Pilot Testing of Mercury Oxidation Catalysts for Upstream of Wet FGD … Library/Research/Coal/ewr... · • Measured performance to see if trend for ... • Proposal submitted in

Pilot Testing of Oxidation Catalysts for Enhanced Mercury

Control by Wet FGD

Gary Blythe

URS Corporation

Austin, TX

Hg Control Technology Concept

• Catalytic oxidation of Hg0 in flue gas to increase overall Hg removal across wet FGD systems

• Initial development is focused on PRB and lignite fuels - higher Hg0

percentages in flue gas• Catalyst to be installed at ESP outlet

Illustration of Process Concept

Wet FGD System (SO2/Hg Removal)

ESP

Mercury Oxidation Catalyst

Process Development Background

• Initial concept development funded by EPRI (early 1990s)

• Further development in DOE NETL/ EPRI co-funded MegaPRDA Project– Lab screening of candidate catalyst

materials– 6-month sand bed tests with promising

catalysts at three coal fired sites– Completed in 2001

Current Project

• Conduct pilot-scale tests of honeycomb Hg0

oxidation catalysts at two sites– 4 catalysts tested in parallel (~2000 acfm each)– 14-months automated operation at each site– Monthly activity measurements with Hg SCEM– Ontario Hydro relative accuracy tests at beginning,

middle, end of test periods

• DOE/NETL, EPRI, utility co-funded

• Host stations include ND lignite (GRE’s Coal Creek), PRB fuels (CPS’s Spruce)

Catalyst Types Tested

• Metal-based– Palladium (Pd #1) - both sites– Ti/V from Argillon (SCR) - both sites– Gold (Au) - Spruce only

• Carbon-based– Experimental activated carbon (C #6) - both sites

• Fly-ash-based– Subbituminous coal ash from one particular plant

(SBA #5) - CCS only

Pilot Unit Concept

From ESPor Baghouse

DPI

DataLogger

InducedDraft Fan

Mercury OxidationCatalyst

FIC

HgHg

TE

TI

To FGD

Pilot Testing Status

• First pilot unit started up at Coal Creek in October 02– 2 of 4 catalysts installed (Pd #1 and SCR)– Delivery of other two catalysts delayed due to

developmental nature of their production

• 3rd catalyst (SBA #5) installed December 02

• 4th catalyst (C #6) installed June 03

• Second pilot unit to start up at Spruce Plant later this month

Pilot Unit at Coal Creek Station

Catalyst Dimensions for Pilot Unit

Catalyst

Cells perin.2

(cpsi)

CrossSection(in. x in.)

Length(in.)

AreaVelocity(sft/hr)

Pd #1 64 30 x 30 9 49

C #6 80* 36 x 36 9 27

SBA #5 80* 36 x 36 9 27

SCR 46 35.4 x 35.4 19.7 14***Die sized for 64 cpsi, cores shrank during drying**1500 acfm, other catalysts operate at 2000 acfm

Photo of Argillon GmbH SCR Catalyst Module

One of Three SBA #5 Catalyst Modules

Close-up of One SBA #5 Block

Initial CCS Pilot Activity ResultsOct. 02 field results vs. lab predictions

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60Area Velocity, sft/hr

Hg0 O

xida

tion

Acr

oss

Cat

alys

t, %

Pd #1 - CCSPd #1 - PredictedSCR - CCSSCR - Predicted

December 02 Results

• Measured activity of Pd #1 and SCR catalysts after ~2 months– Both catalysts showed activity loss

• Pressure drop across both catalysts had substantially increased

• Suspected fly ash buildup in horizontal catalysts

January 03 Results

• Measured performance to see if trend for loss of activity continued– Activity slightly improved since December– Pressure drop indicated continued ash buildup

• Opened catalyst boxes, confirmed fly ash buildup

• Used plant air and vacuum to clean

• Re-measured performance

Surface of Pd #1 Catalyst Prior to Cleaning

Surface of Pd #1 Catalyst after Cleaning

Catalyst Activity Results

Hg0 Oxidation across Catalyst (%)Catalyst(Flow Rate,acfm) October December January

January(aftercleaning)

Pd #1 (2000) 93 53 58 91SCR (1500 ) 67 28 37 61SBA #5(2000)

na* na* 59 75

Efforts to Resolve Ash Buildup

• Identified sonic horns as a likely mechanism to limit fly ash buildup– Commonly used to clean SCR catalysts

• Tested horn (Analytec 17”) on Pd #1 catalyst chamber– Installed perpendicular to chamber inlet

transition duct– Appeared effective in 2 mos. operation– Installed on other chambers June 03

Sonic Horn Installation on Pilot Unit

Sonic Horn Location

Catalyst Pressure Drop thru 5/27 (shows sonic horn effect on Pd #1 catalyst)

Pd #1

0

1

2

3

4

5

6

7

8

9

10

0:01:4210/1/02

5:01:4310/31/02

6:01:4111/30/02

8:01:4212/30/02

8:01:591/29/03

10:35:103/1/03

6:01:404/1/03

10:43:385/11/03

Date/Time

Cat

alys

t Pre

ssur

e D

rop,

in. H

2O

Catalyst Activity Trends over First 8 Months at Coal Creek

0

20

40

60

80

100

1-Oct 1-Dec 31-Jan 2-Apr 2-Jun 2-Aug

Date

Hg0 O

x. A

cros

s C

atal

yst,

%

Pd #1

C #6 (initial)

Cleaned Catalyst

Installed Sonic Horn Cleaned

Catalyst (removed light dusting)

Catalyst Activity Trends over First 8 Months at Coal Creek

0

20

40

60

80

100

1-Oct 1-Dec 31-Jan 2-Apr 2-Jun 2-Aug

Date

Hg0 O

x. A

cros

s C

atal

yst,

%

SCR

SBA #5

Cleaned Catalysts

Installed Sonic Horn,

Cleaned Catalysts

Catalyst Pressure Drop since 6/5 (sonic horns in all 4 compartments)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

12:53:126/5/03

9:04:376/10/03

5:01:536/15/03

20:01:516/20/03

16:01:526/25/03

12:01:526/30/03

9:00:397/5/03

Date/Time

Cat

alys

t Pre

ssur

e D

rop,

in. H

2O

SBA #5SCRC #6Pd #1

Flue Gas Characterization Results - Other Species

• Ontario Hydro results show good agreement with EPRI Hg SCEM

• Controlled Condensation results showed no oxidation of SO2 across catalysts– Catalyst inlet and outlet SO3 ~0.1 ppmv

• Little or no oxidation of NO across catalysts– Inlet and outlet NO, total NOX values agree within 10

ppm (precision level of technique)

• M26a showed no change in HCl (~1 ppm) or HF (~6 ppm) across catalysts

Lab Testing

• 1 to 2 l/min testing of catalyst cores in simulated flue gas (SO2, HCl, NOX, etc.)

• Tested TVA’s patented gold catalyst in honeycomb form, varied gold loadings– Activity compares favorably with Pd #1– May be effective at lower wash coat loading on

alumina than Pd #1– Selected as 4th catalyst for Spruce (SBA #5 fly

ash not a likely commercial catalyst source)

• Tested other catalyst materials at Spruce conditions

Results of Spruce Laboratory Simulations

20

30

40

50

60

70

80

90

100

0 20 40 60 80 100 120Area Velocity (std ft/hr)

% H

g0 Oxi

datio

n

Pd #1, 3x, 1" lengthSBA #5, 0.5" lengthAu, 1x, 1" length (preliminary)Au 1x, 1"length, repeatSCR, 1"lengthC #6, 1"length

Catalyst Selection for Spruce Pilot

• Pd #1 and C #6 selected based on high activity at CCS

• Au selected based on positive lab results– Need field results to compare activity, life to Pd #1

• Will test SCR catalyst in spite of activity loss at CCS– Loss could be site specific based on previous

PRDA sand bed results with other catalysts

• All catalysts re-ordered at current pitch (Au the same as Pd)

Current Schedule

• Continue SCEM measurement trips to Coal Creek through early 04 – Conduct 3rd set of Ontario Hydro SCEM

relative accuracy tests ~March 04

• Start up 2nd pilot unit at Spruce later in August– Initial shake-down operation w/o catalysts– Expect Pd and Au catalysts by end of August– SCR and C #6 catalysts late September

Proposed Follow-on Project• Proposal submitted in NETL large-scale Hg control

testing solicitation

• Would use existing pilot units to test Hg oxidation catalysts at 2 new sites starting Spring 04– TXU’s Monticello Station (Tx lignite/PRB blend)– Duke Energy’s Marshall Station (low S Eastern bit.)

• Proposed effort would integrate new wet FGD pilot unit downstream of oxidation catalysts– 2000 acfm flow rate to match catalysts– Would test LSFO vs. Mg-lime chemistries

Summary

• Pilot tests results verify previous sand-bed results for the ability to catalytically oxidize Hg0 in flue gases– Honeycomb catalysts have achieved over 90%

oxidation of Hg0 in ND lignite flue gas

• On-line cleaning (sonic horns) needed to prevent fly ash buildup in horizontal gas flow catalysts

• Continued testing will establish catalyst life for two coal types