sorbents for desulfurization of natural gas, lpg, and ......
TRANSCRIPT
TDA Research Inc. • Wheat Ridge, CO 80033 • www.tda.com
Sorbents for Desulfurization of Natural Gas, LPG and Transportation
Fuels
Gökhan O. Alptekin, PhD
Sixth Annual SECA WorkshopPacific Grove, California
April 21, 2004
Introduction
Advances in fuel cell technologies have the potential to revolutionize the way the power is generated and distributed Fuel cells require an ample supply of high qualityfuelPipeline natural gas is the fuel of choice because of its abundance and well-developed supply infrastructureIn addition to naturally occurring H2S, chemical odorants made with sulfur-containing compounds are added to natural gas for leak detection
Common odorants include mercaptans, sulfides and thiolsThe total sulfur content of the pipeline gas averages about 4 ppmv but can be as high as 10-12 ppmv
TDAR e s e a r c h
Sulfur As an Electrocatalyst Poison
TDAR e s e a r c h
Sulfur compounds contaminate the catalysts used in fuel cell systems and degrade power generation performance
Traditionally sulfur removal is carried out with a two-step process:HDS of the organic sulfur compounds and subsequent H2S removalIt is not practical for small-scale residential units or for transportation systems
Source: Israelson, “ Results of Testing Various Natural Gas Desulfurization Sorbents”, J. of Materials
Engineering and Performance, Vol. 13 (3), June 2004
Source: De Wild, “The removal of sulfur-containing odorants from natural gas for PEMFC”, Proceedings of
Fuel Cell Seminar, p227, 2002
Project Objective
TDAR e s e a r c h
TDA Research, Inc. is developing a passive adsorbent for ambienttemperature natural gas desulfurization
Requirements of the sorbentHigh sulfur capacity (minimum replacement frequency, small size)Low costReducing total sulfur concentration to sub-ppm levels Inertness (no side reactions or chemisorption of hydrocarbons)Tolerance to possible natural gas contaminants (hydrocarbons, CO2, H2O)Ease of disposal (no toxicity, flammability, pyrophorocity)
Fuel
Water
Steam
Burner
Steam Reformer
HEX
WGS PROX
Air
PEMFC
Air
Fuel Cell Off Gas
DesulfurizerPre-
reformer
PEM System SOFC System
WaterSteam
SOFC
HEX
DesulfurizerPre-
reformerFuel
Outline
IntroductionOutlineNatural Gas Desulfurization
Experimental SetupPerformance Comparison with Other SorbentsParametric TestsSorbent RegenerationDemonstrations
LPG DesulfurizationSorbent Performance
Desulfurization of Transportation FuelsExperiments with Model FuelsBattlefield Fuels (i.e., JP-8)
Conclusions
TDAR e s e a r c h
Experimental Setup
TDAR e s e a r c h
A high pressure quartz reactor (with internal pressure stabilization) was used for bench-scale testing
All lines and system components were either quartz, Teflon or Silcosteel– to minimize sulfur interaction with system components
Con
cent
ratio
n,
ppm
v
Time, min
0.5
1.0
2.5
1.5
2.0
3.0
3.5
Sulfur Analysis
Two gas chromatographs with sulfur chemiluminescence detector (SCD) and flame photoionization detector FPD) were used to analyze organic sulfur speciesThe detection limit of the SCD and FPD were 1 and 50 ppbv, respectively
signal/noise ratio greater than 10Restek RTX-1 column was used for separation of the sulfur species
Component name
Retention time
H2S= 2.1 ppmvDMS= 1.8 ppmvTBM= 1.3 ppmvTHT= 1.6 ppmv
Odorants Tested Formula AcronymIsopropyl Mercaptan (CH3)2CHSH IPMTert-butyl Mercaptan (CH3)3CSH TBM
Dimethyl Sulfide CH3SCH3 DMSHydrogen Sulfide H2S H2S
Tetrahydrothiophene CH2CH2CH2CHS THTEthyl Mercaptan EMC2H5SH
TDAR e s e a r c h
Natural Gas Analysis
Natural gas composition during sorbent testingComponent Volume % Volume %
Sample Jan. 2004 Sample June 2004Methane 92.83 92.39Ethane 3.30 3.42Propane 0.61 0.56Butane 0.13 0.11Isobutane 0.12 0.12Pentane 0.10 0.11Isopentane 0.10 0.10Neopentane 0.10 0.10Hexane 250 ppm 280 ppmCarbon Dioxide 0.70 0.81Nitrogen 2.01 2.08
In selected tests, the natural gas contained up to 200 ppmv water vapor
TDAR e s e a r c h
Two gas chromatographs with a TCD and FID detectors were used toanalyze natural gas componentsThe FID detector could measure as low as 1 ppmv hexane in natural gas
Typical Test Profile
0
10
20
30
40
50
60
70
0 1000 2000 3000 4000 5000 6000Run Time, min
DM
S C
onco
ncen
trat
ion,
ppm
v
Feed Analysis Feed Analysis
DMS Breakthrough
T= 22oC, P= 50 psig, DMS Inlet= 60 ppmv, GHSV= 1,500 h-1
TDAR e s e a r c h
Prior Work
Prod. Name
Source: Israelson, “ Results of Testing Various Natural Gas Desulfurization Sorbents”, J. of Materials Engineering and Performance, Vol. 13 (3), June 2004
M anufacturer Active Com ponent Product Nam e
DM S (ppm v)
Sulfur (ppm v)
DM S Index
United Cata lysts, Inc. (Süd-Chem ie)
ZnO @ 350°C (No preceding CoMo catalyst bed or hydrogen addition) G -72E 1.5 6.7 668
Norit Carbon w. chrom ium & copper salts RG M-3 0.8 5.1 24000
Calgon Carbon Carbon PCB 1.5 6.2 26550
United Cata lysts, Inc. (Süd-Chem ie) Carbon w. copper oxide C8-7-01 1.8 6.7 27900
Süd-Chem ie N ickel, N ickel O xide (Unheated) C28 1.1 4.2 44992
G race Davison
Molecular S ieve; 13X (10 Å pore) zeolite-X 544HP 0.8 4.4 110376
Supplier C Unknown Proprie tary Adsorbent S 1.2 5.1 182573
Pacific Northwest National Lab (PNNL)
Copper im pregnated zeolite-Y substrate Unnam ed 0.8 4.1 2416800
Synetix Copper O xide & Z inc Oxide 100°C top, 170°C bottom
Puraspec 2084 3.4 8.4 3661800
Supplier E Unknown Proprie tary Adsorbent T 2.3 8.2 3814300
DMS Index = average ppm DMS x cubic feet of natural gas/cubic feet of adsorbent
TDAR e s e a r c h
Performance Comparison at Bench-scale
0
2
4
6
8
10
12
0 100 200 300 400 500 600 700 800 900 1000Time (min)
DM
S C
once
ntra
tion
(ppm
v)
TDA's SulfaTrap-BR3
Siemens Sample 5
Siemens Sample 4
Grace X Zeolite
Norit RGM3, Activated Carbon
TDAR e s e a r c h
T= 20oC, P= 3 psig, DMS = 12.3 ppmv, TBM = 9 ppmv, THT= 9 ppmv in Nat. Gas GHSV= 60,000 h-1
TDA’s sorbent showed the highest sulfur adsorption capacity 60% higher sulfur capacity than Siemens Sample #5
Norit RGM3 Activated Carbon
Grace Zeolite X
Siemens Sample 4
Siemens Sample 5
TDA SulfaTrap-
BR3
0.36%Grace X Zeolite0.18%Norit RGM3 Activated Carbon
0.85%Siemens Sample 41.96%Siemens Sample 53.12%TDA’s SulfaTrapTM
Pre-Breakthrough Capacity (% wt.)
Sample
0.36%Grace X Zeolite0.18%Norit RGM3 Activated Carbon
0.85%Siemens Sample 41.96%Siemens Sample 53.12%TDA’s SulfaTrapTM
Pre-Breakthrough Capacity (% wt.)
Sample
Hydrocarbon Adsorption
TDAR e s e a r c h
The sorbent does not adsorb any hydrocarbon species from the natural gas
Even the heavy hydrocarbons such as hexane were not removed
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 500 1000 1500 2000 2500 3000Run Time, min
Con
cent
ratio
n, %
vol
.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Propane Iso Butane N-butane Iso-PentaneNeo-Pentane Pentane Hexane Ethane
Etha
ne C
once
ntra
tion,
% v
ol.
Ethane--->
<---Propane
<---Hexane
<---C4s&C5s
Potential Side Reactions
T= 20oC, P= 17 psia, DMS = 17.0 ppmv, TBM= 7 ppmv, THT= 5 ppmv, GHSV= 60,000 h-1
02.2
4.46.6
8.80
120
240
360
480600
Rel
ativ
e In
tens
ity
GC Test Time (min)
Test Time ...
Feed Analysis
Breakthrough
DMS
B-Merc
Test Start
TDAR e s e a r c h
The organosulfur species do not undergo any side reactions that cause formation of complex sulfur species
TDAR e s e a r c h
Effect of Water Content in Natural GasT= 20°C, P= 3 psig, DMS Inlet= 12.4 ppm, TBM= 9 ppm, THT= 9 ppm in Natural Gas, GHSV= 60,000 h-1
0
2
4
6
8
10
12
14
0 100 200 300 400 500 600 700 800Time (min)
DM
S C
once
ntra
tion
(ppm
v)
Norit RGM3, Activated Carbon Grace Zeolite X 2%Cu 4%Fe/MFI (Seimens)
Dashed Lines - Tested in natural gas with 45 ppmv waterSolid Line - Tested in dry natural gas
SulfaTrap-R1 (tested at Siemens)
Pipeline gas may contain up to 155 ppmv of water vaporSulfaTrapTM-R1 sorbent was impacted the least by presence of water
Cyclic CapacityT= 20oC, P= 3 psig, DMS = 7 ppmv, TBM= 7 ppmv, THT= 15 ppmv, GHSV= 75,000 h-1
T Adsorption= 20oC T Regeneration = 300oC
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
0 2 4 6 8 10 12Cycle
Bre
akth
roug
h C
apac
ity
Hydrogen RegenerationNatural Gas Regeneration
TDAR e s e a r c h
Sorbent maintains its capacity for 10 adsorption/regeneration cyclesSorbent regenerates using clean natural gas as well as hydrogen
DMS Breakthrough Profiles in the 10-Cycle Test
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 50 100 150 200 250 300 350 400 450 500
Time (min)
DM
S C
once
ntra
tion
(ppm
v)
Cycle 1, Hydrogen RegenerationCycle 2, Natural Gas RegenerationCycle 3, Natural Gas RegenerationCycle 4, Natural Gas RegenerationCycle 5, Natural Gas RegenerationCycle 6, Natural Gas RegenerationCycle 7, Natural Gas RegenerationCycle 8, Natural Gas RegenerationCycle 9, Hydrogen RegenerationCycle 10, Hydrogen Regeneration
T= 20oC, P= 3 psig, DMS = 7 ppmv, TBM= 7 ppmv, THT= 15 ppmv, GHSV= 75,000 h-1
TDAR e s e a r c h
Temperature-Programmed Desorption
0
20
40
60
80
100
120
140
0 50 100 150 200 250
Time (min)
Con
cent
ratio
n (p
pm)
0
50
100
150
200
250
300
350
400
Tem
pera
ture
(°C
)
DMS
TMB
THTTemperature
The sorbent DMS interaction is the weakest as evident by low desorption temperature Close sulfur balance between the adsorption and regeneration indicates full regeneration potential for the sorbent TDA
R e s e a r c h
5 kWe SOFC Alpha Testing at SWPC
TDA supplied its sorbent to Siemens Westinghouse to provide natural gas desulfurization during alpha testing of their SOFCsSorbent bed was sized to provide 1 year continuous operation
2.2 L sorbent (assuming 10 ppmv sulfur content for the gas all of which is DMS)1/16” cylindrical pellets
2,700 hrs testing was completed in March 2005The same canister will be used for additional tests
TDAR e s e a r c h
Alpha Testing Results
0
2
4
6
8
10
12
14
16
18
29-Nov-04 19-Dec-04 8-Jan-05 28-Jan-05 17-Feb-05 9-Mar-05 29-Mar-05Test Duration
Fuel
Flo
w, s
lpm
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Fuel Flow Total Sulfur, ppmv
Tota
l Sul
fur C
once
ntra
tion,
ppm
v
0
2
4
6
8
10
12
14
16
18
29-Nov-04 19-Dec-04 8-Jan-05 28-Jan-05 17-Feb-05 9-Mar-05 29-Mar-05Date
Fuel
Flo
w R
ate,
slp
m
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0Fuel Flow, slpmDMS, ppmv
DM
S C
once
ntra
tion,
ppm
v
TDA’s SulfaTrapTM-R1 sorbent removed all sulfur for over 2,700 hrsThe average concentration of the sulfur species in Pittsburgh pipeline gas were as follows:
Variation in sulfur concentration Variation in DMS concentration
*Data provided by Gordon Israelson, SWPC
H2S EM DMS TBM THT NPM Total Sppmv ppmv ppmv ppmv ppmv ppmv ppmv0.10 0.07 0.45 0.59 0.61 0.52 2.39
TDAR e s e a r c h
Water Vapor in Natural Gas
Natural Gas Water Vapor
0200400600800
100012001400
17-F
eb-0
5
22-F
eb-0
5
27-F
eb-0
5
4-M
ar-0
5
9-M
ar-0
5
14-M
ar-0
5
19-M
ar-0
5
24-M
ar-0
5
29-M
ar-0
5
3-A
pr-0
5
8-A
pr-0
5
Date
Wat
er V
apor
, ppm
v
SWPC measurements showed that natural gas used during the testing contained ~500 ppmv water vapor
*Data provided by Gordon Israelson, SWPC
TDAR e s e a r c h
Field Demonstration with a 5 kWe SOFC
A 5 kWe SWPC SOFC combined with TDA’s desulfurizer will be delivered to CTC Fuel Cell Test Facility in Johnstown, Pennsylvania
Desulfurizer dimensions 3” x 21”The fuel cell will be delivered to US Army base at Fort Meade, MD
TDAR e s e a r c h
Field Demonstration with a 125 kWe SOFC
TDA is also preparing a desulfurizer unit for a 125 kWe CHP SOFC system
Sorbent size = 150 L1/8”cylindrical pellets
TDA scaled-up the production capacity two-folds
The sorbent produced with high throughput equipment (e.g., spray dryer, screw extruder) exhibited good performance
The unit will be shipped to Hanover, Germany
TDAR e s e a r c h
Desulfurization of LPG
TDAR e s e a r c h
Liquified petroleum gas (LPG) has higher power density than natural gas on volume basis LPG-fed systems are suitable fuel for portable systems and applications in remote locationsMercaptans (mostly ethyl, n-propyl and isopropyl mercaptans) are the primary sulfur species in LPG
The presence of unsaturated hydrocarbons affects the performance of the desulfurization sorbents
0 60 120 180 240 300 360 420 480
GC Time, sec
AAA PropaneIndustrial Grade Propane
Met
hane
Etha
ne
Prop
ane/
Prop
ylen
e
But
ylen
e
Iso-
But
ane
n-B
utan
e
Sorbent Performance for LPG Desulfurization
TDAR e s e a r c h
0 2000 4000 6000 8000Time (min)
Eth
yl D
isul
fide
(Rel
ativ
e In
tens
ity)
C2H5SH C4H10S2
Ethyl Mercaptan Diethyl disulfide
TDA’s SulfaTrapTM-P sorbent achieved sulfur capacity of 0.63% and 2.35% wt. at GHSV of 30,000h-1 and 5,000h-1, respectively (based on diethyl disulfide breakthrough)
GHSV = 30,000 h-1
13 ppmv EM
GHSV = 5,000 h-1
25 ppmv EM
T=20°C, P=5 psig, GHSV= 5,000-30,000 h-1, Ethyl Mercaptan Inlet = 13 ppmv–25 ppmv
Field Demonstration with a 100 We SOFC
TDA will supply MesoscopicDevices with its LPG desulfurization sorbent75-100 We portable devices with up to 10 day continuous operation6 units will be developed for propane-fed SOFC by June 2005 and shipped to the Army for field testingUltimate goal is to develop an effective sorbent for transportation fuels (gasoline and diesel) and logistics fuels (e.g. JP-8 fuel)
TDAR e s e a r c h
TDAR e s e a r c h
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0 50 100 150 200 250 300 350 400Time, min
Ben
zoth
ioph
ene
Con
cent
ratio
n, p
pmv
Cycle #1 Cycle #2
Cycle # Benzothiophene Capacity Total Sulfur Capacity(% wt.) (% wt.)
1 0.35 0.622 0.33 0.61
A Regenerable Sulfur Sorbent for Liquid Fuels
Benzothiophene breakthrough earlier than 2-methyl benzothiopheneThe sorbent was regenerated in hydrogen with a mild temperature swing
T= 60oC, LHSV= 1 h-1, 1,100 ppmw Benzothiophene, 900 ppmw 2-methyl Benzothiophene in a model fuel (70% aliphatic HCs, 30% aromatics)
Mo
S
4,6-dimethyl dibenzothiophene
In the refractory sulfur molecules, the sulfur atom is sterically hindered and difficult to activate
TDAR e s e a r c h
Sulfur Removal From JP-8
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 50 100 150 200 250 300 350
Time (min)
Per
cent
Rem
aini
ng S
ulfu
r
t= 60 min
t= 90 min
JP-8 Feed Analysis
TDA’s sorbent can remove all the sulfur in the JP-8 fuelSulfur speciation and capacity calculations are underway
T=60°C, Sulfur Content= ~3,000 ppmw, LHSV= 1 h-1 (Flow = 0.12 ml/min, 3.0 g sample)
Conclusions
TDA's SulfaTrapTM Sorbent Activated CarbonOperating Temperature Ambient AmbientBed Volume 1-2 L 30-50 LHydrocarbon Adsorption Minimal SubstantialEnd-of-Life Indication Yes NoRegenerability Yes NoFlamability No YesDisposability Easy Difficult
small volume toxic, pyrophoric
TDA’s sorbent can be effectively and economically used for natural gas desulfurization
Sorbent cost is estimated ~$10-25/lb (depending on production scale)Based on the conditions of the alpha test (~14 slpm fuel flow, 2.4 ppmv sulfur), sorbent cost is estimated as $4.71/1,000 m3 natural gas or $16 to $40/year ($3-$8/kW)TDA’s SulfaTrapTM-P sorbent can achieve 2.35% wt. capacity for desulfurization of LPGTDA’s sorbent also shows promise for JP-8 desulfurization
TDAR e s e a r c h
Acknowledgements
Funding for the research is provided by DOE Phase II SBIR Grant, Contract No.DE-FG02-03ER83795 and DOD Phase I SBIR Project, Contract No. W6HZV-05-C-0121
Authors thank:Donald Collins, NETL, DOE Magda Rivera, NETL, DOEKevin Mills, US Army TACOMGordon Israelson, Siemens WestinghouseJoe Poshusta, Mesoscopic Devices, LLC
TDAR e s e a r c h