PARTIAL OXIDATION OF PROPANE TO HYDROGEN

1
PARTIAL OXIDATION OF PROPANE TO HYDROGEN OVER
Pt/Ceria CATALYSTS Confirming The Sequence Of
Reactions And Role Of Ceria
Tapan K. Das1, Edwin L. Kugler Dady B.
Dadyburjor Department of Chemical
Engineering West Virginia University, Morgantown
WV 26506 Annual Technical Meeting, Consortium
for Fossil Fuel Science, Lexington KY July 29,
2007
ACKNOWLEDGMENTS USDOE Collaboration with
CFFS ____ 1 current address Chevron Research,
Richmond CA
2
OUTLINE

• Introduction, The Story So Far
• The Ambiguity
• The Resolution, Experiments
• Conclusions

3
INTRODUCTION, THE STORY SO FAR

• Hydrogen from gasoline or diesel by on-board
  reforming
• Propane used as a surrogate for gasoline
• Ceria has interesting oxidation characteristics
• (also good catalyst for WGS)
• Hence POX of Propane over 1 Pt / CeO2
• different surface areas H 80.4 m2/g
• L
  9.9 m2/g

• Experiments with varying T, W, V, O/F in
  Fixed-Bed Reactor
• Exit Species H2, CO, CO2, C3H6, O2, C3H8
• Large number of possible reactions TO, PO,
  WGS, SR, DR, WF, DH,

• Questions
• How to analyze results ?
• Sequence PO then TO or TO then PO
• Influence optimum reactor
  type

4
THE STORY SO FAR (contd)

• Gauss elimination gt Reaction Matrix is of order
  4
• i.e., 4 independent reactions exist
• Which four (out of 7) ?
• Use reasonableness and physical sense
• Calculate rates for these four as f (W, Contact
  Time)
• to determine sequence of PO / TO

• In all cases, find as Contact Time R
  (PO) R (TO)
• Hence PO first, then TO
• Very small Contact Time (or W) before TO
  is large
• gt Microchannel reactors in parallel ?

• The four reasonable and physical independent
  equations are
• For High-SA CeO2 support, largest W (0.02
  g) TO, PO, WG, DH
• 
  smaller W (0.01 g) TO, PO, WF, DH
• For Low-SA CeO2 support, both loadings TO,
  PO, WF, DH

5
THE AMBIGUITIES

• The set of reactions TO, PO, SR, DR are
  acceptable
• but not physically reasonable
• Why is WGS acceptable for one situation, but not
  for others ?

THE RESOLUTION
Carry out independent experiments on SR, DR and
WGS
6
THE RESOLUTION (contd.)
Steam Reforming Activity for 1 Pt/CeO2 High-SA
(H) and Low-SA (L) catalysts, Large W (0.02
g) Total Flow rate 300sccm, H2O / C3H8 5, (Ar
N2) 90

• SR is negligible at 600C

7
THE RESOLUTION (contd.)
Dry Reforming Activity for 1 Pt/CeO2 High-SA
only, Large W and Small W catalyst, 600C Total
Flow 300sccm, CO2/C3H8 1, (Ar N2) 90
Catalyst Loading W (g) 0.01 0.02 C3H8
conversion () 0.2 0.3 Outlet flow (sccm)
H2 0.01 0.02
CO n.d. 0.01

• Reaction Sets with SR, DR are not physical

8
THE RESOLUTION (contd.)
WGS Activity for 1 Pt/CeO2 High-SA (H) and
Low-SA (L) Catalyst, Large W (0.02 g) Total Flow
rate 300sccm, H2O / CO 8.3, (Ar N2) 90

• No WGS activity for L
• Significant WGS activity for H
• Total amount of ceria important in determining
  net loss/formation of water

9
THE RESOLUTION (concld.) Why is the total amount
of ceria present important ?

• Surface CeO2 is reduced for High-SA supported
  catalyst,
• negligible for
  Low-SA
• per our earlier TPR data

• Lo-T peak surface ceria
• Hi-T peak bulk ceria
• Pt catalyzes reduction of ceria
• (peaks moved to lower T)
• Lo-T peak for H reduced significantly,
• No change in Lo-T peak for L
• No change in Hi-T peaks for L or H

• Surface-reduced CeO2 H2O gt gem OH groups
• active
  sites per Shido and Iwasawa (1993)

10
CONCLUSIONS

• Pt/CeO2 a possible catalyst for on-board
  reforming
• Matrix Reduction a useful technique for
  complicated systems
• PO first, then TO
• Small amount of catalyst before TO occurs
• Possibly use microchannel reactors in parallel
• SR, DR not reasonable reactions to consider in
  the overall process
• WGS depends upon the total amount of CeO2
  support present