U.S. Army Research, Development and Engineering Command
Regenerable Desulfurization of JP-8 for Fuel Cell Power Units
Terry G. DuBois, Ph.D., Research Engineer; Command, Power and Integration Dir.,
CERDEC
26 August 2015
2015 Joint Service Power Exposition
2
Presentation Outline
Background
Program Objectives/Goals
JP-8 Desulfurizer Design and
Operating Parameters
Performance Results
Summary
Questions
3
Logistics Fuel Processing
Challenges
JP-8 Fuel
• JP-8 is a complex hydrocarbon mixture.
• Boiling range is broad (160 oC – 300 oC) making vaporization a
challenge.
Complex Blend of 100’s of
hydrocarbons
Varies by petroleum
feedstock, refinery, season,
and daily.
Contains relatively high
levels of sulfur.
alkanes cyclo-alkanes aromatics poly-aromatics
4
JP-8 Fuel Processing
Challenges
Logistic Fuel Reforming
• Catalysts deactivation by carbon deposition and sulfur are primary
technology challenges
• Improper control and/or mixing can lead to both hot and cold spots
within the reactor contributing to deactivation.
• Good operating window can be narrowly defined. [temperature,
steam/carbon ratio, oxygen/carbon ratio, fuel flow (space velocity)]
System Integration
• Thermal and physical integration
• Water balance
• Operating considerations
Poor
Mixing/Vaporization
Carbon
Deposition
Catalysts
Vaporization
Sulfur
Poisoning Catalysts
Sintering S S S S
Carbon
Deposition
Support
Failure
Catalysts Deactivation
5
Logistics Fuel Processing
Challenges
Logistics Fuel (Diesel and JP-8)
• Sulfur content
Numerous organo-sulfur
compounds can be found in JP-8.
2,3-dimethylbenzothiophene and
2,3,7-trimethylbenzothiophene have
been suggested as best
representative compounds.
Steric hindrances present in
more refractory compounds makes
sulfur sorption difficult.
Subramani Velu, Xialiang Ma, Chunshan Song, Mehdi Namazian,
Sivakumar Sethuraman, and Guhanand Venkataraman. Desulfurization of JP-
8 Jet Fuel by Selective Adsorption over a Ni-based Adsorbent for Micro Solid
Oxide Fuel Cells. Energy Fuels, 2005, 19 (3), 1116-1125.
6
Effects of Sulfur on Catalysts
Performance
Some catalysts exhibit good “sulfur tolerance”, but most, if not all catalysts
perform better with low or no organic sulfur content.
25
20
15
10
5
0
0 5 1 2 3 4 6
Hyd
rog
en
Co
mp
os
itio
n (
%)
Time On Stream (hrs.)
Source: Spivey, J, Haynes, D, Salazar-Villaphando, M., Berry, D., Shekawat, D.,
Gardner, T, “Evaluation of fuel reforming catalysts for logistics fuel cell applications,”
6th Annual Logistics Fuel Reforming Conference, May 15-16, 2006.
Fuel: n-tetradecane with 1000 ppmw as dibenziothiophene (DBT)
7
Effects of PAH on Catalysts
Performance
Poly-aromatic content in fuels has been shown to result in irreversible
deactivation in CPOx/ATR reformers.
1-methylnaphthalene (MN)
C11H8
CPOx at O/C = 1.2;
Fuel = n-tetradecane w/
5% 1-methylnaphthalene.
– Significant
negative effect with
permanent catalysts
deactivation.
Source: Haynes, D., (2007), “The Catalytic Partial Oxidation of N-tetradecane
on Rh and Sr Substituted Pyrochlores,” M.S. Thesis, Louisiana State University,
May 2007.
Time on Stream (min.)
0 50 100 150 200 250 300 350
0
10
20
30
40
50
60
70
80
90
100
Yie
ld %
5 wt% MN
added MN removed
CO
H2
CO2
CH4
Step response of LRZ after addition of 5 wt% MN (O/C = 1.2, 0.23 MPa,
900 oC and 50,000 scc/gcatalysts/hr.
8
Introduction
Objective:
Evaluate mixed metal oxide compounds for air
regenerable JP-8 desulfurization.
• JP-8 fuel: desulfurization to ≤ 10 ppmw
with 1000 ppmw JP-8 input fuel.
• Single pass yield ≥ 90%.
• Lifetime of ≥ 1000 hrs.
Fuel (Diesel/JP-8) Objective Threshold
Production Rate (ml/min.) 12 6 Organic sulfur content (ppm) 1000 400 Aromatic (%) 25 20 Start-up Lab Demo (@21 ⁰C, min.) ≤20 ≤30 Thermal Cycles 100 60 Lifetime (hrs.) 1000 800 Turn down ratio 10:1 5:1 Acoustic signature (dBA @ 7m) 50 55
9
Experimental Setup
Dual Bed Design
Absorbent: mixed metal
oxides and some catalytic
materials
Bed Volume = 0.80 L/bed
Absorbent mass = 680
gm/bed
Operating Pressure = ~150
psig
Nominal adsorption
temperature = 450 C
Power Consumption:
- Ctrl = 15.1 W
- Fan = 40.8 W
- Bed Heater = 1500W
10
Analytical Methods
• Elemental Analyzer - Analytik-Jena Ea3100
- Sulfur – ultraviolet (UV) fluorescence
Calibration Range: 0.2 ppmw S to 1000 ppmw S
- Carbon – nondispersive infrared sensor (NDIR)
Calibration Range: 84 wt% to 90 wt% carbon
• Gas Analysis
- Agilent Micro GC 3000 (2 channel)
- Columns:
(i) molecular sieve column (10 m x 0.32 mm)
(ii) Plot U column (8 m x 0.32 mm)
Multiple level calibration curves were used for all compounds with the
calibration gases at 1 % blend tolerance and 0.02% analytical
tolerance.
• Functional Group Analysis via GC-FTIR
- GC: Thermo Scientific (Trace 1310)
The GC column used was an RTX1 15 meters, 0.53mm ID, 1um
dimethyl polysiloxane stationary phase
- FTIR: Thermo Scientific (Nicolet iS50)
11
Operating Characteristics
(Temperature/Pressure)
Fuel: JP-8 w/ 314.3 ppwm total sulfur
Hours Ops: 28.9
Desulfurized JP-8 @ <1 ppmw total sulfur
• A thermal wave within the
beds can be seen for both
desulfurization and in
regeneration.
• Cycle time and regeneration
air flow are largely dictated
by the regeneration cycle
and the need to control
exothermic reactions. 0
50
100
150
200
250
300
350
400
450
500
550
600
650
55 75 95 115 135 155 175 195 215 235 255 275 295 315 335 355 375 395 415
Tem
pe
ratu
re (⁰
C) a
nd
Pre
ssu
re (p
sig)
Time (minutes)
Desulfurization RunAspen, Army Regenerable Desulfurizer
January 31, 2012
Fuel Pressure (psig) Temp, Zone 1, InletTemp, Zone 1, Outlet Temp, Zone 2, Inlettemp, Zone 2, Outlet
Zone 1 - DesulfurizingZone 2 - Regenerating
JP-8 input fuel at 6 ml/min. JP-8 feed; 450 C
adsorption temp., 150 psig bed pressure
12
Desulfurized JP-8, sulfur content
as function of input fuel
• The input fuel does affect the sulfur content of the desulfurized fuel.
• Relationship appears to be nonlinear.
6 ml/min. JP-8 feed; 450 C adsorption temp., 150 psig bed pressure
0
1
2
3
4
5
6
0 100 200 300 400 500 600 700
Su
lfu
r c
on
ten
t, d
es
ulf
uri
ze
d J
P-8
(p
pm
w t
ota
l s
ulf
ur)
Sulfur content of input JP-8 (ppmw total sulfur)
13
Desulfurized JP-8
Gaseous Component
• In addition to desulfurized fuel, the desulfurization process results in a
significant amount of hydrogen (70-80 mol%) and methane (5-9 mol%).
• Total gaseous component flow peaks at 110 minutes with a flow of approx. 820
cc/min.
• For a desulfurizer integrated into a power system, the hydrogen and methane
would be retained and would contribute the overall efficiency and yield of the
process.
6 ml/min. JP-8 feed; 450 C adsorption temp., 150 psig bed pressure
0
200
400
600
800
1000
1200
1400
1600
0 50 100 150 200 250 300
Flo
w r
ate
(cc/m
in)
Time (min.)
Flow Rate_GC data
0
2
4
6
8
10
12
14
16
18
0
10
20
30
40
50
60
70
80
90
0 50 100 150 200 250 300 350 400
Mo
lar
co
nc. o
f C
H4, C
O, C
2H
6 (%
), d
ry b
asis
Mo
lar
co
ncen
trati
on
of
H2 (%
), d
ry b
asis
Time (minutes)
H2
CH4
CO
C2H6
14
Performance throughout a
Sorption Cycle
• System warm-up time ~17 – 18 minutes. Desulfurized fuel begins to
flow ~9 minutes into the desulfurization cycle.
• Desulfurized fuel production is linear (mass basis). Desulfurization
appears to be less efficient at the beginning and end of each cycle.
• With every cycle approximately 60 g of fuel is captured as “condensate”
representing fuel purged from the bed during transition to regeneration.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
31 80 118 166 206 260
To
tal s
ulf
ur,
pp
mw
Cycle Time (min.)
Sulfur
6 ml/min. JP-8 feed; 450 C adsorption temp., 150 psig bed pressure
0
10
20
30
40
50
60
70
80
90
100
0
200
400
600
800
1000
1200
0 50 100 150 200 250 300
Pro
cess Y
ield
Fu
el C
on
su
med
/Pro
du
ced
(g
)
Time (minutes)
JP-8 Consumed
DeS JP-8 Produced
Yield
15
Process Yield
75
80
85
90
95
100
0 50 100 150 200 250 300 350 400 450 500
Yie
ld (
%)
Time on stream (hrs.)
Yield
Yield with Condensate Recycle
• Yield with and without
condensate recycle. (JP-8
input fuel at 400 ppm total
sulfur).
• The Yield, as shown, is
mass based and liquid
content. The condensate is
fuel that has not been fully
desulfurized fuel contained
in the beds during transition
from desulfurization to
regeneration. Condensate
JP-8 typically contains 15 –
25 ppmw total sulfur and
within a system would be
recycled back to the fuel
tank or used directly.
JP-8 input fuel at 400 ppmw total sulfur. 6
ml/min. JP-8 feed; 450 C adsorption temp., 150
psig bed pressure
16
Impact of TOS and
Thermal Cycling
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
To
tal su
lfu
r (p
pm
w)
Time on stream (hrs.)
Thermal Cycles
• Sulfur content of
desulfurized fuel as a
function of time-on-
stream. (JP-8 input
fuel at 400 ppmw total
sulfur)
• Achieved ≥ 475
hrs/bed and ≥ 100
thermal cycles/bed.
• Some degradation in
performances is seen,
but operation is well
within goals of ≤ 10
ppmw total sulfur.
JP-8 input fuel at 400 ppmw total sulfur. 6
ml/min. JP-8 feed; 450 C adsorption temp., 150
psig bed pressure
17
IR Spectral Results
• JP-8 versus desulfurized JP-8
• Significant changes in Gram-Schmidt from 3 to 5 minutes and at 10
to 12 minutes, but little change elsewhere.
18
IR Spectral Results (con’t)
Desulfurized JP-8
Selected spectra collected at 3 to 5 min. retention times
• The desulfurizing process has increased the number of branched alkanes in
JP8. Note increase in CH3 bands at 2954 and 1388cm-1.
19
80
81
82
83
84
85
86
87
88
89
90
31 80 118 166 206 260
Ca
rbo
n (
wt%
)/ d
en
sit
y X
100
(g
/ml)
Cycle Time (min.)
Carbon (%) Density X 100 (g/ml)
Source Fuel Density
Source Fuel Carbon wt%
Desulfurized JP8 Carbon
Content and Density
• JP-8 Fuel specification:
- Carbon max (wt%)* 86.6
- Density (g/ml) 0.775 – 0.840
• Desulfurization process alters fuel density and increases amount of carbon.
- May be increasing mono-aromatic content.
* Based on minimum hydrogen specification
20
Summary
JP-8 Fuel Desulfurization
• Successful operation of an air regenerative JP-8 fuel desulfurizer has
been demonstrated.
• JP-8 fuel with organic sulfur content up to 700 ppm total sulfur was
desulfurized to levels below 7 ppmw total sulfur.
• Very good absorbent durability has been demonstrated, achieving > 950
hours on-stream operating hours and > 105 thermal cycles.
• Yields of greater than 90% can easily be achieved.
• The desulfurized fuel is altered a bit with increased lower molecular
weight hydrocarbons, increased branched hydrocarbons, and possibly
increased aromatic content.
Fuel Work
• Continued characterization of desulfurized fuel.
• Evaluate changes in operation parameters (bed residence time,
operating pressure and temperature, etc.) on performance.
21
Acknowledgement
This presentation is based on the work of my colleagues at
CERDEC, Command, Power and Integration Directorate: Mr.
Michael Seibert and Mr. Richard Scenna.
In addition, we gratefully acknowledge the contributions of Dr. Mark
Fokema (Aspen Products, Inc.), Mr. Will Wihlborg (Thermo Fischer
Scientific), and Dr. Tim Edwards (AFRL).
22
Questions
