is the most
widely used method with 75 percent of the applications using this method.
NDIR and ITDUV are used in 11 and 14 percent of the applications, respectively.
TABLE 37. MEASUREMENT METHODS FOR NO*
Type of Analysis
Principle
Measurement Range (ppm)
Chemical
Zn-Reduction-NEDA
10M.000

Phenol-Disulfonic-Acid
10V300

Ion Electrode

Continuous
Chemiluminescence
0^25 — OM.OOO

Nondispersive IR
0-VjO	CK1000

Nondispersivo UV
OV.0	OMOOO

Electrochemical
0^200 	OVL000
For measurement of NH3 in addition to other components" "in the duct or
stack, direct inview gas analysis (DIGA) is preferred. In this metnod a
light source is placed on one side of the stack and a detector on the other.
A large tube or pipe travels between the light and detector to prevent inter-
ference. Figure 62 shows a schematic of the techniques. This method offers
quick response, 5 seconds, and can be used with gas temperatures of 3 50 to
225

-------
Light
Source
UV or I.
<({	 several meters
Duct or Stack
 d Pipe across duct 3D
"lotted to
allow gas flow
Detector
Method	Measurable Gas Components
ND1R	CO, C02, H20
Nll'.V	S02, NO, NH3
Figure 62. Schematic of DIGA technique for
continuous in-duct gas analysis.
226

-------
400°C thereby making it useful for NH3 control at the SCR reactor outlet.
Also, by making a direct measurement, sampling error is eliminated. How-
ever, the accuracy of the NH3 measurement is affected by the presence of
SO2 which causes an artificially high result. Reportedly, some modifications
are necessary to counterbalance this interference. The ash from coal-fired
boilers will also affect the total absorbance, however, this can be accounted
for by measuring the specific wavelengths absorbed and calibrating the in-
strument accordl -gly. Tokyo Electric is supposedly using this technique
for NH3 and obtains ± 10 percent accuracy at 5 ppra NH3 on a 100 ppm full scale
instrument. CRIEPT representatives indicate that the equipment is produced
in the US by Environmental Data Corporation which is a subsidary of Thermo-
electron Corporation.
CRIEPI has also examined the chemiluminesc.ent methods which use an
NH3 converter and split sample stream to measure NH3 by difference. They
have found that errors occur due to incomplete NH3 conversion and feel that
the most advanced converter systems are those manufactured by Fuji Electric,
Shimadzu, and perhaps Horiba.
227

-------
JAPAN ENVIRONMENT ACEN'CY
Headquarters
Tokyo
March 19 v L980
Taka Hiraishi, Deputy Diiector - International Affairs
Ijamu Yokota, Deputy Director - Planning
Shigehisa Hidehira, Deputy Director - Air Pollution Control
David Mob ley
Cary Jones
Doug Maxwell
Preceding page blank
229

-------
JAPAN ENVIRONMENT AGENCY
The Japan tnvironerant Agency is similar in regulatory Function to tlie
US t^PA in that it sets ambient air quality standards that apply throughout
the country. The agency also sets emission standards. Local governments
have the option of adopting the national emission standards or establishing
their own stricter emission standards, except for SOx emissions.
Environmental emission standards for NOx were first established in 1973
and have been revised three times since, most recently in August of 1979.
The most recent revision covers 95 percent of the NOx emission sources in the
country. 144,000 source*; are registered. The ambient air quality standards
cover three levels of ambient NO2 concentration as shown m Table 38.
TABLE 38. RANGE OF AMBIENT NOx CONCENTRATIONS IN JAPAN
Ambient NOx	Number of Areas Percent of Total Percent of Total
Concentration (ppm) in this Range	Population	Land Area
>0.06	6	22.3	1.04
0.04-0.06	18	19.2	3.29
<0.04	Balance	58.5	95.67
The Environment Agency has set 1985 as the target date for areas with
ambient NO^c concentrations >0.06 ppm to comply with the regulations by reduc-
ing the concentration to at least 0.06. A policy of nondegradation is
established for areas in the 0.04 to 0.06 ppm range. The policy allows no
change in the ambient concentration if there is no industrialization of the
area. If there is industrialization, a significant change is allowed.
"Significant" is not defined. Areas with concentration <0.04 are allowed to
increase the ambient concentration to 0.04.
The Environment Agency also sets emission limits for both mobile and
stationary sources. A complete list of emission standards for new and existing
231
Preceding page blank

-------
sources is shovn in Table 39. These national standards are based on combus-
tion modifications for NOx control and do not require application of SCR or
thermal de-NOx. For new, large utility boilers, an environmental assessment
is required and it is anticipated that this nay result in catalytic dt>-NOx
applications.
While the national emission standards require only combustion modifica-
tions, individual sources have to negotiate their own standard with local
authorities. These locally set standards have resulted in a rapid increase
in SCR installations as shown in Figure 63. By 1978, there were 58 SCR
facilities and now there are 104 de-NOx facilities including both SCR and
thermal de-NOx.
The Environment Agency indicated that Japanese gas- and oil-fired
installations operate at 80 percent NOx removal rather than 90 percent because
this technology is what is commercially available and guaranteed by the pro-
cess vendois. The Agency is more concerned with overall N0X concentrations
and environmental qualify than with individual control efficiencies.
232

-------
TABLE 39, EMISSION STANDARDS FOR NO IN JAPAN AS
OF AUGUST 2, 1979	X
(Source: Japan Environmental Summary, Sept., 1979)

r»r« *





««« IM«I



1 •-
1 s»'r»

V v.. *< » M

«« 10 ">)
Vn « n
•ftrt
N«n 19 *7)
t.w 17 *71
(l(rt
Im II *77
brlMi
Ati » n
•f*t
A«r fO 7f
Date of Construction

Cm
O-*
S«"» 1*- M
l«JU - ll«l

ito
Id)
00
*v




~ ft • IfWl
j


no


l«0




IU - 4ft

1)0

| 1 4





5 • 10
Irx U«it S

IW



S«(*
•*4 C >1
100 ,*4 «k. •«

tifl Apr » to
kfO











*»o

419







«0 - iUO
*
till Awf $ •'
im
Uom 10 I1
4*0














400


i

10 - 40

t¦>*
I'O j*J »W»«

iril Apr '0 lb
»>0
rm Mj* t KU
IIO


IM

l»




IOC - >00
4
Id) Ap» »•>
1 k>
'/v« 1 "»ft
ItO

110
ISO





*n - i<«-»

	"0









<0 - 40

1 »Ui>
>0








i - 10

1 - Vpi * 7*i
linn  7
-1




kr« (Ki« J

. -Vp« » >».
If m AM
0 M
'JO


IBO


V«ir

|W j-vJ «

1
rft M,\
I in

>*0




11 - ioo
IS
	
U,.
1 >0 1^0



I.tTi'"*

10
¦
> m A >i
io • i smj


2O0

Date of Enforcement
rr

14
«
um A«i
to f :*o


?:o
) and
¦u<
f- •*.«

1)
f

10 f no


ino

Emission Standard
««1J
twn


II
tll>n Aki
10


no

Wfljl ffi
I'lO ltd j>vf


TOO

00



¦ It.
«>M t»pf»
4" -
II
(rem M«» | *%A
yo

SO





10 - 4ft

>
no








s - to

N««" Mj. 1
>0
Ml

ISO




k.» «*,• j

Irom A q
o »:
luO

uo


«f|jl Iwilpf
tOO ii4 ibo*«
II
«lE Ap* *0 to
2.0
fiom M»» 1 ¦>.»
I»0
utt Apt jo m
MO
lltxa Ui» 1 10
1*0
too




40 - 100
I.J Apt )0 »0
}20
from Mi | l'i
1 >0
1 n Ar» » to
¦>00
r#o«* mj* i \o
XiO
l>0

IK»




10 - 40

t.n A„t « i:
>00
lrnf"» Aw* 10 12
no







J - 10

».««¦ u,. i
»0
170

ISO




W»« lfl» J

'•rm Awt
0 «*
mo

110


Ktr,
hr«<
lews
<~*
K»
|W* i«J iVnt
40 - 100

no JO to
)i0
llOA Mj» > *>o
1 0
170
too




10 - 4"

m |

ISO
IN>




J - 10

' €>•" Mj. 1
•to
tto

ISO




»r >• *

ir«*. to i;
100

ltd


fin
t+1
l«K> t-e iK. •« 1
I'l^l AfH
11
410
IM>





*«"**
|l\1

1
u-n ^ r<
1 11 ««'*


ISO



7"-"

15


ll.HV Auy. to
1*

4 50


t-%-
« i«*t i

tt


i «>»>0

}X>


%i«t

4«> »«J ,S .«
l)
l.«« A«f
n ¦»
KO


jso


•k ~*ina
k. i«i» «t>


fx fi Am 10
i: K1o

150


V.I K ».*

CH



100




l«rJ
,









>r>
|«V ,*< ,VI

!..«»«» |
to
1)0
no






*»« iiuk ion

'
iwr. Mi,
i **o jjo



J

I »* f«*<» f I* t r(| M,
2. TK« SO,	lr
<1 * T,M .K It.	.
"	*b» »» Vu t
I>i IV •» Ih»l fc
S^h iff
o«.*«
m Aw

?35

-------
100C-
60
50
30
21
20
10
0
78
Year
Figure 63. Growth of SCR facilities
in Japan.
234

-------
AICHI ENVIRONMENTAL RESEARCH CENTER
Aichi Kogai Center
Aichi Prefecture
March 17, 1980
K. Yoshimoto, General Manager
V. Kobayashi
Y. Takai
K. Sato, Engineer
H. Noma
M. Ito
David Hobley
Jumpei Ando
Gary Jones
Doug Maxwell
235

-------
AICHT ENVIRONMENTAL RESEARCH CENTER
The Aichi Research Center is a prefectural air, water, solid waste and
noise quality monitoring center in charge of operation and maintenance of a
large, centralized air quality monitoring network. (The Aichi prefecture
includes Nagoya and the surrounding area.) The network consists of 72 remote
monitoring stations and 3 movable stations that transmit information on NOx,
S0X, hydrocarbons, CO, oxidants, and particulates to a central data processing
center. In addition, soire of the larger strtionary sources have their own
\'0X and S0y monitors and are required to keep records. At the data processing
center, the concentration and meteorological data are analyzed to determine if
any area of the prefecture is experiencing air quality problems. If problems
exist, the Center takes appropriate action to limit emissions in that area.
For example, if high concentrations of oxidants are observed, soot and smoke
emitters in the area are contacted and directed to reduce emissions as per
previous agreement. The data collected by the Center is also used to make
trend studies and track progress in air quality.
The people at the Aichi Research Center feel that the problem with SO2
is almost over and that NO* should be the next focus. Accordingly, they plan
to convert the telemetry system for emission sources currently in use for SO2
data transmission, to NOx data transmission. Currently, the NOx data is
manually transmitted to the data processing center; however, the telemetry
system transmits measurements on 30 minute intervals. The NOx control methods
currently in operation, shown in Table 40, appear to be effective in reducing
the total ambient NOx concentration (Figure 64). The large SCR and thermal
de-NOx systems are located at Chubu Electric's Chita Power Station. SCR is
planned for installation at Chubb Electric's Atsirmi Power Station. Th'Is
system will treat flue gas from residual oil firing on a large boiler. At
Atsumi, combustion modification will reduce the NOx to 80 ppm and the SCR
will lower this to 16 ppm. The Center's experience is that SCR systems are
very reliable and have not caused any plant shutdowns.
Preceding page blank
237

-------
TABLE 40. NOx CONTROL
TECHNIQUES IN AICHI
PREFECTURE

Technique
Number of
applications

Large boilers
Small boilers
Total
Selective Catalytic Reduction
2
2
6
Thermal de-NOx
1
0
4
Staged Combustion, Burner
22
3
26
Staged Combustion, Over Fire Air
21
1
25
Flue Gas Pecycle
3
35
54
Low NO Burners
X
18
5
54
They have experienced instrument problems with stack gas instruments
located downstream from SCR systems due to the combined pressure of NH3, NOx
and SOx in the flue gas. In these situations, manual methods are used rather
than instruments.
Overall, the monitoring and feedback system are very successful, although
a strong commitment to instrument maintenance is necessary for reliable opera-
tion. To date, there have been no violations since the companies operate with
a safety factor on emission rates. Funding for the Aichi Environmental Research
Center is provided by the prefectural government. This is an unusual situation
however, since in all other prefectures the companies pay for the monitoring
and Lhe prefecture pays for the telemetry system.
238

-------
100 -I
1000 tons
per year 50 ~
0 J—J									
1974 1975 1976 1977 1978
Figure 64, N0X reduction in Aichi prefecture for
the period 1974-1978.
239

-------
KANAGAWA PREFECTURAL GOVERNMENT
Kanagawa Monitoring Center
Yokohama City
Kanagawa Prefecture
March 27, 1980
Yasuji Himi, Manager of Air Quality Management Division
David Mobley
Jumpei Ando
Gary Jones
Doug Maxwell
Preceding page blank
241

-------
KANACAWA PREFECTURAL GOVERMMENT
The prefectural government of Kanagawa operates an Air Pollution Moni-
toring Center to track emissions from all major sources in the aru-a, monitor
ambient concentrations of pollutants, and coordinate action to reduce emis-
sions if warranted by ambient conditions. In nJuition to this "real-time"
monitoring function the Center manages the historical data base for use in
trend analysis.
In spring 1980, the ambient NO2 concentration based on a 24-hour average,
exceeded the Environment Agency's upper limit of 0.06 ppm even with all
sources meeting their respective KO^ emission limitations. The prefectural
government is planning to begin a program of redicing overall emissions until
an ambient concentration of 0.04 ppm NO2 is reached. This may require in-
stallation of SCR systems on the coal-fired boilers at EPDC's Isogo Power
Station. Stricter emission control of all sources, including automobiles,
will be considered and the level of control necessary in each area of the
prefecture will be determined by analyzing the data received by the center
from its reix>te monitoring stations.
Throughout the prefecture, there are 70 monitoring stations which mea-
sure 13 separate items. The monitoring stations include fixed location air
monitors both in specific land areas and along major roads. There are also
mobile monitoring stations and a meteorological monitoring tower. The pri-
mary pollutants monitored and the measurement techniques are shown in Table 41.
Thu output of the monitoring systems is relayed to data processing
equipment in the Air Pollution Monitoring Center which analyzes and records
the data as well as displaying the results on a large graphic panel. The
display panel indicates pollutant concentrations at each of the monitoring
stations quantitatively and also qualitatively by the color of the lamp
behind each station location on the display panel map.
Preceding page blank
243

-------
TABLE 41.
POLLUTANT MEASUREMENT TECHNIQUES OF AMBIENT

MONITORS IN KANACAWA PREFECTURE
Pollutant
Measurement Technique
S02
Conductivity
NOx
Absorptiometry (Saltzman Method)
Hydrocarbon
Gas Chromotography with Flame

Ionization Detector
CO
Nondispersive Infared
Particulate
Light Scattering
Oxidants
Absorptiometry (KI Colorimetry)
When photochemical smog is anticipated or foirid to exist, the Center
contacts high level, stationary sources and asks them to take appropriate
corrective action. The Center also issues a smog notice and warning to homes
via TV and radio and to schools via local administration centers. If neces-
sary, traffic control can be initiated to limit mobile source emissions. The
facility and information flow are shown in Figure 65.
The Monitoring Center has operated since 1970. The capital and opera-
ting costs to operate the Center are shown in Table 42 which indicates that
the largest costs lie in the purchase and operation of the instruments and
associated equipment. Major cities in the prefecture have their own monitor-
ing systems, but the prefecture provides the systems for the smaller cities.
244

-------
Continuous Air Pollution Monitoring: System
Monitoring Stations	Environnant Dapartmant
"*	to NkH iMIHni C«*W
K3 in

Figure 65. Air pollution monitoring center and information flow diagram.
245

-------
TABLE 42. CAPITAL ANT) OPERATING COSTS FOR KANACAWA PREFECTURE AIR POLLUTION
MONITORING SYSTEMS (COSTS ARE PRESENTED IN THOUSANDS OF DOLLAHS)
Instruments	Telemetry	Other
Purchase Running Purchase Running Purcnase Running Total,
Year Cost	Cost	Cost	Cost	Cost	Cost	$1000
19 70
1624

-
23
-
50
-
1,697
1971
307

109
169
22
-
1
608
1972
192

144
670
26
154
2
1,188
1973
172

220
94 7
13
-
9
1,361
19 74
144

176
74
23
4
30
451
19 75
21

326
-
123
-
35
505
19 76
29

304
-
133
-
35
501
1977
455

361
-
158
32
33
1,039
19 78
672

361
-
160
8
36
1,237
19 79
301

399
213
149
7
39
1,108
1980
166

444
402
141
-
39
1,192
Total
4,083
2,
844
2,498
948
255
259
10,887
Total
6,
927


3,446
514

10,887
(Dollar
amounts
based on
exchange
rate of ¥220
per $1.)


246

-------
YOKOHAMA CITY POLLUTION CONTROL BUREAU
Yokohama
Kanagawa Prefecture
March 27, 1980
Katsumi Saruta, Director General
Koichi Ishikawa, Vice Manager
David Mobley
Jumpei Ando
Cary Jones
Doug Maxwell
247

-------
YOKOHAMA CITY POLLUTION CONTROL BUREAU
Yokohama is located near Tokyo and is the capital of the Kanagawa Pre-
fecture. The area is highly industralized and, on some days, photochemical
saog is severe enough that warnings are issued. Since the smog is a result
of atmospheric NOx and hydrocarbons, the City's Pollution Control Bureau has
a program to reduce the average level of NOy emission from all stationary
sources. A similar program for reducing hydrocarbon emissions has been
completed. The monitoring center facility is shown in Figure 66.
In terms of NOx emissions there are 1165 plants classified as emission
sources as of December 1978. 82 of these are large plants (>1 kl/hr of oil
consumption) and of these 13 (Table 43) account for 95 percent of the total
NOx emissions.
TABLE 43. MAJOR EMISSION SOURCES IN YOKOHAMA CITY
Emission Source	Number
Power Stations
coal	1
oil	1
LNC	1
Oil Refineries	3
Gas Smelters	2
Petrochemical Plants	1
Alumina Plants	1
Others	3
The City has established the goal of reducing N0X emissions to 36 per-
cent of the 19 73 level by 1985. This is equivalent to a daii\ n/erage ambient
concentration of 0.04 ppm NOj. To accomplish this objective they will request
the esiission reduction s,hovn in Table 44.
Preceding page blank
249

-------
Figure 66. Yokahama City pollution monitoring centtr.
250

-------
TABLE 44. EMISSION REDUCTION REQUESTED BY YOKOHAMA CITY
TO MEET 1985 TARCET
Sources
Requested Emission Reduction
(Based on 1974 Levels)
13 Major Sources
69 Large Sources
1083 Other Sources
70%
50-70%
30%
In terms of ppm concentrations, this means that emissions from coal-
fired boilers will be about 160 ppm and those from oil-fired boilers will be
about 40 to 50 ppm for power stations.
In order to meet the goal, it will also be necessary to ,educe emissions
from mobile sources to 30 percent of the 1974 levels. Currently, all gasoline
powered cars have 90 percent NOx control; however, this is not the case for
trucks and diesel powered passenger cars. The City cannot act directly on
mobile source emissions since that authority lies with the Central Government.
In fact, the emission regulations that are enforced by law are those of
the Environmental Agency, although some enforcement can be made based on
Perfectural Ordinance. If more stringent emission limitations for stationary
sources are desired on the local leve], they are achieved through a cooperative
spirit between industry, local government and the people. There is no active
enforcement of the regulations because, in almost every case, the regulations
are met by industry to insure the goodwill between all parties. In the few
cases where emission limitations were exceeded, the news media swayed public
sentiment against the violators and this was sufficient to convince these
sources to change their policies. In the spirit of cooperation, the local
authorities will not require emission reductions that do not appear to be
technically feasible.
251

-------
APPENDIX D
JAPANESE SCR PROCESS VENDOR CONTACTS AND US AFFILIATES
Preceding page blank
253

-------
PRINCIPAL VENDOR CONTACTS FOR JAPANESE NOx
SELECTIVE CATALYTIC REDUCTION TECHNOLOGY
Process	Japan Representative	U.S. Licensee or Partner
Babcock-Hitachi
(Hitachi, Ltd.)
Hiroshl Kuroda
Kure Works. Babcock-
Hitachi K.K.
No. 6-9 Takara-Machi
Kure-shi, Hiroshima-Ken,
737 , Japan
<0823) (21) 1161
Telex 6624-21 BHK KRE-J
Greg T. Bielawski
Babcock and Wilcox
FPG-ESC
P. 0. Box 351
Barbarton, Ohio 44203
216/753-4511
Hitachi-Zosen
(Hitachi Ship-
building and
Engineering)
Hideya Inaba
Hitachi Shipbuilding k
Engineering Co., Ltd.
Palaceside Building
1-1, Hitotsubasln, 1-Chome
Chiyoda-Ku, Tokyo, Japan
Tokyo (213) 6611
Telex J22363 • J24490
Phil Winkler
Cheraico Air Pollution
Control
One Penn Plaza
New York, NY 10001
212/239-5344
Ishikawaj ima-
Harima
Heavy
Industries
(IHI)
Nobuo Aoki
Ishikawaj ima-Haritna Heavy
Industries Co., Ltd.
Tokyo Genboku Kaikan
30-13 " 5-Chome, Tokyo
Koto-Ku, Tokyo, 135, Japan
(03) 649-1111
Telex (IHI CO) J22232
John Cvicker
Foster Wheeler Energy Corp.
9 Peach Tree Hill Road
Livingston, NJ 07039
201-533-2687
Kawasaki
Heavy
Industries
(KHI)
Senji Niwa
Kawasaki Heavy Industries,
Ltd.
14, 2-Chome
Higashikawasaki-Cho
Ikuta-Ku, Kobe, 650-91,
Japan
Tel. Kobe (078) 671-5001
Telex 5626-032 KHI CPL J
Lee Coe
Joy Manufacturing
Western Precipitation Div.
P. 0. Box 2744
Los Angeles, CA 90051
213/240-2300
Mitsubishi
Heavy
Industries
(MHI)
Tadamasa Sengoku
Mitsubish Heavy Industries,
Ltd.
Shin-Tamachi Bldg.
34-6, Shiba 5-Chome
Minato-Ku, Tokyo 108
Japan
(0.S) 455-5711
Telex J22282 HISHIJU
J28578 M1IITAM
Donald J. Frey
Manager, Fuel Systems
Engineering
CE Power•Systems
Ccjmbustion Engineering, Inc.
1000 Prospect Hill Road
Windsor, Conn. 06095
203/688-1911 Ext. 2241
Telex 99297
Preceding page blank
255

-------

-------