Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Biogas reforming: conventional and reactive separation processes and the preparation and characterization of related materials
(USC Thesis Other)
Biogas reforming: conventional and reactive separation processes and the preparation and characterization of related materials
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Biogas Reforming: Conventional and Reactive Separation Processes and the Preparation and
Characterization of Related Materials
By
Sasan Dabir
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMICAL ENGINEERING)
May 2017
Copyright 2017 Sasan Dabir
i
Table of Contents
Dedication ..................................................................................................................................... iii
Acknowledgement ........................................................................................................................ iv
List of Figures ............................................................................................................................... vi
List of Tables ................................................................................................................................ xi
1. Introduction ........................................................................................................................... 1
1.1 Steam Reforming of Methane (SRM) ........................................................................... 2
1.2 Autothermal Reforming (ATR) .................................................................................... 4
1.3 Partial Oxidation ............................................................................................................ 5
1.4 CO2 (dry) Reforming ..................................................................................................... 5
1.5 Membrane Reactors ....................................................................................................... 8
2. A Feasibility Study of Biogas Reforming .......................................................................... 13
2.1 Introduction .................................................................................................................. 13
2.2 Experimental................................................................................................................. 21
2.3 Modeling ........................................................................................................................ 28
2.4 Results and Discussion ................................................................................................. 36
2.4.1 Reformer Performance ......................................................................................... 36
2.4.2 Engine Performance ............................................................................................. 42
2.5 Conclusions ................................................................................................................... 49
ii
3. Fabrication of Nanoporous SiC Membranes via Dip-Coating Technique .................... 52
3.1 Introduction .................................................................................................................. 52
3.2 Experimental................................................................................................................. 59
3.3 Results and Discussion ................................................................................................. 62
3.4 Conclusions ................................................................................................................... 74
4. Fabrication of Nanoporous SiC Membranes via Initiated Chemical Vapor Deposition
(iCVD) Technique ....................................................................................................................... 77
4.1 Introduction .................................................................................................................. 77
4.2 Previous Studies............................................................................................................ 83
4.3 Fabrication of Disk-shaped Nanoporous SiC Membranes ....................................... 85
4.3.1 Research Approach ............................................................................................... 85
4.3.2 Pyrolysis of Pre-ceramic Polymer ....................................................................... 88
4.3.3 Preparation of the Disk-Shaped SiC Substrate for iCVD ................................. 93
4.3.4 iCVD Process and Preparation of Nano-porous SiC Membrane ..................... 96
4.3.5 SiC Membrane Permeation Test ......................................................................... 98
4.3.6 Results and Discussion .......................................................................................... 99
4.3.7 Conclusions .......................................................................................................... 108
5. Suggestion for Future Work ............................................................................................ 109
Appendix .................................................................................................................................... 112
References:................................................................................................................................. 125
iii
Dedication
To my beloved father Bahram Dabir, my mother Maryam Dabir, and my brother Ashkan
for their unconditional love and support throughout all these years and to all who helped me in
this journey
iv
Acknowledgement
I would like to express my deepest gratitude and appreciation to my advisor and committee chair
Professor Theodore T. Tsotsis, who has the attitude and the substance of a genius, who is the reason
for all the motivations I had, and who persistently guided me through this long journey. Without
his help and guidance, it would not have been possible to achieve what I have now.
In addition, my special thank goes to Mr. Richard Prosser, who not only guided me through
challenges we faced during our research study at Santiago Canyon landfill but also taught me very
big lessons and helped me discover my capabilities that were beyond my imagination.
I would like to thank my committee members, Professor Thieo E. Hogen-Esch and Dr. Malancha
Gupta who kindly accompanied me through my qualifying exam and my final defense.
I would like to also thank my dear, Leila, my family, and my friends Dr. Wangxue Deng, Mingyuan
Cao, Mark De Luna, Dr. Xiaojie Yan, Devang Dasani, Soheil Soltani, and Dianrang Bai who
directly or indirectly helped me by all means.
I wish to thank our Mork Family department staff members: Mr. Andy Chen, Ms. Karen Woo, Mr.
Shokry Bastorous, Mr. Martin Olekszyk, Ms. Angeline Fugelso, Ms. Heather Alexander, Ms.
Aimee Bernard, and Ms. Laura Carlos and I would like to send my deepest gratitude to Ms. Tina
Silva who continually supported me and helped me throughout my study.
The support of the Energy Innovations Small Grant Program of the California Energy Commission
(Grant #: 57827A/13-10G) is gratefully acknowledged. We also express our gratitude to Orange
County Waste and Recycling for allowing the research to be performed at the Santiago Canyon
landfill, and in particular to Mr. Larry Adams for his kind support during the experiments at the
v
landfill. We also acknowledge Mr. Daniel Waineo, Dr. Jerry Ren, Mr. Kambiz Jozitehrani, Mr.
Dustin Stickney, Mr. Tommy Smith, and Ms. Farideh Kia for the many useful technical
discussions and their support during this project, and Mr. Don Wiggins and his team at USC for
their efforts in the construction of the pilot-scale system. And Finally, the financial support by the
United States Department of Energy, the National Science Foundation, and the Mork Family
Department of Chemical Engineering and Materials Science is gratefully acknowledged.
vi
List of Figures
Figure 1-1 Schematic view of the conventional multi-stage reforming process for hydrogen
production using through the SRM reaction (adapted from Iulianelli et. al. [27]). ........................ 3
Figure 1-2 Schematic view of the MR: a) co-current; b) counter-current flow (adapted from
Gallucci et. al. [23]). ....................................................................................................................... 9
Figure 2-1 CR process configuration that was filed-tested. .......................................................... 22
Figure 2-2 Schematic view of the experimental set-up. a) solid-view, b) transparent-view. ....... 24
Figure 2-3 Experimental conversions and calculated equilibrium conversions for different (S/C)
ratios as a function of Tex .............................................................................................................. 37
Figure 2-4 Experimental compositions and calculated equilibrium compositions as a function of
feed temperature TF for two different (S/C) ratios. Left, (S/C)=1.68, right (S/C)=2.33. Other
conditions such as feed flow rate and feed composition are listed in the pap .............................. 39
Figure 2-5 Calculated conversion vs. W/F for various steam to methane ratios. q V=1.6 kW ...... 39
Figure 2-6 Calculated H2 composition (dry basis) vs. W/F for various steam to methane ratios.
qV=1.6 kW ..................................................................................................................................... 40
Figure 2-7 Experimental conversions and calculated equilibrium conversions as a function of
feed temperature TF for two different feed flow rates of biogas. Other conditions are discussed in
the document. ................................................................................................................................ 41
Figure 2-8 The NOX (NO, NO2) emission data in the exhaust gas for the engine running on a
syngas/biogas mixture with 50% of the total flow of biogas going into the reformer. Three
different operating engine regimes: No load, 2 Amp, and 3 Amp load. ....................................... 44
Figure 2-9 The NOx (NO, NO2) emission data in the exhaust gas for the engine running on
propane for three different operating regimes: No load, 2 Amp, and 3 Amp load. ...................... 45
vii
Figure 2-10 Adiabatic flame temperature vs. the air-to-fuel equivalence ratio λ for various gas
feed compositions. ........................................................................................................................ 47
Figure 3-1 Structure of an idealized SiC membrane ..................................................................... 62
Figure 3-2 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes
employing supports made from a single 0.6 µm SiC starting powder via the use of a regular slip-
casting solution, prepared as described in the Exp. Sec. Measurement temperature 473 K;
pressure difference across the membrane 2.41×10
5
Pa; permeate side pressure 1 atm; each
individual data point reflects results of three different membranes prepared under identical
conditions. ..................................................................................................................................... 66
Figure 3-3 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes
employing supports made from a single 0.6 µm SiC starting powder via the use of the PS-slip-
casting solution, prepared as described in the Exp. Sec. Measurement temperature 473 K;
pressure difference across the membrane 2.41×10
5
Pa; permeate side pressure 1 atm; each
individual data point reflects results of three different membranes prepared under identical
conditions. ..................................................................................................................................... 67
Figure 3-4 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes
employing supports made from a blend of 50 wt.% 0.6 µm and 50 wt.% 6 µm SiC particles via
the use of a regular slip-casting solution, prepared as described in the Exp. Sec. Measurement
temperature 473 K; pressure difference across the membrane 2.41×10
5
Pa; permeate side
pressure 1 atm; each individual data point reflects results of three different membranes prepared
under identical conditions. ............................................................................................................ 68
Figure 3-5 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes
employing supports made from a blend of 50 wt.% 0.6 µm and 50 wt.% 6 µm SiC particles via
viii
the use of the PS-slip-casting solution, prepared as described in the Exp. Sec. Measurement
temperature 473 K; pressure difference across the membrane 2.41×10
5
Pa; permeate side
pressure 1 atm; each individual data point reflects results of three different membranes prepared
under identical conditions. ............................................................................................................ 69
Figure 3-6 Cross-sectional SEM images of SiC membranes employing supports made from a
single 0.6 µm SiC starting powder via the use of the PS-slip-casting solution, prepared as
described in the Exp. Sec., at different stages of preparation. ...................................................... 71
Figure 3-7 Top view SEM images of SiC membranes of SiC membranes employing supports
made from a single 0.6 µm SiC starting powder via the use of the PS-slip-casting solution,
prepared as described in the Exp. Sec., at different stages of preparation. ................................... 72
Figure 3-8 Cross-sectional SEM images of SiC membranes employing supports made from a
single 0.6 µm SiC starting powder via the use of the regular slip-casting solution, prepared as
described in the Exp. Sec., at different stages of preparation. ...................................................... 74
Figure 4-1 Schematic showing the mechanism associated with initiated chemical vapor
deposition (iCVD). I-I represents the initiator molecule and M represents the monomer. .......... 78
Figure 4-2 Silicon trenches coated using a) liquid-phase spin-coating versus b) iCVD [291]..... 80
Figure 4-3 Silicon trenches coated using a) iCVD versus b) pulsed plasma CVD [294]. ............ 81
Figure 4-4 Deposition rate of PPFDA onto silicon wafers as a function of a) substrate
temperature and b) PM/PSat [282]. ................................................................................................. 82
Figure 4-5 Chemical structures of silicon-containing monomer. ................................................. 88
Figure 4-6 a) Experimental setup for pyrolysis of the pre-ceramic polymer poly(1,3,5-tetravinyl-
1,3,5-tetramethylcyclotetrasiloxane) inside the DRIFTS cell, b) Collector II with the reaction
chamber and heating assembly ..................................................................................................... 91
ix
Figure 4-7 The pressure/temperature chart of the chamber. ......................................................... 92
Figure 4-8 Helium Compressibility Factor (Z) Vs. Pressure (P, psi/atm) for various temperatures
up to 300K [311]. .......................................................................................................................... 96
Figure 4-9 Membrane permeance test set-up [312]. ..................................................................... 98
Figure 4-10 The IR absorbance spectrum of alpha-alumina powder (DX Type (1µm particle
size), Electron Microscopy Sciences), diluted with KBr. ........................................................... 100
Figure 4-11 The IR absorbance spectrum of alpha alumina powder coated with pre-ceramic
polymer poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane), diluted with KBr. ............ 101
Figure 4-12 The IR absorbance spectrum of alpha alumina powder coated with pre-ceramic
polymer poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane) and pyrolyzed outside the
DRIFT, diluted with KBr. ........................................................................................................... 103
Figure 4-13 The change in IR absorbance spectrum of alpha alumina powder coated with pre-
ceramic polymer poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane), as being pyrolyzed
at different temperatures from room temperature to 1000 ºC, diluted with BaF 2 matrix. .......... 105
Figure 4-14 The IR absorbance spectra of alpha alumina powder coated with pre-ceramic
polymer poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane); a) room temperature, b)
pyrolyzed at 1000 ºC, both diluted with BaF2 matrix. ................................................................ 106
Figure A-1 a) The generator used to power-up the system, b) The generator used for the
experiments. ................................................................................................................................ 115
Figure A-2 The compressor driven by a 1 hp electrical motor. .................................................. 116
Figure A-3 Schematic view of the 60.96 cm (24”) long stainless steel reformer. ...................... 118
Figure A-4 a) Cross-flow heat exchanger b) Air-fin cooler with a fan. ..................................... 119
Figure A-5 Thermoelectric Peltier refrigeration cooling system. ............................................... 120
x
Figure A-6 The PLC unit with the total number of 8 cards connected to the HMI through
Ethernet card/cable and a Network Switch. ................................................................................ 122
Figure A-7 The HMI window panels. ......................................................................................... 122
Figure A-8 The system control panel. ......................................................................................... 123
Figure A-9 P&ID drawing of the system. ................................................................................... 124
xi
List of Tables
Table 1-1 Various methods used for H2 separation/purification (adapted from Iulianelli et al.
[27])................................................................................................................................................. 7
Table 2-1 Properties of the catalyst (JM-R44-PGM) .................................................................... 26
Table 2-2 The reaction rate expressions (adapted from Schouten and coworkers [218]) ............ 29
Table 2-3 Reactor properties used for the modeling ..................................................................... 35
Table 4-1 Weight Average Molecular Weight From GPC Measurements ([282]) ...................... 83
Table 4-2 Average Porosities of different support material.......................................................... 87
Table 4-3 He and H2 permeance results and ideal separation factors of He and H2 over Ar for
Slip-casted SiC support coated via iCVD, pyrolyzed at 1100ºC and tested at 35psig and 473K.
..................................................................................................................................................... 107
Table A-1 Cost estimation of a 848 SCMH (500 SCFM) biogas-to-energy project .................. 113
1
1. Introduction
Air pollution and global warming remain a challenging issue for the world today due to the
continuing high emissions of greenhouse gases (GHGs). The primary causes of the greenhouse
effect in the Earth’s atmosphere are CO2, CH4. CO2 and CH4 are thought to contribute from 9%–
26% and 4%–9%, respectively to the total effect and, therefore, minimizing the presence of these
gases in the atmosphere is a major concern and goal [1]. Introducing a method to recover and reuse
gases that are flared (e.g., in refineries, oil and natural gas extraction facilities, landfills and
municipal water plants, etc.) and removing/sequestering CO 2 from flue-gas emissions from power
generation are today a key focus. For example, the World Bank has estimated that 110 billion m
3
of associated gas (from oil and gas operations) is being flared annually resulting in huge quantities
of GHGs, mostly CH4 and CO2 that are vented into the atmosphere. Finding a way, for example,
to efficiently utilize such gas will help recover a huge amount of energy that is presently being
wasted, and will assist in controlling the level of GHGs that are exhausted into the atmosphere [2].
Along the same lines, a number of refineries are planning today to build co-generation plants in
which gas turbines burn refinery gases [3, 4] or a mixture of refinery gases and natural gas [3, 5]
to produce electricity [6-9].
Two other approaches for reducing CHG emissions include (i) CO 2 capture and sequestration
(CCS) [10-13] from the flue-gases of power plants burning fossil fuels, and (ii) replacing fossil
fuels altogether with clean-burning fuels like hydrogen. The production of hydrogen (H 2) [14-16],
in particular, is attracting quite a bit of attention today [17] (in a number of these studies [18],
including the one in this Thesis, hydrogen production is part of a co-generation scheme whereby
the waste heat in flue-gas is used as an energy source in the hydrogen production step). The reason
for the enhanced interest in H2 in recent years is because it is a clean fuel that does not generate
2
CO2 emissions during electricity production, e.g., when used as a fuel gas in polymer electrolyte
membrane (PEM) fuel cells (PEMFCs) [19] (very high purity H 2 is needed for this application
with a CO content less than 10
–4
%) or in gas turbines.
Hydrogen, as one of the most important energy carriers, produced in reforming processes has
attracted a lot of attention [17]. Hydrogen can be used in combustion engines or, at high purities
(the CO content in a fuel hydrogen containing gas less than 10
–4
%), in PEMFCs for generation of
electricity [19]. Hydrogen production is, therefore, a key focus of this Thesis, particularly from
renewable resources like biogas. Hydrogen use in PEM (and other types) of fuel cells for
distributed electricity production, e.g., for residential systems, and for emergency power
generation in hospitals and other municipal and industrial facilities is an area of high future growth.
In what follows, therefore, we review some of the current technologies that are used to produce
syngas (a mixture of H and CO) from which H2 can be separated from. Such technologies include
traditional steam reforming of methane (SRM), CO 2 (dry) reforming, or combination of both,
(which is known as mixed reforming), autothermal reforming (ATR) and partial oxidation,
1.1 Steam Reforming of Methane (SRM)
Most hydrogen is produced today by the steam reforming of methane (SRM), which produces a
syngas from which H2 is separated by cryogenic distillation, membrane separation or pressure-
swing adsorption (PSA). Despite its large energy consumption, SRM is still considered today as
the most attractive fuel processing technique for hydrogen and synthesis gas production [20-22].
In conventional SRM processes the reaction is carried out inside multi-tubular fixed-bed reactors.
High temperatures favor the process, due to the endothermicity of the SRM reaction. Methane
conversion >80% is typically achieved with H2O/ CH4 feed ratios in the range of 3-4 at
3
temperatures as high as 850°C. High pressures (ranging from 1 to 4 MPa) are also used in order to
improve the energy efficiency of the process [23].
Almost 50% of the global production of hydrogen (~ 55×10
6
tons/year) comes from natural gas
(which mainly contains methane) [24] (in addition, ~30% is supplied through oil/naphtha
reforming, 18% from coal gasification, 3.9% by water electrolysis, and 0.1% by other sources
[25]). SRM, as noted above, is the most common process for hydrogen generation from natural
gas, despite its negative impact on the climate through the emission of CO 2 [26]. Conversion of
natural gas to hydrogen is carried out in a conventional reformer (CR) –see above – which is
followed by water-gas shift (WGS) reactors (high temperature and low temperature WGS reactors)
to convert the CO in the syngas from the CR into H 2 and CO2, followed by other equipment for H2
separation (e.g., PSA), and further purification (e.g., preferential oxidation (PrOx) or methanation
reactors, etc.). A schematic of the multi-step reforming process is shown in Figure 1-1.
Figure 1-1 Schematic view of the conventional multi-stage reforming process for hydrogen production
using through the SRM reaction (adapted from Iulianelli et. al. [27]).
During SRM, methane and steam are fed into the CR, which typically contains a Ni-based catalyst,
at temperatures in the range of (800–1000°C) and pressures (14–20 bar), The following two
linearly independent reactions are known to place inside the CR.
CH4 + H2O ↔ CO +3H2 ΔH(298K) = 206 kJ/mol (R1)
4
CO + H2O ↔ CO2 + H2 ΔH(298K) = -41 kJ/mol (R2)
The reforming process produces a hydrogen-rich gas mixture with a significant CO content
(usually greater than 5 vol% [28]). Subsequent to the CR, two WGS reactors are used in series to
further reduce the CO content and to increase the hydrogen content of the reformed gas. The first
WGS reactor (high-temperature shift or HTS) is typically loaded with high-temperature WGS
catalyst such as Fe3O4/Cr2O3 with rapid kinetics, and usually operates at temperatures between 310
and 450°C. The HTS reactor helps to reduce the CO syngas content by ~50% [29, 30]. The gas
mixture then enters the second WGS reactor (low-temperature shift or LTS) operating between
180 and 250°C, which is loaded with a low-temperature catalyst such as Cu/ZnO/Al 2O3 [29, 31,
32]. At the exit of the second reactor, the CO concentration is decreased from ~3 vol% to ~500
ppm [27].
Despite its industrial maturity, studies on SRM are still continuing both at the academic and
industrial level, mostly focused on improving the catalyst performance (e.g., maximizing hydrogen
yield), its resistance to sintering, and minimizing the effect of carbon deposition and Sulphur
poisoning. Though Ni-based catalysts, are most commonly utilized because they are less expensive
and show reasonable performance [27, 33-37], a number of experimental studies have also been
done evaluating other, both noble and non-noble, metal-based catalysts and novel catalyst supports
[38-41].
1.2 Autothermal Reforming (ATR)
In autothermal reforming (ATR) , oxygen is added into the methane/steam feed mixture (in some
instances, as in this study, which focuses on biogas reforming, oxygen is naturally present in the
gas itself, see Ch. 2). The oxygen in the feed mixture oxidizes part of the methane, which then
5
helps to compensate for some of the endothermicity of the methane reforming reaction. One can,
in principle, add enough oxygen to provide all the endothermic heat of the reaction (thus, the name
autothermal), which then significantly simplifies reactor design.
The outlet temperature of the ATR reactor is, typically, between 950-1100℃ and the outlet
pressure can be as high as 100 bar. In addition to the aforementioned reactor simplicity, another
reported advantage of the ATR process is that the presence of O 2 provides an added degree of
control in adjusting the syngas (H2/CO) ratio, and this is particularly useful for producing certain
second-generation biofuels, such as DME which requires a 1:1 H 2:CO ratio. On the other hand,
the use of oxygen is associated with explosions and other safety concerns.
1.3 Partial Oxidation
Though not practiced commonly, partial oxidation (POX) which is the reaction of methane with
oxygen under strongly fuel-rich conditions can also produce a syngas. The advantage of POX is
that it is a non-catalytic reaction (homogeneous combustion) and it requires no external energy
input. The disadvantage is the very high temperatures required and the generation of other
pollutants (e.g., NOX).
1.4 CO2 (dry) Reforming
The reforming of gaseous fuels that contain large quantities of CO 2, such as sour natural gas with
a large CO2
content or biogas, which is the focus of this Thesis, is also known as dry reforming. It
has attracted increased attention in recent years, because of concerns about the contribution of CO 2
to global warming and renewed interest to utilize more effectively these low-BTU content fuels
[42]. One disadvantage with dry reforming is that syngas produced contains a higher content of
CO because of the enhanced reverse water gas shift reaction (RWGS)
6
CO2 + H2 ↔ H2O + CO
Another disadvantage is potential catalyst deactivation due to coke formation. Typically, Ni-based
catalysts [43-48] are utilized, but when the feed contains a large fraction of CO 2 fast deposition of
carbon on the catalyst has been reported in the process [45]. Deactivation and coke deposition
diminish significantly by the addition of steam to the feed (the process is also known as ‘mixed
reforming) [49-52] [48, 53-55]. The proper choice of catalyst and operating conditions has been
shown to have a positive impact on improving the conversion of methane, hydrogen yield and the
optimal (H2/CO) syngas ratio [55-58].
In the recent studies on the mixed reforming process, the main focus has been on catalyst
performance [59-62]. Active metals, such as noble metals [27, 63-66] and transition metals are all
used to prepare active catalysts. Ni-based catalysts still remain a common choice [57, 67-76].
Noble metal catalyst are pricier and their availability [77] is limited. However, they show higher
coke resistance compared to transition metals [78-84]. A noble-metal catalyst was chosen for the
investigations in this Thesis (see Chapter 2).
Following the reforming process of methane, various processes need to be employed for separation
of H2 from the rest of the syngas components, and for further purification to meet the purity
requirements for the specific process (e.g., for use in PEMFC). Some of these methods for H 2
separation and purification are listed in Table 1-1 along with their advantages and disadvantages.
7
Table 1-1 Various methods used for H 2 separation/purification (adapted from Iulianelli et al. [27]).
Cryogenic distillation, membrane separations and PSA are techniques commonly used for H 2
separation from syngas. Distillation is an effective technique, but it requires very low operating
temperature, and has a high energetic consumption [85]. For that reason, currently the PSA process
is the conventional technique that is used, and typically results in a high- purity hydrogen product.
However, significant hydrogen loss, nearly 20%, during the PSA operation is one of the big
disadvantages of this process [27]. In membrane separations, polymeric membranes are still the
common choice. However, these membranes have limited thermal and mechanical resistance, and
relatively high sensitivity to swelling and compaction, and that makes them less attractive for use
in hydrogen separation application. A key emphasis in this Thesis, therefore, is the development
of inorganic membranes, particularly nanoporous SiC membranes which overcome some of the
challenges polymeric membranes face for this application.
For application where highly pure H2 is required, one has to employ additional purification steps.
As can be seen in Table 1-1, the PrOx technique is one method by which we can remove CO
8
impurities in the hydrogen fuel for PEMFC down to the single-digit ppm level. In the PrOx reactor
selective oxidation of CO into CO2 takes place over noble metal supported catalysts (Pt, Ru, or Rh
on Al2O3). The use of the PrOx reactor makes the overall system more complex, in addition to the
fact that it requires adding pure oxygen to the syngas with its associated costs and safety concerns.
1.5 Membrane Reactors
Though SRM technology is pretty mature with many facilities presently operating worldwide, for
the application that we study here (hydrogen production from low-BTU content biogas) the ability
to employ the technology in a co-generation scheme and to be able to utilize waste heat that is
plentifully available in these processes is critical. There is need, therefore, to employ novel reactors
that operate efficiently at lower temperatures as this will make it easier to recoup the waste heat
energy that may be available. One such technology that may be utilized for such an application is
membrane reactors (MR), and a key focus, therefore, of this study is to develop and test the robust
materials (i.e., SiC nanoporous membranes) that can be used in such an application (see Chapters
3 and 4 detailing such efforts).
The use of membranes for hydrogen separation and purification has been discussed above, and the
need for developing robust membranes has been noted. Membrane reactors that employ reaction
and separation in one single unit, have also attracted considerable attention to date, in the context
of hydrogen production [86] and their performance has been bench-marked to that of conventional
reactors. MR have been proven a viable alternative to fixed-bed type conventional reactors, by
enhancing the conversion of the reforming reaction, and by delivering a pure hydrogen product
which is acceptable for use in most applications, other than those that require ppm-level of impurity
content.
9
Palladium (Pd) and Pd-alloy type membranes have been extensively studied in different membrane
reactor configurations for the SRM reaction, as well as to improve the conversion or selectivity of
other dehydrogenation or hydrogenation reactions [87, 88]. The combination of the reaction with
selective permeation of hydrogen through the membrane increases the conversion of the reaction
at constant temperature significantly for the case of the equilibrium-limited SRM reaction. Thus,
one can employ lower operating temperatures to achieve conversion levels which require much
higher temperatures of operation to achieve in traditional reactors (as noted above, this then allows
for a greater use waste heat in a co-generation environment). When employing Pd membranes (or
the high-quality SiC nanoporous membranes which are the focus of this Thesis), not only is the
equilibrium in the SRM reactor shifted towards the hydrogen-producing side, high-purity
hydrogen is also produced from the permeate side of the membrane reactor, see Figure 1-2.
Figure 1-2 Schematic view of the MR: a) co-current; b) counter-current flow (adapted from Gallucci et. al.
[23]).
10
In MR applications, Pd is usually used in a form of an alloy with other metals to increase the
mechanical strength, solubility of hydrogen and activity of the membrane towards hydrogen
dissociation (a key step preceding hydrogen atom transport through the membrane). Pd-Ag
membranes have been commonly used for purification of hydrogen [23]. Pd-Ru alloy membranes
also show very good stability and long-term resistance at high temperatures up to 550°C,
reportedly five times greater than that of pure Pd [87]. Unlike pure Pd membranes, which are
sensitive to poisoning and coking, Pd-Ru membranes have shown higher resistivity towards
poisoning resulting in a more stabilized hydrogen permeation rate during MR applications [89].
The research in the membrane reactors area has intensified with the development of inorganic
membranes suitable for high-temperature applications [90], like the Pd-alloy membranes discussed
above, and microporous membranes (like the SiC materials which are the focus of this Thesis).
The conversion of methane into hydrogen is a good potential application for MR as it is an
equilibrium limited reaction and because it is overall endothermic and thus increasing the
temperature increases the equilibrium conversion attained. This is the reason why most SRM
reactors operate at temperatures of 800°C or higher. Utilizing membrane reactors makes it possible
to reduce the operating temperature by shifting the SRM reaction towards hydrogen production
via the use of hydrogen permselective membranes such as those made from Pd alloys [90, 91], and
this has the beneficial effect of being able to use waste heat available to compensate for the reaction
endothermicity.
However, the broad use of Pd membranes in catalytic reactors for steam reforming of methane
[92] is hampered by their high cost [88, 90] and limited availability of the metal. These challenges
may be partially overcome with the fabrication of super-thin (1-2 µm) membranes [93] but robust
such membranes are presently lacking. (Pd-alloy membranes on the other hand may more
11
appropriate for micro-membrane reactors (MMR) where price and availability are not key
concerns. These MMR, employing planar microchannel, hollow-fiber, and monolithic
configurations, are a promising and fast developing area with both a scientific and technological
interest [94-103]).
This has then motivated the use of porous membranes for such an application like alumina [104,
105], but mostly microporous silica [106, 107] [108, 109]. However, silica membranes have been
proven unstable in the presence of high-temperature steam, which makes the use of such
membranes for the SRM application rather questionable. The SiC microporous membranes, which
are the focus of this Thesis, have many of the same excellent characteristics with the silica
membranes (high permeability and permselectivity towards hydrogen), but on the other hand are
quite stable to high temperature steam.
As noted previously, the key goal of this Thesis is to investigate the production of hydrogen from
low-BTU gaseous fuels, specifically biogas. There has been growing attention recently towards
hydrogen generation from renewable and bio-derived sources such as biogas, in which methane is
the main product [56, 110-124]. The use of such fuels as natural gas substitutes has the potential
to significantly reduce GHGs emissions. Chapter 2 of this Thesis describes a field study, whose
objective was to validate the use of SRM technology in a co-generation environment employing
low-BTU content biogas. Such biogas produced from the anaerobic digestion of municipal solid
waste is composed of methane and carbon dioxide, but also in addition of trace amounts of H 2S,
NH3, hydrogen, nitrogen, oxygen and water vapor [125] as well as numerous other impurities
known as non-methane organic compounds (NMOC) containing various heteroatoms (e.g., various
halides, Sulphur, silicon, etc.). The presence of such compounds presents challenges for the SRM
catalyst and processes which are detailed in Chapter 2.
12
The remaining two Chapters of the Thesis focus on the use of membrane reactors for the proposed
biogas into hydrogen application. As noted above, the use of MR for this application shows
excellent promise, because of the ability of such reactors to operate efficiently at much lower
temperatures than conventional SRM reactors. Specifically, the focus in this Thesis is the
preparation and characterization of the critical materials, i.e., robust SiC nanoporous membranes
needed for the proposed application. We conclude with a Summary and Future Work Chapter.
13
2. A Feasibility Study of Biogas Reforming
2.1 Introduction
Lean, premixed fuel-gas combustion has been receiving increased attention recently because of its
ability to reduce emissions of nitrogen oxides (NO X). The reason for the reduction in emissions is
because under lean-burn conditions, peak flame temperatures are reduced, and the thermal NO X
formation rate decreases because it is strongly temperature dependent [126-129]; and, also,
because for a fixed fuel energy input the overall combustor volume flow rate increases, which
reduces the residence time available for NOX formation[130]. Using a mixture that is highly fuel-
lean, on the other hand, may result, potentially, in incomplete/unstable combustion, which may
then increase the concentration of carbon monoxide (CO) and unburned hydrocarbons (UHC) in
the exhaust gas. It will also reduce the engine energy efficiency. Maintaining good combustion
stability in highly fuel-lean, premixed gas combustion is, therefore, a challenging technical task.
A number of studies, in recent years, [131] including the work performed by this team[128, 129]
have reported that addition of hydrogen (H2) to the fuel mixture can enhance combustion stability,
thus allowing the mixture to burn leaner, and as a result reducing the emissions of NO X. However,
the economics of hydrogen production and utilization during power generation still remain a key
challenge.
The most common approach to produce a hydrogen-rich syngas feedstock, that could potentially
be used to enhance fuel-lean combustion, is via the catalytic steam reforming of methane (CH 4),
the key component of natural gas (NG), described, typically, by the following two independent
reactions (reaction R2 is also known as the water-gas shift step):
14
CH4 + H2O 3H2 + CO H298 = 206 KJ/mol (R1)
CO + H2O H2 +CO2 H298 = -41 KJ/mol (R2)
Ni-based supported catalysts are commonly utilized, however, in recent years more costly noble
metal type catalysts have also found use[132] because of their enhanced activity and stability. Such
catalysts are of particular value when reforming gas mixtures, like biogas, containing large
fractions of CO2, as is the case in this study. A key challenge with catalytic steam reforming is that
it is an energy-intensive process.[128, 133] One way to improve the energetics (and the economics)
of the process is to improve conversion efficiency, by employing various novel reactive separation
technologies (e.g., membrane reactors or adsorptive reactors) that help to overcome the
thermodynamic constraints of the reforming reactions. Increased conversion and H 2 yield are
among the reported benefits of using such processes, which, however, are still at an early
developmental stage [134, 135]. (Other potential methods to produce H 2 for use during power
generation include electrolysis which has been extensively developed in recent years, and direct
photocatalytic or thermochemical water splitting, the latter processes being still under
development. However, the economics of these approaches are not competitive, at present, with
conventional steam reforming. Gasification of waste biomass is currently studied by a number of
groups, including this team, [136] as a method to improve the economics of H 2 production by
utilizing what is, in principle, a zero-cost renewable resource. Though progress has been made, the
process is still hampered by high capital and operational costs[137, 138].)
Another means to improve the energetics of hydrogen production during power generation is via
the employment of combined heat and power (CHP) advanced cycle concepts that center around
the idea of waste-heat recovery in the engine exhaust. An interesting such concept, known as the
15
chemical recuperation (CR) cycle, which is the focus of the present project, was first proposed in
the late nineties [139-141] , and has been studied since. It involves the use of “waste-heat” from
the combustor exhaust to produce from NG, via catalytic reforming, a H 2-rich syngas (SG) mixture
(containing, in addition, CO, CO2 and unreacted CH4) which is then used for power generation. In
past (mostly bench-scale) investigations the application of the CR concept to NG combustion has
shown promise for attaining ultra-low NO X emission levels, reduced CO emissions, and significant
improvements in operability.
In this study the focus is on power generation from biogas, produced by the anaerobic digestion of
organic waste materials in landfills or in digesters. Such biogas, typically, contains CH 4 diluted
with CO2, N2, and smaller concentrations of O2, and shows promise for use as a renewable fuel, in
place of NG, to produce electricity. In California, for example, the law mandates that by the year
2020 one third of the State’s electricity must be from renewable sources, and 50% by the year
2030, and as a result biogas-based power generation is attracting renewed attention. Because of
concerns with NOX emissions, lean-fuel air combustion is also employed when using biogas as
fuel. As noted in the case for NG, lean-burn combustion conditions result in relatively low flame
temperatures that do not favor NOX formation, but lead, on the other hand, to poor combustion
stability and reduced engine power output. The problem is exacerbated for biogas, as it contains
diluents like CO2 and N2 that negatively impact combustion stability. The addition of H 2 with its
potentially favorable impact on flame stability shows, therefore, good promise for application in
“biogas-to-energy” processes [142, 143]. Hydrogen production/utilization during power
generation from biogas via the CR cycle is, therefore, a key focus of this study. Our particular
interest is in low-BTU content biogas, a cheap (effectively zero-value, as it is today mostly being
16
flared) and readily available renewable fuel, and our approach aims to enhance its energetic
content, and thus make it possible to generate electricity from it in an energy-efficient manner.
To the best of our knowledge, prior to this study, the CR cycle had not been tested for power
generation from biogas, particularly the low-BTU content gas that is the focus of this investigation.
Such gas contains, besides CH4, substantial quantities of CO2 and N2, and numerous trace
impurities known as non-methane organic compounds (NMOC), which contain heteroatoms like
sulfur, silicon (siloxanes), and various halides. Catalytic reforming of CH 4 in the presence of
substantial quantities of CO2 (as in biogas) is also known as “dry reforming”, and presents unique
challenges to catalyst and reactor design. The NMOC are a significant hurdle as well, as they are
a potential threat to catalyst performance. Ni-based reforming catalysts are the common industrial
choice, because of their good activity and relatively low cost, when compared to noble metal-based
catalysts [144, 145]. Much research has been performed on Ni-based catalysts, and their reaction
mechanism and phenomena like catalyst deactivation due to coking and poisoning by sulfur
impurities are well understood [51, 146-150]. Dry-reforming of CH 4 is quite challenging, however,
for Ni-based catalysts. Li and co-workers [51], for example, have found a higher propensity
towards coke formation. Ni-based catalysts are, nevertheless, able to function in that environment
by appropriately adjusting the (H2O/CH4) feed ratio or the temperature of operation; however,
noble metal catalysts are better suited to function in such an environment, particularly one that
also contains small, albeit finite, concentrations of O 2, and have, as a result, been used in this
investigation.
The combustion of NG and SG mixtures has received attention, particularly in recent years with
the advancement of the integrated gas combined cycle (IGCC) concept for the production of
electricity from biomass and coal [151-153]. In one of the earliest studies, this team[128, 129, 154]
17
in fundamental studies in a single jet-wall configuration reported that adding H 2 or CO individually
to CH4 increased flame stability and resistance to extinction, but addition of H 2O and CO2 had the
opposite effect. (Similar observations were reported by numerous studies since. [155-173] Delatin
[174, 175], who studied wet and dry SG combustion in a microturbine and compared it to NG
combustion, report, for example, lower blow-off limits with no noticeable changes in pollutant
production). In fuel-lean flames, for a constant (CH 4/air) ratio, adding H2 (or CO) increases NOX
emissions, but adding CO2 and H2O decreases NOX.[129] However, during combustion of
(H2/CH4) mixtures, when the fraction of H2 in the fuel mixture increased [128, 129, 154], while
the total equivalence ratio (Φtot) stayed constant, little change in NOX emissions was observed (this
also agrees with the observations of Coppens [176] of H 2/CH4/air non-stretched flames stabilized
on a perforated plate burner); NOX emissions, on the other hand, substantially increased (for
constant fraction of H2 in the fuel) as Φtot increased. For (H2/CH4) flames with the same laminar
flame speed (taken as a measure of flame stability), those with a higher H 2 content show lower
NOX emissions (for the same H2 content, flames with higher flame speeds show higher NOX
emissions) [128, 129, 154]. For flames with the same maximum temperature (T max, taken as a
measure of the flame’s power output), H2 addition shows a small beneficial effect, particularly for
flames with higher Tmax [128, 129, 154]. (Similar observations are reported by Jeong et al. [142]
who studied the combustion of biogas/hydrogen blends in a spark-ignited (SI) engine under fuel-
lean conditions. They defined an efficiency per NOX emissions ratio (EPN) to describe the
relationship between the generating efficiency and NOX emissions. It was shown to be maximum
at a H2 concentration of 15%).
For flames with the same laminar flame speed, Ren et al. [128, 129, 154] reported that CO addition
also has a beneficial effect (albeit smaller than that of H 2), but for flames with the same Tmax,
18
addition of CO increases NOX. For flames with the same laminar flame speed or T max (less so for
the latter flames), CO2 addition increases NOX emissions (similar observations were made by
Fackler et al. [177, 178] who studied CH4/CO2 flames in a jet-stirred reactor (JSR) with the same
combustion temperature). For flames with the same laminar flame speed, H 2O addition increases
NOX emissions, though not as much as CO2 does. For flames with the same T max, on the other
hand, H2O addition decreases NOX emissions, the effect being even stronger than that of H 2.
Similar observations were made by Goke et al. [179] who found that steam dilution is very
effective for NOX reduction in NG/H2 flames and for preventing flashback; and by Lee et al.[180],
who studied the combustion performance in a turbine of SG composed of H 2/CO/N2/CO2/steam
and reported that NOX emissions per unit power generated decreased as the amount of SG diluents
increased.
By keeping the initial (CH4/air) ratio (Φin) constant and then assuming that a fraction R1 of the
CH4 is directed to a catalytic reformer with the reformate mixture (including unreacted CH 4 and
H2O) exiting the reactor being blended back with the unreacted (1-R1) CH 4 fraction, Ren et al.
[128, 129, 154] observed lower NOX emissions for sufficiently high R1 (though for a fixed
incomplete reactor conversion, for real high R1 flame stability begins to suffer – this can be
remedied, however, by removing the unreacted H2O). For flames with the same laminar flame
speed or Tmax and various R1, increasing the R1 results in decreased NO X emissions. Ren et al.
[128, 129, 154] investigated the mechanistic causes of the observed behavior for flames with the
same Tmax. They reported that the impact of H2 addition for such flames can be explained by a
reduction of NO formation by the prompt mechanism, which agrees also with subsequent
observations by others.[181, 182] The impact of CO (increase in NO X) can be attributed to an
increase of NOX formation through the N2O mechanism. For CO2 addition, since richer flames are
19
needed to maintain the same Tmax, this results in an increase in the prompt mechanism. For H2O
addition, though richer flames are required, and thus the rate of the prompt mechanism is higher,
the rates of the thermal (Zeldovich) and N2O mechanisms are lower because the lower
concentration of oxygen radicals which are consumed by the H 2O+O→2OH reaction (similar
observations were since reported by Giles [182]. For further discussion about the mechanistic
causes of NOX formation during combustion of SG/biogas mixtures, see, e.g..[183]).
Alavandi and Agrawal [184] studied experimentally an equimolar mixture (CO/H 2), in which CH4
was blended, in a 2-zone porous burner. For the same adiabatic temperature, increasing the
(H2/CO) content in the fuel lowered the CO and NOX emissions. Arrieta and Amell [185] studied
mixtures of CH4 with simulated SG from coal/biomass gasification in a porous burner under fuel-
lean conditions. For the same air/fuel ratio and thermal input, mixing the SG into the CH 4 resulted
in lower CO but higher NOX emissions. Arroyo et al.[186, 187] studied the combustion of SG with
compositions relevant to biogas reforming in a SI engine. SG with the highest H 2 content resulted
in the highest NOX emissions. Azimov et al. [188] in their engine study of SG combustion reached
similar conclusions: Increased H2 content led to higher combustion temperatures and efficiency,
lower CO and HC but higher NOX emissions. Increased CO2 content influenced performance and
emissions only when it reached a certain level, with thermal efficiency, and NO X emissions
decreasing only with CO2 content past that level. In a numerical study of premixed CH 4/CO/air
flames under fuel-lean conditions, when maintaining the Φ tot constant while adding CO to the fuel,
Chen et al. [189] noted a remarkable reduction in NO X emissions. When studying the combustion
of H2–CO mixtures blended with CO2 in a SI engine under fuel-lean condition Chen et al. [190]
reported that CO2 dilution resulted in a remarkable decrease in NOX emissions with little decrease
in brake thermal efficiency.
20
In their study of H2/CO/CH4/air opposed-jet flames Cheng et al. [191] report that H 2 addition
increases the laminar flame speed, as have also numerous other investigators.[160, 164, 192-203]
This is mainly due to the high reactivity of H2 leading to high production rate of H and OH
radicals.[204-207] Das et al. [208] experimentally studied the effect of H 2O on the laminar flame
speeds of moist H2/CO/air mixtures using the counterflow twin-flame configuration under fuel-
lean conditions. They report that the laminar flame speed varies non-monotonically for CO-rich
mixtures, first increasing, reaching a maximum value, and then decreasing with H 2O addition
(similar non-monotonic impact was also recently reported by others [209, 210]). In contrast, for
H2-rich mixtures the laminar flame speed monotonically decreases with increasing H 2O addition
(with similar behavior also reported by Singh et al.[209]). The effect of CO content on the laminar
burning velocity of SG was studied by He et al.[211], who reported it to increase monotonically
with it. Hinton and Stone[212] report that addition of CO2 to CH4 decreases the laminar burning
velocity of such mixtures, with similar observations reported by numerous other investigators[201,
202, 213-216] attributed to the active participation of CO 2 in the chemical reactions through the
intermediate step CO + OH ↔ CO2 + H. [217]
In summary, several prior (mostly lab-scale) studies have reported enhanced combustion stability
and in some instances improved pollutant emissions when employing reformate products during
NG combustion. We know of no prior study, however, where the operation of a practical device
(i.e., an internal combustion (IC) engine) operating on a mixture of real biogas and reformate
products is documented. This is, to the best of our knowledge, the first time such data are being
reported. Specifically, the goal of this research was to evaluate the technical and economic
feasibility of using reforming products during biogas combustion in a lean-burn gas engine in order
to reduce NOX emissions and to enhance combustion stability. The project involved both
21
experiments and related modeling for data analysis and interpretation, and for preliminary process
design and economic evaluation. The project results/outcomes are described below:
2.2 Experimental
The experimental study took place in its entirety at a field-site, specifically at the Santiago Canyon
landfill in Southern California. The experimental system design was based on the CR concept
previously outlined. There are a number of different CR process configurations one could
potentially employ to implement the proposed technology. In this “proof-of-concept” effort,
attention was focused on the most straightforward of such processes, shown in Figure 2-1. In this
process, a portion P1 of the biogas available for power generation (P1=biogas used for
reforming/total biogas) is fed into the catalytic reforming reactor in order to generate the synthesis
gas (the hydrogen-rich product mixture of reactions R1 and R2 above). In this approach, heat
recuperation takes place via a conventional heat exchanger step, and only pure biogas (plus steam)
is fed into the reforming reactor, and the synthesis gas produced is mixed with the rest of the biogas
(1-P1) in order to serve as the fuel for the engine.
Another approach that can be utilized to implement the heat recuperation step (which, however,
was not investigated in this study, as it went beyond the scope and available resources) is to employ
two engines (both operating on biogas), one that runs under fuel-rich conditions and the other
under lean-burn conditions. In this configuration, the flue-gas from the rich-burn engine is mixed
directly with the P1 portion of the biogas used for the lean-burn engine prior to feeding it into the
reformer. The remaining portion of the biogas (1-P1) together with the synthesis gas product of
the reforming reaction is then used for power generation in the lean-burn engine. In that case, the
rich-burn engine exhaust’s energy is directly recuperated by mixing it with the biogas, thus not
requiring the use of heat exchangers, and not incurring the unavoidable heat losses. This second
22
approach, however, though advantageous from a heat-recuperation standpoint, presents unique
challenges for catalyst selection and catalytic reactor design, which to the best of our knowledge
have not been previously addressed.
Figure 2-1 CR process configuration that was filed-tested.
The proposed process (Figure 2-1) provides for plentiful opportunities for waste heat recuperation.
During the field-testing, for example, the waste heat in the engine exhaust was used to pre-heat the
biogas prior to flowing it into the catalytic reformer to promote the catalytic reaction, as well as to
heat the water in the boiler to produce steam that is used as a reactant in the steam reforming
reactions. The energy content of the products of the reformer was another source of waste heat
utilized to boil the water to produce steam. It should be noted, however, that it was not the objective
of this early-phase study to maximize/optimize waste heat utilization, a task that will be a key goal
of a follow-up investigation.
Ceramic
Heaters
P1
Steam
Biogas C H 4, C O 2
1-P1
Reforming
Reactor
Engine
Engine
Exhaus
t
Water
23
For the studies, a pilot-scale set-up was constructed and installed at the Santiago Canyon landfill
(see Figure 2-2 for a schematic of the system-- the P&ID drawing of the process can be found in
the Apendix). It consists of different pieces of equipment (e.g., the engine, the compressor, various
sensors and analytical equipment, etc.) that were purchased from outside vendors, and various
other components (the reformer, the heat-exchanger, the boiler, etc.) which were designed and
custom-fabricated at the machine shop at the University of Southern California (photographs and
brief descriptions of all system components can be found in the Appendix). The system was sized
for a maximum flow of ~5 standard cubic meters per hour (SCMH) (3 SCFM) of landfill gas
(LFG), with the reformer operating in the pressure range of 1.36-1.70 bar (5-10 psig), with a
maximum boiler pressure of ~2 bar (15 psig). There is no “biogas-to-energy” system presently
installed at the Santiago Canyon landfill, so the biogas that is produced there is collected and then
combusted via the use of three flares, each fed by a separate blower that delivers the LFG. To
provide the biogas to the pilot-scale set-up, a line was connected to the exit pipe of one of the
blowers (blower #1); the discharge pipe from the set-up was connected to the flare-stack #1. The
system discharge that consists of the back-pressure regulator discharge flow and outlet stream of
the sensor box was recycled through a line connected to the suction point of blower #1.
a)
24
b)
Figure 2-2 Schematic view of the experimental set-up. a) solid-view, b) transparent-view.
25
The LFG at the Santiago Canyon site contains about 30% CO 2, 30-35% CH4, 1-2.5% O2, balanced
with N2 and water vapor. There are also small amounts of other compounds (<1%), that include
H2, NMOC, including siloxanes, and trace amounts of sulfur-containing inorganic compounds
(primarily H2S). Before the biogas could be fed to the system, these contaminants had to be
removed. For that, four adsorbing columns were used, two columns upstream of the compressor
and two columns downstream. The adsorbing columns upstream of the compressor, the first
containing DARCO activated carbon (AC) and the other silica gel (SiG), were used to remove the
H2S and water vapor, respectively from the LFG before the gas was fed into the compressor. The
two adsorbing columns downstream of the compressor, also containing AC and SiG were used to
remove the remaining contaminants as well as any condensed water after the gas was cooled.
The total flow rate of the gas, as well as of the portion of the gas going into the reformer were
measured via the use of an orifice plate and a differential pressure transducer (DPT) that measures
the pressure drop across the orifice, which is then used to calculate the volumetric flow rate. The
pressure of the system was regulated using a back-pressure regulator (BPR) downstream of the
system where the dried reformate gas was connected to the engine. In order to provide the steam
for the reformer, a boiler was used, the pressure of which was set to values between 1.7-1.9 bar
(10-13 psig), with the maximum designed pressure being ~2 bar (15 psig); the steam flow rate into
the reformer was measured using another orifice plate and a DPT. The flow rate of the steam was
adjusted to the value required to provide the selected steam/methane ratio for a given experiment.
After the LFG was mixed with the steam flow, the mixture entered the first-phase, cross-flow type
heat exchanger in order for its temperature to be raised to 260-282.2 C (500-540℉ )(before
entering the reformer) using the exhaust gas of the 7.5 kW power generator as the heating fluid.
26
Once the flowing mixture was preheated to the desired temperature, it would enter the catalytic
reformer, which was built from a 60.96 cm (24 in) long stainless steel pipe (7.62cm (3 in) ID, 8.89
cm (3.5 in) OD, Schedule 40). The bottom half of the reformer was packed with two different sizes
of stainless steel balls (5.6 mm (7/32 in) and 7.9 mm (5/16 in)) and served as the pre-heating
section. The top-half of the reactor was packed with Pt/Al 2O3 catalyst pellets (JM-R44 PGM,
purchased from Johnson Matthey, Inc.) and served as the reaction zone. The properties of the
catalyst, as provided by the manufacturer, are listed in Table 2-1.
Table 2-1 Properties of the catalyst (JM-R44-PGM)
Diameter of the pellet 1.8 mm
Density ~1 gr/ml
Surface area of the catalyst support ~120 m
2
/gr
Thickness of the catalyst layer
(Coated on the surface of the Alumina (Al 2O3) support at 20%wt
cat./wt. pellet)
Approximately
100 m
Two K-type thermocouples were inserted in a thermo-well, made from a 6.35 mm (1/4 in) stainless
steel tubing inserted axially in the reactor; their tips were placed at a distance of 15.24 cm (6 in)
and 30.48 cm (12 in) from the entrance of the reactor. One additional thermocouple was inserted
from the top of the reactor, its tip placed precisely at the exit of the reactor. The reactor was heated
from the outside using four high-temperature ceramic heaters, nominally 0.8 KW each, completely
surrounding it. The two bottom heaters were installed in parallel and were controlled by the system
control panel through a silicon-controlled rectifier (SiCR); the two top heaters, installed in parallel
again, were controlled by the system control panel through another separate SiCR. The
temperatures of the reformer were monitored and recorded in real time by the system control panel.
The outer surface of the reformer was completely insulated using high-temperature insulation
(Further details about the reformer are provided in the Appendix).
27
Inside the reactor, reactions R1 and R2 take place. Because of the presence of a small fraction of
oxygen in the LFG, combustion reactions of methane (R3) or of H 2 and CO (R4 and R5 in the
modeling Sec. 2.3) may also take place with their heat of the reaction being recuperated by the
endothermic reaction R1).
CH
4
+2O
2
→CO
2
+2H
2
O (R3)
The reformer reaction products, after exiting the reactor, would pass through a coil inside the boiler
in order to recoup their heat to produce the steam for the reformer. The gases exiting the boiler
would then flow into an air-fin cooler equipped with a fan to cool down the gas, mainly by
removing the latent heat from the water vapor and condensing it into liquid water, which was then
removed using a manual valve on the air-fin cooler as well as a water knock-out system that would
trap the water and drain it automatically. The gas product downstream of the air-fin cooler had an
approximate temperature of 37.8 C (100℉). (Note, that the use of the air-fin cooler was one of
convenience to limit the system complexity and its capital costs; in a pre-commercial or
commercial-scale system, this effluent could be used for additional heat recuperation, e.g., for pre-
heating the LFG or the water going into the boiler). Once the water was removed from the gas
product through a water trap and an additional adsorbing column, the mixture, now as a fuel, would
either go directly into the engine or would be blended with the by-pass flow of the raw LFG first
and would then be sent to the engine.
The composition of the various components was measured using a sensor block which contained
five different sensors for CH4, CO2, O2, nitrogen monoxide (NO), and nitrogen dioxide (NO2). A
portable analyzer was also used to measure the composition of H 2 in the dry gas mixture. (For
more information, please refer to the Sensors and Analyzers part of the Appendix, and Figure A5.
28
As can be seen from the P&ID drawing of the system, Figure A9, the various gases were sampled
using a total of four solenoid valves labeled as SOV-01, SOV-11, SOV-31, and SOV-32, from
upstream of the reformer, downstream of the reformer, after the blending point, and at the exhaust
of the engine respectively). All conversion values presented in the Figures are molar-based for
both the calculated equilibrium and experimental values, and are based on the feed molar flow rate
of methane.
2.3 Modeling
As part of the study, the catalytic packed-bed reformer was also simulated in order to gain insight
into the observed experimental behavior and to further guide the on-site biogas reforming
experiments. Since this team signed a non-testing agreement (NTA) with the catalyst
manufacturer, and also a non-disclosure agreement (NDA), for the simulations in this paper the
true experimental reaction rate expressions are not utilized, but the rates for the three chemical
reactions (R1, R2, R3) shown in Table 2-2, from the technical literature[218] are used, instead.
29
Table 2-2 The reaction rate expressions (adapted from Schouten and coworkers [218])
Reaction Rate expression
R1: CH4 + H2O 3H2 + CO
r
1
=
K
1
(p
CH4
-p
H
2
3
p
CO
/K
eq,1
)
p
CH4
m
1
p
H
2
O
n
1
a
R2: CO + H2O H2 +CO2
r
2
=
K
2
(p
CO
-p
H
2
p
CO
2
/K
eq,2
)
p
CH4
m
2
p
H
2
O
n
2
a
R3: CH
+ 2O
→ CO
+ 2H
O
r
3
=
k
31
p
CH4
p
O
2
(1+k
CH4
ad
p
CH4
+k
O
2
ad
p
O
2
)
2
+
k
32
p
CH4
p
O
2
(1+k
CH4
ad
p
CH4
+k
O
2
ad
p
O
2
)
a
m 1=0.0, n 1=0.596, m 2=0.0, n 2=0.0, as obtained by Numaguchi and Kikuchi [219].
A pseudo-homogeneous packed-bed reactor (PBR) model is used, in which the reaction rate is
expressed per unit mass of catalyst in the bed, and in which diffusion inside the catalyst pellets is
not explicitly accounted for. This assumption is consistent with the “egg-shell” type catalyst used,
with the active noble metal embedded in a thin skin on the top of a non-porous spherical pellet, so
that one can assume that the reactions only occur on the surface of the pellet. With no additional
information about the properties of the active layer or the properties of the inner solid core (because
of the NTA with the catalyst manufacturer, no attempt was made to measure these properties), the
model also does not account for intraparticle temperature gradients. In addition to not explicitly
accounting for the intraparticle concentration and temperature gradients, the pseudo-homogeneous
model also does not explicitly account for external mass and heat transfer limitations. The gas
phase reformate mixture was, for simplicity, assumed to obey the ideal gas law, which was shown,
based on the calculations of compressibility factors, not to be a bad assumption given that the
reformer was operating at near atmospheric pressure and high-temperature conditions.
In the PBR model, the axial dispersion of mass is neglected, as suggested by Schouten and
coworkers.[218] The equations solved are detailed below:
The mass balance equation for the pseudo-homogeneous PBR 1-D model is given as
30
dF
i
dV
= 1-ϵ
bed
)ρ
s
R
i
(2-1)
where F
i
(mol/s) is the molar flow rate of species i, V (m
3
) the reactor volume variable, ϵ
bed
the
void fraction of the PBR, which was measured to be equal to 0.43, ρ
s
the density of the catalyst
(kg-cat/m
3
), and Ri the rate of reaction for species i (i=CH4, CO, H2O, CO2, H2, O2, N2) in the PBR
(mol/kg-cat s) described by the following equation
R
i
= ∑ α
ij
r
j
(2-2)
where α
ij
is the stoichiometric coefficient of species i in reaction j (j=1, for the methane steam
reforming reaction, j=2 for the water gas shift reaction, and j=3 for the CH 4 oxidation reaction –
negative for reactants, positive for products, and 0 for inerts like N 2), and r
j
(mol/kg-cat s) is the
rate of reaction j, described by the expressions in Table 1, with the partial pressure P
i
(bar) for
each species i described by the following Eqn.
P
i
=
F
i
∑ F
i
P (2-3)
where P (bar) is the total reactor pressure. The volumetric flow rate V
(m
3
/s), along the axial (Z)
direction, in the PBR is described as:
V
=
R
g
T
P
∑ F
i
(2-4)
where Rg (8.3144x10
-5
m
3
∙ bar/mol∙K) is the gas constant, and T (K) is the reactor temperature.
The superficial fluid velocity u (m/s) is defined as follows
u=
V
A
c
(2-5)
31
where (m
2
) is the reactor cross-sectional area.
The pressure drop in the PBR is described by the Ergun equation as follows: [220]
-
dP
dV
=
1-ϵ
bed
)G
A
c
ρd
p
ϵ
bed
3
[150
1-ϵ
bed
)μ
mix
d
p
+1.75G]=
1-ϵ
bed
)u
A
c
d
p
ϵ
bed
3
[150
1-ϵ
bed
)μ
mix
d
p
+1.75G] (2-6)
where ρ is the fluid density (kg/m
3
), d
p
(m) the diameter of the catalyst pellets, μ
mix
the fluid phase
viscosity for gas mixture (kg/m s) and G (kg/(m
2
s) the superficial mass velocity given by the
following Eqn. as
G=uρ=u
f
ρ
f
(2-7)
where subscript f signifies the feed conditions. The viscosity of the gas mixture μ
mix
in Eqn. 2-6 is
calculated using the Wilke’s equation: [221]
μ
mix
= ∑
η
i
∑ Φ
ij
n
j=1
n
i=1
(2-8)
where Φ
ij
=
1+ η
i
η
j
0.5
Mw
j
Mw
i
0.25
2
√8 1+(
Mw
i
Mw
j
) 0.5
.
In Eqn. 2-8 above, xi and xj are the mole fractions for species i and j, M wi and Mwj are their
molecular weights (g/mol), and η
i
and η
j
are the corresponding single-gas viscosities (Pa s)
calculated from Eqn. 2-9 below. [222]
η=
AT
r
2
1+0.36T
r
(T
r
-1) 1/6
1+270(μ
r
0
)
4
T
r
+270(μ
r
0
)
4
(2-9)
32
where η is the viscosity for single gas (Pa s), Tr is the reduced temperature, A is a constant
determined by group contribution methods, and μ
r
0
the modified reduced dipole moment which is
calculated from:
μ
r
0
=52.46μ
r
(2-10)
μ
r
= μ)
2
P
c
) T
c
)
-2
(2-11)
Where μ is the dipole moment (debyes), P c (bar) is the critical pressure, and Tc (K) is the critical
temperature.
The heat balance equation is as follows:
( ∑ F
i
C
p,i
n
i=1
)
dT
dV
=Q
V
+ 1-ϵ
bed
)ρ
s
∑ -r
j
∆H
T.j
3
j=1
(2-12)
In the above Eqn. 2-12, Q
V
is the heat per unit time per unit volume (kW/m
3
) delivered to the
reactor from the heating elements, C
p,i
(kJ/(mol K) is the specific molar heat capacity of species
i, described by the following Eqn.
C
p,i
=Ai+BiT+CiT
2
+DiT
3
(2-13)
where the values of the individual coefficients are taken from Elliott and Lira.[223]
∆H
T.j
(KJ/mol) is the heat of reaction j described by the following Eqn.
∆H
T.j
=∆H
R,j
298.16
+ ∆C
p,i
jdT
T
298.16
(2-14)
where ∆H
R,j
298.16
(kJ/mol) is the standard heat of reaction at reference conditions (298.16 K) and
∆C
pj
is described by
33
∆C
pj
= ∑ α
ij
C
p,i i
(2-15)
Eqns. 2-1, 2-6, 2-12 constitute a system of (N+2) ordinary differential equations, which describe
the (N+2) dependent variables (N=7 molar flow rates, Fi, for the seven components in the mixture
flowing through the reactor plus the temperature and total pressure) as a function of the
independent reactor volume variable V. Together with the following initial/boundary value
conditions
V=0; Fi= Fi0; P=PF; T=TF (2-16)
the Eqns. 2-1, 2-6, 2-12 constitute an initial value problem that has been solved in MATLAB. The
value of the total heat per unit time q
v
= Q
V
V
0
dV added to the reactor can be back-calculated
from the experimentally measured exit temperature (T ex) and composition according to the
following overall energy balance Eqn. 2-17, which does not require a knowledge of reaction rate
expressions (the exact q
V
values utilized in the simulations are indicated in the captions of the
relevant Figures)
q
v
= Q
V
V
0
dV= ∑ F
i0
C
pi
dT+ 1-ϵ
bed
)ρ
s
∑ ∆H
j,T
ex
r
j
dV
V
0
3
j=1
n
i=1
T
ex
T
F
(2-17)
where ∆H
j,T
ex
is the heat of the reaction calculated at T ex.
The integrals of the rates in Eqn. 2-17 can be calculated (based on Eqn. 2-1) from the inlet F
i
|
0
and outlet F
i
|
e
molar flow rates of various species as follows
F
CH
4
|
0
X -
F
O
2
|
0
2
= 1-ϵ
bed
)ρ
s
r
1
dV
V
0
(2-18)
F
CO
2
|
e
-F
CO
2
|
0
-
F
O
2
|
0
2
= 1-ϵ
bed
)ρ
s
r
2
dV
V
0
(2-19)
34
F
O
2
|
0
=2 1-ϵ
bed
)ρ
s
r
3
dV
V
0
(2-20)
where X is the conversion of methane defined by Eqn. 2-21 below.
X=
F
CH
4
|
0
F
CH
4
|
e
F
CH
4
|
0
(2-21)
The exit temperature for the preheating zone is calculated by the following overall heat balance
equation
(
T
F
T
in
∑ F
i
C
pi
)dT=q
tot
N
i
(2-22)
where T
in
is the inlet temperature in the preheating zone (K), q
tot
(W) is the heat delivered per unit
time by the two ceramic heaters in the preheating zone.
Of the five overall global reactions that can potentially take place during the steam reforming of
biogas, specifically reactions R1, R2 and R3 as well as the oxidation reactions R4 and R5 below
2H2+O2 => 2H2O (R4)
2CO+O2 => 2CO2 (R5)
only three are linearly independent. For calculating the equilibrium conversion of CH 4 and the
corresponding equilibrium species compositions in the reactor at the measured exit temperature of
Tex, three independent reactions, R1, R2 and R3, were chosen. For the calculations, thermodynamic
equilibrium constants from the technical literature [224] were utilized. Other properties used for
the modeling can either be found in Table 2-3, or are indicated in the captions of the corresponding
Figures.
35
Table 2-3 Reactor properties used for the modeling
Diameter of the pellet 1.8 mm
Density of Catalyst pellet ~1 g/cm
3
Inside diameter of reactor 7.62 cm (3 in)
Reactor cross section area (Ac) 0.0045604 m
2
Length of reactor (L) 30.48 cm (12 in)
0.43
We have extensively, in the past, analyzed in the laboratory the combustion characteristics of
NG/syngas mixtures using model experimental combustion configurations
[128, 129, 154]. The
blending of syngas with NG was shown to increase combustion stability and to decrease NO X
emissions, in agreement with the behavior observed with the biogas/syngas mixtures in this field
study (See Sec. 2.4 below). It goes beyond the scope of this preliminary study, however, to
simulate the complex, turbulent reactive flows within the experimental test engine (a DuroMax
XP4400EH dual-fuel generator, see the Appendix for a photograph and further technical details)
utilized. We have calculated, instead, here the adiabatic combustion temperatures at the
experimental engine feed and pressure conditions in order to compare such temperatures among
the various fuels utilized. These calculations are important to provide guidance about the amount
of waste heat that is being chemically recuperated, and the level of potential increase in engine
efficiency when using the biogas/reformate blends. We have also calculated the equilibrium NO X
compositions at the corresponding adiabatic combustion conditions. Under fuel-lean conditions,
NOX emissions consist primarily of NO and NO2, and for the calculations the following two global
reactions were chosen:
N2+O2<=>2NO (R6)
2NO+O2<=>2NO2 (R7)
36
For the equilibrium calculations, the engine cylinder compression ratio, which represents the
ratio of the volume of its combustion chamber from its largest capacity to its smallest capacity,
was assumed to be 10, which is typical for the general-purpose gasoline/propane IC piston engines
employed in this study.
As part of the study, the technical feasibility of employing catalytic reforming to improve the
energy efficiency and to reduce NOX emissions during power generation from biogas has also been
evaluated. In the effort, the experimental results from field-testing and the related modeling have
been used in a financial and process model to access the commercial feasibility for the proposed
process. The capital costs, including the cost of the reformer, the heat exchangers, and engine, and
the fixed/variable operation and maintenance (O&M) cost were estimated for a 848 SCMH (500
SCFM) “biogas-to-energy” project, which is typical of sites producing low-BTU content LFG.
The focus of efforts was to evaluate whether the proposed process is potentially financially viable
for such a small size project, and the preliminary findings are summarized in Sec. 2.4 below, with
the calculation details shown in the Apendix.
2.4 Results and Discussion
2.4.1 Reformer Performance
The catalytic reformer operated stably throughout the field-testing, with the catalyst showing high
activity and no evidence of catalytic deactivation. During the studies, the effect of control
parameters, such as the steam/methane (S/C) ratio, and the total flow rate of gas going into the
reformer (which is determined by the reformer by-pass ratio), on the performance of the biogas
catalytic reforming process was quantitatively analyzed and assessed. Figure 2-3 shows the
experimental and the calculated equilibrium conversions for three different (S/C) ratios, plotted
with respect to the measured reformer exit temperature, T ex,, for the same approximate biogas flow
37
of ~2 SCMH (1.23 SCFM) at (corresponding to a catalyst weight to biogas molar flow rate ratio
of W/F=117 kg-cat·s/mol) and feed pressure of 1.7 bar (10 psig). Figure 2-4 shows the
experimental and calculated equilibrium species compositions (molar basis as is the case for all
Figures in this paper) plotted as a function of the feed temperature into the reaction zone (The
biogas composition at the field site was a bit variable. Specifically, for the experiments in Figures
2-3 and 2-4 for the (S/C) ratio of 1.3 the biogas composition was O 2: 2.1%, CO2: 29%, CH4: 30%,
N2: 38.9%, for the experiments with a (S/C) ratio of 1.68 it was O 2: 1.8%, CO2: 29.8%, CH4:
30.6%: N2: 37.8%, and for the experiments with a (S/C) ratio of 2.33 it was O 2: 1.7%, CO2: 29.8%,
CH4: 32.2%, N2: 36.3%. In the simulations, the real experimental biogas feed composition and
flow rate were utilized).
Figure 2-3 Experimental conversions and calculated equilibrium conversions for different (S/C) ratios as
a function of T ex
Higher (S/C) ratios imply higher calculated equilibrium methane conversions, though the effect
saturates out at the higher Tex. Higher (S/C) ratios though lead to higher equilibrium concentration
of H2 (on a per dry basis) in the syngas product, as the presence of steam helps to shift the
38
equilibrium of the reactions R1 & R2 to the right, see Figure 2-4 (On the other hand, higher (S/C)
ratios, generally, imply higher required qV values. The estimated qV (kW) for the data shown in
Figure 2-4 are 0.33-1.4 kW for the S/C=1.68 case, and 0.64-1.54 kW for the S/C=2.33 case). The
experimental behavior is a bit more “nuanced” though, as the reactor operates away from
equilibrium. Under these conditions, higher (S/C) ratios imply higher reaction rates due to an
increase in the concentration of reactant steam, but simultaneously the total flow (steam+biogas)
of gas going through the reactor increases (for a fixed flow of biogas), which implies a decreased
residence time and, thus, lower conversions.
Mathematical simulations of the reformer’s performance manifest similar behavior, see Figures 2-
5 and 2-6 below (For these simulations the feed pressure was 1.7 bar, the T F=1041 K, and the feed
gas consisted of O2: 1.7%, CO2: 29.8%, CH4: 32.2%, N2: 36.3%. For all cases the qV=1.6 kW,
corresponding to the total nominal heating capacity of the two ceramic heaters utilized). As these
Figures indicate (see the insert in the Figures), for the region of conditions where the reactor
operates substantially away from equilibrium (e.g., lower W/F which in this case means higher
biogas flow rates as the mass of catalyst in the reactor is maintained constant equal to that used in
the experiments) higher (S/C) correspond to lower conversions. At higher W/F (lower biogas flow
rates), however, for which the reactor behavior tracks closer the equilibrium conditions, the reactor
behavior reverses and higher (S/C) ratios imply also higher reactor conversions as well.
39
Figure 2-4 Experimental compositions and calculated equilibrium compositions as a function of feed
temperature T F for two different (S/C) ratios. Left, (S/C)=1.68, right (S/C)=2.33. Other conditions such as
feed flow rate and feed composition are listed in the pap
Figure 2-5 Calculated conversion vs. W/F for various steam to methane ratios. q V=1.6 kW
40
Figure 2-6 Calculated H 2 composition (dry basis) vs. W/F for various steam to methane ratios. q V=1.6 kW
Interesting is the behavior of the calculated concentration of H 2 in the exit stream which passes
through a maximum when plotted against W/F. This relates to the operational characteristics of
the pilot-scale reactor, which is heated from the outside with electric heaters. For larger W/F,
equilibrium conversion is attained at earlier position within the reaction zone and the exit
temperature is higher, as result shifting the equilibrium conversion of R2 to the left (For the data
presented in Figures 2-5 and 2-6 the calculated exit temperatures range from 814-1146 K for the
S/C=1 case, 829-1170 K for the S/C=2 case, and 844-1180 K for the S/C=3 case). Similar non-
monotonic behavior in H2 composition has also been observed experimentally see, e.g., Figure 2-
4.
41
The impact of increasing the biogas feed flow rate into the reformer, while maintaining the (S/C)
ratio (~ 2.33), feed pressure (~1.7 bar), and feed composition approximately constant, is shown in
Figure 2-7. (As noted previously, the biogas composition at the field site was a bit variable. For
the case of the biogas feed flow rate of 1.5 SCMH (0.91 SCFM) the feed composition was O 2:
1.8%, CO2: 30.4%, CH4: 32.00%, N2: 35.8%, while for the case of biogas flow rate of 2 SCMH
(1.23 SCFM) was O2: 1.7%, CO2: 29.8%, CH4: 32.2%, N2:36.3%). Decreasing the biogas flow
rate into the reformer, while maintaining other conditions constant (e.g., the (S/C) ratio), increases
methane conversion, as expected (consistent also with the mathematical simulations).
Figure 2-7 Experimental conversions and calculated equilibrium conversions as a function of feed
temperature T F for two different feed flow rates of biogas. Other conditions are discussed in the document.
The mathematical model has proven a very effective tool for hypothetical process design and
optimization as well as for providing important qualitative understanding into the reformer
42
behavior. Additional detailed numerical simulations of the catalytic reformer are presented
elsewhere.[225]
2.4.2 Engine Performance
A DuroMax XP4400EH dual-fuel generator was used as the test engine for all the experiments in
this project. During the study, the engine would be typically started on propane gas; once the
reformer conversion had reached a steady-state value, the gas product from the reformer would be
combined with the rest of the flow of the raw biogas, and the gas feed to the engine would be
switched from propane to the syngas/LFG mixture via the use of a three-way valve. The fuel flow
rate could be also adjusted with the use of a separate globe valve. The pressure of the fuel flow
that went into the engine was kept above 1.24 mbar (0.5 inH 2O) in order for the engine to run on
the syngas/biogas mixture (the usual pressure of the fuel in the combustion experiments was
approximately 2.48 mbar (1 inH2O)). The zero governor on the engine, connected to the propane
cylinder along with a pressure regulator, would set the pressure of the fuel at a constant zero
pressure when switched from landfill gas to propane gas fuel. The catalytic reformer and the engine
operated as a single unit continuously, and during operation the different system parts were
monitored continuously and controlled either manually or by the programmable logic controller
(PLC) – see the Appendix.
During the study, the effect of blending the reforming products with unreacted biogas on the engine
combustion characteristics were quantitatively assessed. Because of the low concentration of
methane (typically ~30%) and the high N2 content (>30%) present in the raw biogas at the Santiago
Canyon landfill, the specific engine (DuroMax XP4400EH) utilized in the project did not run on
this particular low-BTU content biogas alone, sputtering and never reaching 3600 rpm during
operation. However, the engine ran smoothly when operating on (reformer off-gas/biogas)
43
mixtures. With hydrogen being present in the fuel, the engine ran stably for various loads under
fuel-lean (i.e., excess air) conditions, resulting in fairly low-NO X emissions, when compared to the
same engine running on propane (see further discussion below). For example, in experiments in
which the flow of fuel was varied (via adjusting the fuel valve), the engine ran under significantly
fuel-lean conditions, i.e., with an air-to-fuel equivalence ratio λ=1.23 without load and λ=1.17
running on a 3 Amp load, where λ is the ratio of actual air-to-fuel ratio (AFR) to that of
stoichiometric combustion conditions, AFR st, which for these complex fuels is defined as follows:
AFR
st
=
3.762*(0.5*x%H2+0.5*y%CO+2*z%CH4)
(x% H2+y% CO+z%CH4)
(2-23)
Figure 2-8, for example, indicates the engine performance when running on a 50% biogas/50%
reformate gas mixture containing hydrogen for three different operating engine regimes (for the
experiments in this figure the reformer conditions were feed pressure of 1.7 bar (10 psig), 1.1
SCMH (0.65 SCFM), S/C=3.4, biogas composition O2: 1.3%, CO2: 31.3%, CH4: 36.3%, N2:
31.1%, TF=699℃; the gas composition (syngas + raw biogas) fed into the engine was CO: 14.9%,
CO2:18.3%, CH4:12.6%, H2:32.4%, O2: 0.44%, N2: 21.2%). The NOX emission data in Figure 2-
8 were recorded by measuring the NO and NO2 concentrations in the engine exhaust. To do so,
before the exhaust gas would flow into the sensor block for the NO X measurements, the water
vapor was removed with the use of thermoelectric Peltier refrigeration cooling system and by
passing it through an adsorption column packed with silica gel. All the measurements were
conducted at atmospheric pressure. In order for the gas to have enough flow to the sensor block,
suction was applied on the exit side of the sensor block by connecting it to the suction port of the
landfill blower on the site.
44
Figure 2-8 The NO X (NO, NO 2) emission data in the exhaust gas for the engine running on a syngas/biogas
mixture with 50% of the total flow of biogas going into the reformer. Three different operating engine
regimes: No load, 2 Amp, and 3 Amp load.
As noted previously, because of the low concentration of CH 4 and the high N2 content of the biogas
at the Santiago Canyon landfill, where the field-testing took place, the specific engine utilized in
the project did not run on biogas alone, sputtering and never reaching 3600 rpm during operation.
However, the engine ran smoothly when operating on (reformer off-gas/biogas) mixtures. It was
not possible, therefore, to compare the performance of the engine running on (biogas/syngas) to
that of the engine running on raw biogas alone. Therefore, as an experimental metric of improved
engine performance the emissions of the engine running on biogas/syngas mixtures were compared
with the corresponding emissions with the engine running on propane (because of the remote
location of the field site that was the only gaseous fuel that was readily available), which are shown
in Figure 2-9. Comparing the NOX emissions from the engine running on propane gas alone (Figure
2-9) with those from the engine running on a 50% biogas/50% reformate gas mixture (Figure 2-
8), under similar loads it is clear that the engine running on (reformer off-gas/biogas) mixtures
shows significantly lower emissions than the engine operating on pure propane under the same
15
32
39
3
1 1
0
5
10
15
20
25
30
35
40
45
No Load 2 Amp 3 Amp
NOx Composition (ppm)
Load
Combined Reformate & Raw LFG Emission Analysis
NO (ppm)
NO2 (ppm)
45
load conditions (65% NOX reduction for the no-load case, 51% reduction for the 2 Amp case, and
41% for the 3 Amp case).
.
Figure 2-9 The NO x (NO, NO 2) emission data in the exhaust gas for the engine running on propane for three
different operating regimes: No load, 2 Amp, and 3 Amp load.
We have previously extensively analyzed in the laboratory the combustion characteristics of
NG/syngas mixtures using model experimental combustion configurations.[128, 129, 154] The
blending of syngas with NG was shown to increase combustion stability and to decrease NO X
emissions, very similarly to the behavior that is observed with the biogas/syngas mixtures in this
field study (e.g., see Figures. 16, 17 in the study of Ren et al.[129]). It goes beyond the scope of
this feasibility-type study to investigate the complex reactive flows within the experimental engine,
and was not attempted here. However, to add further insight into the potential of the proposed CR
concept, additional simulations were carried out, in which as noted in Sec. 2.3 the engine is
modeled as a flow reactor, in which all relevant reactions have reached equilibrium under adiabatic
48
63
64
4
5
4
0
10
20
30
40
50
60
70
No Load 2 Amp 3 Amp
NOx Composition (ppm)
Load
Propane Emission Analysis
NO (ppm)
NO2 (ppm)
46
conditions. Figure 2-10 shows the calculated adiabatic flame temperatures as a function of the air-
to-fuel equivalence ratio (A raw LFG with composition (CH4: 32.2%, O2: 1.7%, CO2: 29.8%,
N2: 36.3%), a 50% reformed biogas/50% LFG mixture with composition (CH 4: 12.6%, O2: 0.4%,
CO2: 18.3%, N2: 21.2%, CO: 14.9%, H2: 32.4%), a 70% reformed biogas/30%LFG with
composition (CH4: 6.7%, O2: 0.3%, CO2: 16.7%, N2: 22.9%, CO: 16.4%, H2: 36.7%) and a 100%
syngas mixture with composition (CH4: 1.4%, O2: 0%, CO2: 14.3%, N2: 20.2%, CO: 18.8%, H2:
45.3%) are studied in this figure).
As can be seen from Figure 2-10, the use of the syngas mixtures results in significant increase in
the adiabatic temperature, which provides a good measure of the amount of waste heat that has
been recuperated, and of the potential for increase in system efficiency. The waste heat recuperated
can also be directly estimated based on the measured gas compositions and the energy content (i.e.,
lower heating value or LHV of the biogas components) and it is significant. For example, for the
case of Figure 2-8, the (syngas/biogas) mixture contains 12.3% more energy than the raw biogas.
47
Figure 2-10 Adiabatic flame temperature vs. the air-to-fuel equivalence ratio λ for various gas feed
compositions.
NOX equilibrium calculations, on the other hand, result in values which are a couple of orders of
magnitude higher than what was observed experimentally during field-testing, which is an
indicator that these processes operate far away from equilibrium, and the improvements in NO X
emissions attained are, instead, the result of complex kinetic processes. [130, 226]
The experimental/modeling results of this preliminary, “proof-of-concept” study have been quite
promising. Based on these findings, a cost feasibility evaluation for an 848 SCMH (500 SCFM)
“biogas-to-energy” project was performed to verify whether it would be financially viable (further
details provided in the Appendix). Two technologies were compared, the conventional technology,
which requires the use of a selective catalytic reduction (SCR) system to allow an engine to operate
while satisfying environmental regulations, and the proposed CR system which is envisioned
capable to operate without needing such a SCR system (The preliminary results reported here are
for building a new, “grass-roots” system, but the most likely initial adaptations of the technology
48
will be as retrofit systems, for landfills with existing biogas-to-energy facilities, for which biogas
quality has declined to the point where the economics of operation are marginal. In these
circumstances, the choice will be between abandoning the existing system, and flaring the biogas,
instead, or retrofitting it with the CR technology to prolong its operating life-time. Though flaring
is presently still permissible, the development of technologies like the CR system proposed here
that permit the beneficial use of such a resource for electricity production, may “spur” new
regulations that discourage the continuation of present flaring practices).
For the 848 SCMH size project, the up-front capital cost of the CR system is estimated to be ~$2.83
million, which is higher (by ~$0.37 million) than the cost of the conventional system without a
reformer. However, the O&M cost (including NOX control) of the CR system is lower, with its
annual cost ($358,347) being less than that of the conventional system ($415,769). The estimated
net annual revenue (annual revenue minus annual cost) for the CR process is $ 755,925 (with the
payback period for the proposed process being 2.7 years), while that for the conventional process
is $ 635,431). It should be pointed out that estimating the costs/revenues of a new, and not yet
fully developed technology, and comparing them against a mature technology is an “in-exact
science”. It is expected, for example, that once the new technology is fully developed, costs will
be reduced. The next phase of development for the proposed CR technology should, therefore,
involve testing the technology at a larger scale with the focus on reducing actual costs, by
improving heat control, minimizing catalyst volume, and reducing operating costs, and increasing
revenue by optimizing waste heat utilization and maximizing energy efficiency. A key potential
“draw” of the proposed CR system for the intended application (closed or minimally-staffed
landfill sites) is that it is passive hence, except for minimal equipment maintenance (e.g., cleaning
the heat exchangers), there is no additional operation and maintenance costs.
49
2.5 Conclusions
The performance of a catalytic reformer operating on raw biogas, after undergoing conventional
pre-treatment for the removal of its trace impurities, was evaluated during this experimental study
with respect to product composition and the conversion of the methane component of the biogas
for different flow rates (and corresponding residence times) and for various (steam/methane)
ratios. The catalytic reformer operated stably for the duration of the tests, with the catalyst showing
high activity. Higher reactor residence times and (steam/methane) ratios resulted in higher carbon
conversions and in higher hydrogen concentration in the resulting syngas. Reactor performance
was also validated by modeling.
Blending the reformer off-gas with the raw landfill gas and using it to generate electricity in an
engine was shown experimentally to generate good benefits for the engine performance. For
example, the specific engine used in this project would not run on low-BTU content biogas
produced at the long-ago closed Santiago Canyon landfill in California, where the field-testing
took place, sputtering and never running at full-speed (3600rpm at a frequency of 60Hz). However,
the engine would run smoothly when operating on (reformer off-gas/biogas) mixtures as a result
of the addition of the reformate mixture, which made the fuel mixture flammable and resulted in a
stable combustion conditions. With the hydrogen present in the LFG, the engine would run stably
for various loads under fuel-lean (excess-air) conditions, resulting in low-NO X emissions (similar
observations were recently reported by Zhen and coworkers [226] in a laboratory study with
incombustible simulated biogas mixtures, which upon H 2 addition became flammable).
Specifically, the engine operating on (reformer off-gas/biogas) mixtures showed significantly
lower emissions than when operating on pure propane under the same load conditions.
50
The results of this preliminary, “proof-of-concept” research are promising; however, it is still early
to begin commercialization or marketing. Significant additional development is required to better
understand the process and how it should be optimized. It is important, for example, that further
process development continues with a larger, turbocharged, lean-burn engine. This will, then,
allow a direct comparison of engine performance and NO X emissions using both the raw LFG and
the LFG/syngas mixtures that did not prove feasible in the present study. The testing should also
focus on optimizing waste heat utilization to maximize energy recovery. Preliminary estimates are
that the technology has the potential to improve energy efficiency by more than 10% from waste
heat recuperation. The break-point between the extent of reforming for increasing the energy
content of the fuel and turbocharger performance when operating on (reformate/biogas) mixtures
needs, however, to be studied. Given the added costs of carrying out field-testing with a larger
engine, in a future study, the focus should be on optimizing the heat recovery, turbocharger
performance, engine kW output, engine heat rate, and NOX emissions via process simulations first,
and then field-testing of the most promising process alternatives. It should be pointed out that just
1% reduction in fuel use in a large engine (via efficiency improvement) is significant. Having the
potential to decrease fuel use by as much as 10% would be a huge advancement for the “waste-to-
energy” field.
Fuel reforming has been proposed and applied in the past for diesel engines with technical success
[227, 228] but has yet to become commercial. However, the economics in diesel engines that do
not face operability issues are different that those the biogas industry faces. Currently, the stringent
air emissions requirements in various parts of the country, particularly in several air districts within
the State of California preclude the use of stationary IC engines operating on biogas, primarily
because they emit large amounts of NOX, a key regulated pollutant. Also, due to poor flame
51
stability when using marginal or poor quality biogas, maximum power during biogas combustion
may be low, or like in this project, the engine may be totally unable to operate on raw biogas. One
way to solve both these problems is to use the proposed CR system in the “biogas-to-energy”
process. Finding a clean way of burning biogas in IC engines would benefit our nation through
utilization of a very important renewable fuel and resource, and could help lead it down the path
of energy independence without the adverse health effects and cost to the economy of burning
fossil fuel resources. This technology, in particular, can help convert poor-quality biogas, that
would otherwise be unusable and likely to be flared, to a usable fuel for power generation, and in
so doing can offer the benefit of offsetting natural gas use elsewhere.
52
3. Fabrication of Nanoporous SiC Membranes via Dip-Coating Technique
3.1 Introduction
Membrane separations have attracted attention over the past three decades due to their low energy
requirements compared to the more conventional separation technologies like distillation.
Polymeric membranes have been the most intensively investigated, and are now widely used
commercially. Inorganic membranes, on the other hand, have received relatively less attention,
despite the fact they also show good promise for broad applications [229-233]. There are presently
a number of commercial liquid-phase separations employing such membranes, but commercial
gas-phase applications are presently lacking. However, high-temperature and high-pressure gas-
phase reactive separations are an area where inorganic membranes have, potentially, a distinct
advantage over polymeric membranes, and thus such applications remain today key drivers for the
further development of inorganic membranes. In particular, hydrogen-selective inorganic
membranes for use in the efficient production of hydrogen, whose demand has been increasing
steadily in recent years due to its potential use in clean energy generation, are attracting significant
attention.
However, currently available inorganic membranes face challenges when used in reactive
separations for hydrogen production, particularly in the context of power generation [136, 234].
They include questionable hydrothermal stability (microporous silica), low hydrogen selectivity
and undesirable reactivity towards hydrocarbons (zeolite membranes) of concern for their use in
catalytic dehydrogenations, high cost and sensitivity to sulfur poisoning and coking (Pd and Pd-
alloys), and poor resistance to oxidizing environments (carbon molecular-sieve membranes). This
has then necessitated the search for alternate materials to prepare these membranes with. A
different type of hydrogen permselective microporous membrane, which shows promise in
53
overcoming some of these challenges, is made of materials with a -[Si-C]- backbone [235-253],
because of their many desirable unique properties, like high corrosion resistance [254], high
thermal conductivity [255], high thermal shock resistance [256], and excellent chemical and
mechanical stability [257].
To prepare such membranes, two different preparation techniques have been employed: chemical-
vapor deposition (CVD)/chemical-vapor infiltration (CVI), and the pyrolysis of pre-ceramic
polymer precursors. Early efforts [235-237],
employing
the CVI/CVD technique and alumina
porous supports prepared SiC membranes with mostly Knudsen-type transport characteristics, and
no permselectivity towards hydrogen. Ciora et al.
[238] were the first to report on the preparation
of truly microporous SiC membranes on -Al2O3 tubular supports via the CVD technique by
employing two different CVI/CVD precursors, namely CH 3SiH2CH2SiH3 (1,3-disilabutane or
DSB) and (C3H7)3SiH (TPS). Both precursors produced hydrogen permselective membranes
[238]. However, the DSB-derived membranes, though thermally stable, were hydrothermally
unstable. The TPS-derived SiC membranes, on the other hand, proved stable in the presence of
high-pressure (1-3 bar) and temperature (<750
o
C) steam. However, the preparation procedure
using TPS involved multiple steps, and required a high temperature (~1000
o
C) post-treatment
[238], which impacted the membrane’s pore structure, thus making it difficult to predict and to
control final product quality, and from a membrane manufacturing standpoint also diminishing the
advantage of the on-line control of the CVD/CVI technique.
As a result, the Group abandoned the CVD/CVI approach [238], in favor of a method that involved
depositing thin pre-ceramic polymer films on macroporous SiC supports followed by pyrolysis in
an inert atmosphere (Ar) to produce a SiC microporous ceramic. The precursor utilized was allyl-
hydridopolycarbosilane (AHPCS), a partially allyl-substituted hydridopolycarbosilane (HPCS).
54
HPCS and AHPCS both produce a SiC-based ceramic with a (Si/C) ratio close to one [239, 240].
However, the allyl groups present in the AHPCS promote polymer cross-linking in Ar, rather than
oxygen, which is typically utilized for other PCS polymers. This is important, since curing in
oxygen may introduce a substantial fraction of -[Si-O-C]- linkages in the bulk of the resulting
ceramics, which have proven to be thermally and hydrothermally unstable [241, 242]. The initial
efforts employed [238] a simple dip-coating method to deposit the thin precursor films, and
prepared membranes with a satisfactory H2 permeance (in the range 10
-8
-10
-7
mol m
−2
s
−1
Pa
−1
),
but relatively low selectivity (an ideal separation factor of ~20 for He/N 2 at 200
o
C). Subsequently,
the Group improved the preparation method by augmenting the dip-coating process with a slip-
casting step (with the aid of SiC nano-powders or nanofibers) that conditions the support surface
prior to film deposition [258, 259]. Combining slip-casting with dip-coating significantly
improved membrane properties, but equally importantly the reproducibility in preparing high-
quality membranes. Also, steam-stability experiments with the membranes lasting 21 days, using
an equimolar (He/H2O) mixture at 200
o
C, indicated good membrane stability at these conditions.
The preparation technique was further improved [260], by alternating the dip-coating of the SiC
precursor layers with coatings of polystyrene sacrificial interlayers on the top of slip-casted SiC
supports. Membranes prepared by this technique exhibited (He/Ar) and (H 2/Ar) ideal separation
factors in the range of (176-465) and (101-258), respectively, with permeances 2-3 times higher
than those prepared by the more conventional techniques [259]. Mixed-gas experiments with the
same membranes indicate separation factors as high as 117 for an equimolar (H 2/CH4) mixture.
This superior performance was attributed [260] to the effect that PS has on the formation of the
three-dimensional (3-D) membrane pore structure during AHPCS pyrolysis, and its role in
55
preventing the pre-ceramic polymer macromolecules from infiltrating into the underlying
membrane layers during dip-coating.
Prior to the publication by this group [238], other investigators had also reported the preparation
of -[Si-C]- type porous membranes via the pyrolysis of thin films made from a variety of other
precursors (e.g., PCS and polysilanes). In an early effort, Morooka and coworkers [241, 242]
pyrolyzed PCS, cured in oxygen, to prepare [Si–O–C] membranes on alumina substrates, using
polystyrene (PS) as a pore former. A membrane prepared with 1% PS in PCS had a H 2 permeance
of 4×10
−8
mol m
−2
s
−1
Pa
-1
, and an ideal H2/N2 selectivity of 20 at 773 K; however, it proved
unstable when heated in Ar at 1223 K, or exposed to steam at 773 K. Lee and Tsai [244, 245]
prepared [Si–O–C] membranes by pyrolysis of polydimethylsilane (the PMS layer was subjected
to a thermolytic reaction at 733 K in Ar, followed by O 2 curing at 473 K, and finally pyrolyzed at
various temperatures from 523 to 1223 K). The membranes prepared via pyrolysis at 873 K had
the best separation characteristics (membranes prepared at the higher temperatures were not
microporous), exhibiting a H2 permeance of ~2.7×10
−9
mol m
−2
s
−1
Pa
−1
, and an ideal (H2/N2)
selectivity of 20 at 473 K. PMS-derived membranes prepared in an autoclave under a N 2
atmosphere at low temperatures were also microporous, but proved unstable to the exposure to
steam [246].
Subsequent to these earlier efforts (and the study of Ciora et al. [238]), other Groups have reported
efforts to produce nanoporous SiC membranes via the pyrolysis of precursor films. Wach et al.
[247, 248]
prepared [Si-O-C] membranes by the pyrolysis of a blend of PCS and polyvinylsilane
films on porous alumina substrates by radiation curing in the presence of O 2. Their membrane
exhibited a H2 permeance of ~3×10
-9
mol m
−2
s
−1
Pa
−1
and ideal (H2/N2) and (He/N2) selectivities
of 206 and 241 at 250
o
C, respectively [248]. Suh et al.
[2009] spin-coated thin HPCS films on
56
flat γ-alumina substrates to prepare, upon pyrolysis, SiC membrane films with H 2 permeances as
high as 1.6×10
-7
mol m
−2
s
−1
Pa
−1
, and with (H2/N2) permselectivity of >42.5 at room temperature.
To reduce the fraction of [Si-O] bonds in the structure of the resulting membranes, several studies
have attempted curing the PCS films in the absence of O 2. For example, Suda et al. [250, 251]
prepared SiC membranes by dip-coating PCS on macroporous α-alumina tubes, using p-
diethynylbenzene (as a cross-linking agent) and Pt 2(dvs)3 (dvs: 1,3-divinyltetramethyldisiloxane)
as a hydrosililation catalyst (PS was used as a pore-former in the preparation of some of these
membranes). Their cross-linked PCS-derived SiC membranes exhibited ideal (H 2/N2) separation
factors of (90-150) and a H2 permeance in the range of (1-3)×10
-8
mol m
−2
s
−1
Pa
−1
at 373K [250].
Nagano et al. [252] prepared SiC membranes by dip-coating of PCS on γ-alumina supports, and
reported a H2 permeance of ~10
-7
mol m
−2
s
−1
Pa
−1
, and an ideal separation factor of (8-12) for
H2/N2 at 873 K. However, curing of PCS in the absence of oxygen resulted in a high carbon ratio
in the final membranes [252]. The same group applied a hybrid approach involving using a CVI
technique (operated in a cyclic fashion, by switching between SiH 2Cl2+H2 and C2H2+H2) to repair
pinholes and cracks on membranes prepared via the pyrolysis of PCS films coated on γ-alumina
substrates; however, not much improvement was reported for the H 2/N2 permselectivity [253].
Takeyama et al.[261] prepared SiC membranes by dip-coating a PCS layer on alumina tubes (α-
Al2O3 coated with a thin γ-Al2O3 layer), exposing the coated film to electron beam irradiation in
He at room temperature, followed by subsequent pyrolysis at a preset temperature (973K, 1073K,
and 1123K) for 0.5 h in Ar. The best group of resulting membranes had a H 2 permeance of 3.1×10
-
7
mol m
−2
s
−1
Pa
−1
and a (H2 /N2) selectivity of 51 at 523K.
A key step in the fabrication of microporous inorganic membranes, in general [262, 263], and of
SiC membranes [238, 258-260], in particular, is preparing proper supports that provide a
57
mechanically-strong template on which the selective layer is deposited. Mechanical strength is a
key need here, but these supports must also have high flux, to not negatively impact membrane
throughput, and their surface characteristics (roughness) must readily permit the deposition of thin
films; they must be, in addition, resistant to high temperatures and pressures, to corrosive
atmospheres encountered during H2 production (e.g., steam), and to thermal and pressure cycling.
Overwhelmingly, the support materials of choice for preparing inorganic membranes, in general,
and with notable exception of our own studies [238, 258-260, 263] for SiC nanoporous membranes
as well (recent efforts to prepare SiC membranes for liquid-phase ultrafiltration (UF) applications
make also use of SiC macroporous supports [264, 265]), are either α-Al 2O3 or γ-Al2O3, the main
reason being their easy commercial availability. γ-Al 2O3, the most frequent choice has, in addition,
a smooth surface and is fairly inert, making it desirable substrate for depositing thin films (this is
a key advantage over other common supports, e.g., porous stainless steel which has a rough surface
and is fairly reactive, thus requiring elaborate surface pre-treatment prior to film deposition [266,
267]). However, γ-Al2O3 supports face their own challenges because of their questionable stability
in high-temperature/pressure steam. The preparation of SiC membranes, via CVD/CVI or the
pyrolysis of pre-ceramic precursors, presents also added challenges for γ-Al 2O3 supports because
it takes place at high temperatures where these materials are no longer stable. For these reasons
(better thermal/mechanical stability and the closeness in thermal expansion coefficient with the
top microporous SiC layer), our team substituted the γ-Al 2O3 supports, used in early efforts [238],
with SiC supports [258-260].
We reported in an early study the preparation of such supports made via the pressureless sintering
of SiC powders in Ar [263]. We revisited the topic in a more recent study whose focus was to
significantly improve the permeation characteristics and mechanical properties of such supports
58
[268]. In the study, the effects of the type of starting powders used and their composition, the
sintering temperature and the amount of sintering aids utilized on the transport characteristics and
the surface roughness of the sintered SiC porous supports were systematically investigated. Using
blends rather than pure powders alone was shown to offer an advantage in terms of preparing
supports with very high permeances and good mechanical stability, because combining powders
with distinctly different sizes allows them to pack better together and to create better sintered and
more highly permeable structures. Increasing the sintering temperature increases the permeance of
the resulting supports, in line with measurements of the average pore diameters which increase
with increasing sintering temperature [Deng et al., 2014]. Decreasing the amount of sintering aids
increases the support’s permeance and decreases the ideal separation factors, consistent also with
porosity and average pore size data with such samples which both increase with decreasing content
of sintering aids. Importantly, also the mechanical properties of supports prepared with smaller
amounts of sintering aids are quite adequate for the proposed application and as good, or even
substantially better, than most of the commercial supports used today for the preparation of
inorganic membranes.
One downside with these highly permeable SiC supports (a He permeance as high as 5.8×10
-5
mol•m
-2
•Pa
-1
•s
-1
, ~2 orders of magnitude larger than that of supports in our original publication
[263], and 3 - 10 times larger than the permeance of SiC supports prepared more recently by other
groups [269]) is their increased surface roughness that makes it challenging to prepare relatively
thin nanoporous membrane films. The difficulty is also compounded by the relatively large
average pore size that makes it challenging to prepare supported nanoporous membrane by dip-
coating or CVD/CVI approaches. The emphasis, therefore, in our efforts reported here has been
on improving the membrane fabrication techniques to be able to prepare quality nanoporous
59
membranes using these supports. We report in this paper the results of such efforts on the use of
the dip-coating technique, employing pre-ceramic polymer precursors, to prepare microporous
membranes on these newly developed supports. In a future publication, we will report on the use
of a novel CVD/CVI method that also prepares quality microporous SiC membranes.
3.2 Experimental
The macroporous SiC support tubes used for membrane fabrication were prepared using uniaxial
cold-pressing of -SiC powders ( H S C 0 5 9, provided by the Superior Graphite Co.). Two different
types of supports were investigated here: the first type was prepared by employing a pure -SiC
powder with an average particle size of 0.6 µm. The second type of support was fabricated using
a blend of two powders (50%/50% by weight), the first being the aforementioned 0.6 µm powder,
the second being a SiC powder with an average particle size of 6 µm. We used no sintering aids to
prepare these supports, which were sintered at the relatively high temperature of 1900
o
C (by
comparison all SiC supports used by our Group in the past to prepare nanoporous SiC membranes
employed sintering aids and were sintered at 1700
o
C). Preparing quality supports without the need
to employ large amounts of sintering aids is important in terms of the eventual use of these
membranes in reactive separations, particularly those involving high-temperature and high-
pressure steam, as some of these sintering aids (e.g., alumina) have been shown to be unstable
under such conditions [268].
The support tubes used in membrane preparation in this study were treated in flowing synthetic air
at 450
o
C (with the purpose of oxidizing any potential carbon contaminants present), sonicated
several times in acetone, and then dried prior to membrane film deposition. For the deposition of
a mid-layer scaffold, which is employed to condition the support surface to make it amenable to
further SiC precursor film deposition, we used a slip-casting step. For that, we utilized a solution
60
that consists of 10 wt.% AHPCS (SMP-10, Starfire Systems, Inc.) and 90 wt.% hexane (HX0299-
5, EMD Chemicals), hereinafter referred to as the dip-coating solution. In the solution we added
SiC particles (5 wt.%), which were obtained by separating the (100-200 nm, as verified by
scanning electron microscopy (SEM) [268]) fraction of the 0.6 µm -SiC powder via a sequential
precipitation/separation approach beginning with a slurry of the original SiC powder in acetone
[258], and separating (specifically, 7 g of the 0.6 µm SiC powder was mixed with 200 ml of acetone
in a beaker, with the top 50 ml of the 200 ml solution being separated) and drying the lighter
particles. In addition to the aforementioned slip-casting solution, we used another slip-casting
solution (referred to as PS-slip-casting solution) which is prepared from the first slip-casting
solution by adding 1.25 wt.% polystyrene (PS, GPC grade, M w=2500, Scientific Polymers
Products, Inc.). For slip-casting, the slurry was first sonicated for 1 min and the support tubes were
placed in it for 12 s, and then drawn out of it at a speed of 0.28 mm/s. The coated tubes were heated
in flowing Ar in a tube furnace (Lindberg/Blue, Model STF55433C) at a rate of 2
o
C/min, first to
200
o
C, where they were kept for 1 h, then to 400
o
C, where they were also kept for 1 h, and finally
to 750
o
C, where they were kept for an additional 2 h. Subsequently, they were cooled down to
room temperature in flowing Ar with a cooling rate of 3
o
C/min. The reason for the relatively slow
heating (and holding at 200
o
C) is that prior studies [239, 240, 259, 260] indicate that using such
heating rates and treatment at lower temperatures result, generally, in better cross-linked
amorphous SiC materials.
The asymmetric porous SiC support tubes thus prepared (with the slip-casted SiC layer on the top
of the original support) were then used to deposit a nanoporous SiC top layer. The technique
utilized involves the use of polystyrene sacrificial interlayers [258, 260]. Specifically, the
aforementioned tubes (support + mid-layer scaffold) are dip-coated in a solution of 1 wt.% of PS
61
(GPC grade, Mw=2500, Scientific Polymers Products, Inc.) in toluene, with a dip-coating time of
12 s and a drawing rate of 0.58 mm/s. The tubes with the PS layer on the top were first dried at
100
o
C for 1 h, and then dip-coated in a solution of 10 wt.% of AHPCS in hexane, with a dip-
coating time of 12 s and a drawing rate of 1.52 mm/s (the choice of hexane as a solvent for the
AHPCS is because it does not substantially dissolve the PS). Following the coating of the top
AHPCS layer, the composite membrane (support + mid-layer + PS layer + top layer) was
pyrolyzed in Ar, following the same heat treatment protocol used for the preparation of the mid-
layer scaffolds, as described above. In order to investigate the impact that the number of dual (PS
+ AHPCS) layers deposited has on the separation characteristics, membranes with one, two and
three such dual layers were prepared (the deposition of each additional dual layer followed the
deposition/pyrolysis of the previous layer). After the desired number of layers were deposited and
pyrolyzed, the composite membranes were treated in flowing synthetic air for 2 h at 450
o
C to
remove carbon-containing residues that may remain. With AHPCS-derived SiC microporous
membranes this air oxidation step results in higher permeances but lower separation factors [258,
260], which is an undesirable outcome, but by employing such treatment, prior to use, one ensures
that no further variation in properties will occur due to accidental exposure to air.
The morphology of the membranes during the various stages of fabrication was characterized by
SEM. The permeation characteristics of the membranes were measured using a Wicke-Kallenbach
type permeation apparatus employing ultra-high purity He and Ar, two noble (non-adsorbing) test-
gases which our team has extensively utilized in the past to characterize these nanoporous
membranes with further details about the technique found elsewhere [238, 258-260]. For all the
measurements the temperature was kept at 200
o
C, the pressure on the membrane permeate side
was kept atmospheric, while the transmembrane pressure gradient was kept constant at 241 kPa.
62
3.3 Results and Discussion
Microporous inorganic membranes have generally an asymmetric structure, consisting of a
macroporous support that is typically highly porous, and highly permeable to gas molecules, and
one or more additional layers deposited on the top that perform the separation function. Though,
ideally, one would like to directly deposit the microporous separation layers on the top of the
support, this typically is not feasible. This is because a highly permeable macroporous support
has, typically, a large average pore size and porosity, and is generally made of large-size particles
sintered (fused) together. This, then, means that one must deposit top layers which are several-fold
in thickness larger than the support’s average particle size. In addition, it is often quite difficult to
prevent infiltration of the precursor top layer into the underlying support, which in turn negatively
impacts the latter’s permeation characteristics. One way to overcome this challenge, is to employ
an intermediate scaffold layer on the top of the support that prevents infiltration during the
deposition of the additional precursor films for the microporous layers, and allows the deposition
of relatively thin such films. The resulting composite membrane structure is shown schematically
in Figure 3-1.
Figure 3-1 Structure of an idealized SiC membrane
The approach followed in this work to prepare the mid-layer scaffolds on the top of the
macroporous SiC supports, as described in the experimental section, involves preparing a slurry
of SiC nano-powders wth sizes comaparble to the average pore size of the underlying supports in
Nano-porous
membrane
Meso-porous layer
63
a precursor solution that is utilized to prepare the subsequent nanoporous SiC films, and pyrolyzing
the composite structure. The resulting mid-layer serves the purpose of forming a surface that is
more amenable to film deposition than the original support surface, thus making it possible to
deposit microporous films on supports with varying pore size and surface roughness. The SiC mid-
layer scaffold, in addition, serves the dual purpose of preventing infiltration into the underlying
support during the deposition of the additional microporous layers. Employing this hierarhcical
pore structure approach has allowed this team in the past to prepare quality SiC membranes with
a He permeance of 0. 9 - 1.2×10
-8
mol•m
-2
•s
-1
•Pa
-1
at 473 K, and with a (He/Ar) separation factor
of 89 – 147, the latter being an order of magnitude better than the membranes we could prepare
using the same supports but without the deposition of the mid-layers [268].
Further improvements in the properties of the resulting membranes, as also noted in the
Introduction, resulted from modifying the process of the deposition of the precursor microporous
SiC films themselves. For that, prior to the deposition of the preusrsor film on the underlying
support surface, the surface is coated with a thin PS film. The goals here are twofold: to minimize
infiltration of the membrane films into the underlying support structure; and for the poymer itself
to serve the dual role as a pore-former during the pyrolysis step, an idea previously used by other
Groups as well when prepraing SiC nanoporous membranes [250, 251]. This modification adopted
for the film deposition step resulted in membranes with signifcantly improved characteristcs, with
He permeances in the range (1.8 - 4.3×10
-8
mol•m
-2
•s
-1
•Pa
-1
) and a (He/Ar) separation factor in
the range of 176 – 465.
With the advances made in the deposition of the mid-layer scaffolds and the nanoporous films, it
became clear that further significant improvements in the throughput (permeance) of these
composite membranes would come from improvements in the permeance of the underlying
64
macroporous SiC supports themselves, whose permeance in the earlier efforts, typically, ranged
from 10
-7
- 10
-6
mol•m
-2
•s
-1
•Pa
-1
. Our focus then shifted on preparing more highly permeable, but
still mechanically strong, supports that could be used to prepare SiC nanoporous membranes. As
noted previously, we have succeded [268] in preparing such highly permeable supports with
permeances as high as ~ 6×10
-5
mol•m
-2
•s
-1
•Pa
-1
. This was accomplished via the appropriated
selection of starting SiC powders, sintering aids and conditions. Compared to the earlier supports,
all these new materials have larger porosities and average pore sizes, and greater surface
roughness, but comparable, if not superior, mechanical properties [268]. As desirable these
characteristics are from the standpoint of enhanced membrane throughputs, they create their own
challenges in terms of thin film membrane deposition.
The focus of this research in recent years, therefore, has been on being able to use these supports
to prepare quality microporous SiC membranes. We report here results of our efforts using two
different types of supports: One fabricated from a single starting SiC powder (~0.6 µm in average
particle size) and another made from a blend of two powders (50 wt.% of 0.6 µm and 50 wt.% of
6 µm size SiC particles). Both types of supports were prepared without using any sintering aids,
but employing the relatively high sintering temperature of 1900
o
C (by comparison, the sintering
temperatures of supports used in the past for SiC membrane preparation was 1700
o
C),
intentionally so, as such supports naturally represent the greatest challenge to prepare quality
membranes. In fact, not employing sintering aids at all to prepare SiC materials, is counterintuitive,
as the commonly held belief is that such materials cannot be prepared without using any sintering
aids [270-272]. However, most of the previous studies focused on dense SiC materials, and our
recent work has shown that highly permeable and mechanically robust membrane supports can be
prepared employing a small amount or even without using sintering aids altogether [268].
65
The permeation results of membranes made using the single starting 0.6 µm SiC powder are shown
in in Figure 3-2, which shows the He permeance (left axis) and the ideal (He/Ar) separation factor
for membranes prepared using the conventional (no PS added to the slurry) slip-casting approach.
In Figure 3-2, and the subsequent Figures, each experimental point represents the average
permeance and separation factor of three different membranes prepared under the same conditions,
while the error bars reflect the range of experimental values measured (this means that three
different supports must are used to generate each experimental permeation point, as the permeation
method employed does not permit the reuse of the membrane to deposit additional layers). As this
Figure indicates, the initial supports are highly permeable (2.25-3.48 ×10
-5
mol•m
-2
•s
-1
•Pa
-1
).
Adding the mid-layer reduces the permeance to (2.21-3.63)×10
-6
mol•m
-2
•s
-1
•Pa
-1
but does not
improve the ideal (He/Ar) separation factor. It takes two dual (PS+AHPCS) coatings before a
microporous membrane is prepared with an average He permeance of ~1.08×10
-8
mol•m
-2
•s
-1
•Pa
-
1
and an ideal (He/Ar) separation factor of ~1000. Adding a third dual layer does not change the
He permeance much (~0.9×10
-8
mol•m
-2
•s
-1
•Pa
-1
), but improves the ideal (He/Ar) separation factor
(~ 1200).
66
Figure 3-2 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes employing
supports made from a single 0.6 µm SiC starting powder via the use of a regular slip-casting solution,
prepared as described in the Exp. Sec. Measurement temperature 473 K; pressure difference across the
membrane 2.41×10
5
Pa; permeate side pressure 1 atm; each individual data point reflects results of three
different membranes prepared under identical conditions.
Figure 3-3 shows the permeation results of membranes made using again the single starting 0.6
µm SiC powder, but in this case fabricated using the PS-slip-casting solution that is prepared by
adding PS to the regular slip-casting slurry. The presence of PS in the slurry seems to have a
positive effect on the properties of the asymmetric membranes formed by the deposition of the SiC
precursor films. Already (and in contrast with the case of the membranes prepared with the regular
slip-casting solution without PS being present) with the deposition of 1 dual layer, a quality SiC
membrane is formed with high permeance of (1.05-2.36)×10
-7
mol•m
-2
•s
-1
•Pa
-1
and an ideal
(He/Ar) separation factor of (60-131). Limiting the numbers of microporous layers one has to
deposit on the supports is a key determinant on the final cost of these materials. It should be noted,
furthermore, that transport through such membranes is activated for the fast component and
follows a Knudsen-like behavior for the slow component (1/T
0.5
), with the activation energy for
0
200
400
600
800
1000
1200
1400
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
Support Slip-casting Layer 1 Dual Layer 2 Dual Layer 3 Dual Layer
Spn Factor of He/Ar
He Permeance/mol.m
-2
.s
-1
.Pa
-1
Permeance
Spn Factor
67
He being ~ 9 kJ/mol [259]. Though experimental limitations with our present permeation
apparatus do not permit us to test the membranes at higher temperatures, the extrapolated average
permeance and ideal separation factor values for the 1-layer membrane in Figure 3-3 at 500
o
C (a
typical temperature for the use of such membranes for methane steam reforming) is ~ 3.0×10
-7
mol•m
-2
•s
-1
•Pa
-1
with an ideal (He/Ar) separation factor of ~ 171.
Figure 3-3 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes employing
supports made from a single 0.6 µm SiC starting powder via the use of the PS-slip-casting solution, prepared
as described in the Exp. Sec. Measurement temperature 473 K; pressure difference across the membrane
2.41×10
5
Pa; permeate side pressure 1 atm; each individual data point reflects results of three different
membranes prepared under identical conditions.
The permeation results of membranes made using the blend of two starting powders (50 wt.% of
0.6 µm and 50 wt.% of 6 µm size SiC particles) prepared using the conventional (no PS added to
the slurry) slip-casting approach are shown in Figure 3-4, which shows the He permeance and the
ideal (He/Ar) separation factor for the membranes prepared. As the Figure indicates, the initial
supports are more permeable with a permeance of (3.5-4.0) ×10
-5
mol•m
-2
•s
-1
•Pa
-1
as compared to
0
100
200
300
400
500
600
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
Support Slip-casting Layer 1 Dual Layer 2 Dual Layer 3 Dual Layer
Spn Factor of He/Ar
He Permeance/mol.m
-2
.s
-1
.Pa
-1
Permeance
Spn Factor
68
~ (2.25-3.48) ×10
-5
mol•m
-2
•s
-1
•Pa
-1
for the supports
in Figure 3-2 prepared with the single 0.6 µm
powder. It takes again two dual (PS+AHPCS) coatings before a microporous membrane is
prepared with an average He permeance of 2.02×10
-8
mol•m
-2
•s
-1
•Pa
-1
and an ideal (He/Ar)
separation factor of ~367. Adding a third dual layer reduces the He permeance a bit (~1.28×10
-8
mol•m
-2
•s
-1
•Pa
-1
) but improves the idea (He/Ar) separation factor (~450).
Figure 3-4 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes employing
supports made from a blend of 50 wt.% 0.6 µm and 50 wt.% 6 µm SiC particles via the use of a regular
slip-casting solution, prepared as described in the Exp. Sec. Measurement temperature 473 K; pressure
difference across the membrane 2.41×10
5
Pa; permeate side pressure 1 atm; each individual data point
reflects results of three different membranes prepared under identical conditions.
Figure 3-5 shows the permeation results of membranes made using again the blend of two starting
powders, but in this case fabricated using the PS-slip-casting solution prepared by adding PS to
the regular slip-casting slurry. The presence of PS in the slurry, once more, appears to have a
positive effect on the properties of the asymmetric membranes formed by the deposition of the SiC
precursor films. In contrast with the case of the membranes prepared with the blended-powder
0
100
200
300
400
500
600
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
Support Slip-casting Layer 1 Dual Layer 2 Dual Layer 3 Dual Layer
Spn Factor of He/Ar
He Permeance /mol.m
-2
.s
-1
.Pa
-1
Permeance
Spn Factor
69
supports employing the regular slip-casting solution (Figure 3-4), the deposition of one dual-layer
results in a microporous SiC membrane with high permeance of ~4.0×10
-7
mol•m
-2
•s
-1
•Pa
-1
and an
average ideal (He/Ar) separation factor of 72 (a range of 50-90).
Figure 3-5 He permeance (left axis) and (He/Ar) ideal separation (right axis) of SiC membranes employing
supports made from a blend of 50 wt.% 0.6 µm and 50 wt.% 6 µm SiC particles via the use of the PS-slip-
casting solution, prepared as described in the Exp. Sec. Measurement temperature 473 K; pressure
difference across the membrane 2.41×10
5
Pa; permeate side pressure 1 atm; each individual data point
reflects results of three different membranes prepared under identical conditions.
SEM is a valuable technique to provide insight into the preparation of asymmetric inorganic
membranes via the dip-coating technique of polymeric pre-ceramic precursors as it is employed
here. It is useful, for example, to determine the location and thickness of the various layers
deposited, to identify causes of the observed behavior, and to pin-point potential areas of
improvement. Figures 3-6, for example, shows cross-sectional SEM images of membranes
prepared employing the PS-slip-casting solution for supports fabricated using the single 0.6 µm
SiC powder, at various stages of preparation, as we deposit more layers on the support (after their
0
100
200
300
400
500
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
Support Slip-casting Layer 1 Dual Layer 2 Dual Layer 3 Dual Layer
Spn Factor of He/Ar
He Permeance /mol.m
-2
.s
-1
.Pa
-1
Permeance
Spn Factor
70
pyrolysis). The impact of adding the various layers is clear from these images. For example, one
clearly distinguishes the mid-layer scaffold with a thickness ~2-3 µm on which the additional
layers are placed. By the time the 3
rd
dual layer is placed on the top of the scaffold the thickness
of the top layer (scafford+3 SiC layers) has grown to ~5-7 µm, which is typical of the thickness of
top layers we prepare with the aid of PS sacrificial interlayers. As we have also commented in the
past, despite the fact that the PS interlayer approach grows thicker layers than those fabricated
without the use of such layers (~2- 3 µm [259]), the permeances of these membranes are
significantly higher, pointing out the dual role of the PS interlayer: acting as a barrier to prevent
further infiltration (thus the apparently thicker films), and also as a pore former (thus the enhanced
permeances).
Slip-casting layer 1 dual layer
71
2 dual-layers 3 dual-layers
Figure 3-6 Cross-sectional SEM images of SiC membranes employing supports made from a single 0.6
µm SiC starting powder via the use of the PS-slip-casting solution, prepared as described in the Exp. Sec.,
at different stages of preparation.
Figure 3-7 shows the top views of the same membranes coated with the various layers. The surface
of the membrane coated only with the PS-slip-casting layer (after pyrolysis) appears granular,
consisting of mostly submicron-sized particles. The SEM image of the 1 dual-layer membrane,
though it is significantly microporous (see Figure 3-3 for permeation data and corresponding
discussion), does not appear significantly different than the corresponding image of the surface of
the membrane consisting only of the slip-casted layer on the top of the support. This is indicative
of the fact that the first dual layer has formed mostly within the pore structure of the slip-casted
scaffold. Subsequent coatings (2
nd
and 3
rd
) appear to deposit on the top of the mid-layer scaffold.
In the SEM images of such membranes not many of these nanoparticles are clearly distinguishable
any longer, and those for which their outlines are distinguishable appear embedded in a solid mass
of black amorphous SiC layer (this is especially true in the image of the 3 dual-layer coated
membrane). These SEM observations are consistent with the permeation data for the
corresponding membranes, as discussed above.
72
Slip-casting layer 1dual-layer
2 dual-layers 3 dual-layers
Figure 3-7 Top view SEM images of SiC membranes of SiC membranes employing supports made from a
single 0.6 µm SiC starting powder via the use of the PS-slip-casting solution, prepared as described in the
Exp. Sec., at different stages of preparation.
Figure 3-8 shows the cross-sectional images of membranes made from supports fabricated using
the single 0.6 µm SiC powder by employing the regular slip-casting solution. A key difference
when comparing these images with those of membranes fabricated with the PS-slip-casting
solution is that the regular slip-casting solution appears to allow less infiltration in the underlying
73
support of the AHPCS during the subsequent dip-coating steps. Specifically, the particles
constituting the macroporous support remain clearly more distinguishable for the membranes
prepared with the regular slip-casting solution, while this is not the case for the membranes
prepared with the PS-slip-casting solution (the differences are more visible with the images from
the 2-dual-layer and 3-dual-layer membranes). This is consistent with the role of polystyrene as a
pore-former during the formation (via controlled-temperature pyrolysis) of the slip-casted layer.
This also explains why we are able to fabricate microporous SiC membranes with the
deposition/pyrolysis of only one dual-layer on membrane supports coated with the PS-slip-casting
solution (and not with the regular solution), as the more porous slip-casted layer formed offers a
more accessible pore space for the PS/AHPCS dip-coating solution to penetrate and to form a
microporous layer upon pyrolysis.
Slip-casting layer 1 dual layer
74
2 dual-layers 3 dual-layers
Figure 3-8 Cross-sectional SEM images of SiC membranes employing supports made from a single 0.6
µm SiC starting powder via the use of the regular slip-casting solution, prepared as described in the Exp.
Sec., at different stages of preparation.
3.4 Conclusions
We have focused in recent years our attention on the preparation of highly permeable and
mechanically strong porous SiC tubular supports via the pressureless sintering of β-SiC powders,
for potential use in the fabrication of nanoporous SiC membranes. By varying the type of starting
powders used and their composition, the sintering temperature and the amount of sintering aids
utilized we have managed to fabricate supports which exhibit very high fluxes compared to
supports prepared previously by our group and others, and which are also mechanically strong to
withstand the pressure drops required in their use as membrane supports. However, compared to
previous supports, all these new materials have larger porosities and average pore sizes, and greater
surface roughness, characteristics that create unique challenges in terms of preparing quality
membrane employing these supports.
75
The focus of this research, therefore, was on using these supports to prepare quality microporous
SiC membranes. We have reported here results of our efforts with two different types of supports,
one made from a single starting SiC powder and another made from a blend of two starting
powders. Both types of supports were prepared without using any sintering aids, by employing
relatively high sintering temperatures. The task of preparing high-quality membranes using these
substrates was made possible via the use of a novel technique that involves depositing on the
substrates, prior to the deposition of the SiC precursor layers, of a mid-layer scaffold made of
highly uniform SiC nano-powders. Utilizing these newly developed highly porous and permeable,
yet mechanically strong, silicon carbide supports, highly permeable SiC microporous membranes
were fabricated, with permeances which are an order of magnitude higher than membranes
prepared with less permeable supports but still with fairly decent separation characteristics.
The dip-coating technique nevertheless presents unique technical challenges as well. The various
pre-ceramic precursors require dissolution in strong and highly toxic solvents, and infiltration in
the underlying support is unavoidable. This then dictates additional steps for preparing the support
surface prior to film deposition (slipcasting using SiC nanotubes) and the need for multiple
dipcoatings, both having a negative impact on the final membrane throughput and performance.
The use of pore-formers mitigates some of these negative impacts (but on the other hand introduces
additional processing complexity). And though the H2 permeance of our membranes at 500
o
C is
~10
-7
mol m
−2
s
−1
Pa
−1
(very competitive to supported CMSM [136, 273, 274]), significant progress
must still be made before these SiC membranes become suitable for large-scale hydrogen
production/separation applications. By comparison, Pd and Pd-alloy membranes, around 500
o
C,
typically, exhibit permeances ~10
-6
mol m
−2
s
−1
Pa
−1
[275], but Pd membranes are sensitive to H2S
and hydrocarbon impurities, as previously noted. A permeance of 10
-6
-10
-5
mol m
−2
s
−1
Pa
−1
is
76
quoted as being the target range for industrial applications [275]. Furthermore, the next frontier
with these membranes is the ability for fine-tuning their pore size characteristics in order to be able
to expand their range of applications beyond the current focus area of hydrogen separations.
77
4. Fabrication of Nanoporous SiC Membranes via Initiated Chemical Vapor Deposition
(iCVD) Technique
4.1 Introduction
Polymer coatings are traditionally made using liquid-phase processes such as dip-coating, spin-
coating [276, 277], and layer-by-layer assembly [278, 279]. These processes present challenges,
however, for the preparation of asymmetric inorganic membranes due to the unavoidable
infiltration into the underlying macroporous structure because of the presence of surface tension
effects. Chemical vapor deposition (CVD) is traditionally used for the production of inorganic
coatings [280, 281], (see further discussion below), however, it requires the use of very high
temperatures, and thus presents challenges in the preparation of large-area membranes. In this
study we propose instead the use of a novel technique for the preparation of asymmetric inorganic
membranes which is named initiated chemical vapor deposition or iCVD, and takes place at low
substrate temperatures (T <40 °C), thus avoiding many of the challenges that conventional CVD
techniques face in the preparation of these materials.
The iCVD technique is a one-step, solventless (this presents a significant advantage over the use
of dip-coating and spin-coating techniques utilized in the preparation of inorganic membranes via
the pre-ceramic polymer pyrolysis route which requires the use of toxic solvents), substrate-
independent process that can be used to create a wide variety of organic polymer coatings,
including those being used in the preparation of inorganic membranes. Polymer film formation
during iCVD is governed by the classical free-radical polymerization mechanism of vinyl
monomers (see Figure 4-1).
78
Figure 4-1 Schematic showing the mechanism associated with initiated chemical vapor deposition (iCVD).
I-I represents the initiator molecule and M represents the monomer.
During the process, monomer and initiator gases are fed into a vacuum chamber where resistively-
heated wires (typically set to 200 °C) are used to thermally decompose the initiator molecules into
free radicals. Acrylates and acrylamides are typically used as the monomers and tert-butyl peroxide
is the most common initiator. The reactor stage is cooled (typically at 25 °C) in order to increase
the adsorption of monomer to the surface.
The iCVD technique is quite flexible and can be used to prepare a variety of polymeric coatings
whose functionality depends on the nature of the monomer precursor (see discussion to follow
about the preliminary choice of monomers to be used in this study). For example, hydrophobic
coatings can be made using 1H,1H,2H,2H-perfluorodecyl acrylate [282], adhesive coatings can be
made using glycidyl methacrylate [283], and hydrophilic coatings can be made using hydroxyethyl
methacrylate [284]. As noted above, lowering the substrate temperature increases the adsorption
of monomer and initiator to the surface, and thus increases the deposition rate and the molecular
weight of the films formed. Deposition rates as high as 400 nm/min and molecular weights as high
as 200,000 can be achieved with appropriate optimization of the reactor conditions [282]. The
polydispersity of the polymer chains ranges between 1.9 - 2.6 [282]. Most importantly,
79
furthermore, the iCVD process can be easily scaled-up for use in large-scale “roll-to-roll”
processing [285].
Although we used SiC macroporous substrates in this work (see further discussion to follow), the
iCVD process is, in fact, substrate-independent, and inorganic, organic, and biological surfaces
can all be coated using this technique. One advantage of the iCVD technique is that it is quite
flexible and can be used to also coat curved and textured surfaces. In this study disk-shaped SiC
macroporous membranes with a diameter of 1.2 cm and thickness of 3 mm were used as our main
focus is the preparation of nanoporous SiC membranes. Such substrates, in fact, are well-suited
for the iCVD technique because of their relatively high thermal conductivity that helps to minimize
the temperature gradient across the membrane thickness during the film deposition and results in
higher in a higher deposition rate.
The iCVD technique, has, so far, been used to coat in addition to membrane substrates [286],
silicon wafers [287], wires [288], carbon nanotubes [289], and fibers[290]. Since, in addition, the
technique does not require the use of organic solvents, pre-ceramic polymer precursor infiltration
(due to surface tension effects) into the underlying substrate is greatly minimized; as previously
noted, and this presents a great potential advantage for the preparation of inorganic membranes,
where such infiltration significantly degrades membrane throughput and performance. The iCVD
technique, furthermore, is known to prepare polymer films of very uniform thickness (a great
advantage for the preparation of inorganic membranes intended for use in reactive separation
applications). For example, Figure 4-2 shows the cross-sectional scanning electron microscopy
(SEM) images of silicon trenches that have been coated using either liquid phase spin-coating
(image on the left) or iCVD [291].
80
Figure 4-2 Silicon trenches coated using a) liquid-phase spin-coating versus b) iCVD [291].
The trench coated using the spin-coating technique appears to only have a polymer film at the
bottom of the trench and the impact of surface tension effects are thus obvious; on the other hand,
the trench that has been coated using iCVD has a uniform polymer layer coating along the top,
bottom, and sidewalls of the trench.
Plasma CVD has also received attention in recent years for the preparation of polymer coatings
via in situ polymerization but also for the preparation of inorganic membranes films via the
decomposition of precursor monomers. The plasma process uses a high-energy source, however,
and presents significant challenges for the preparation of large-area membrane thin films. The
iCVD process, on the other hand, uses resistively-heated wires, whose temperature is set to 200
°C, as the energy source for the initiation step. This low-energy process (0.01 W/cm
2
) only breaks
the initiator molecule but leaves the monomer molecules completely intact. The plasma CVD
process, on the other hand, results in the partial decomposition of the monomer molecules, thus
reducing the concentration that is available for polymerization [292]. Additionally, in the plasma
processes there is a competition between etching and deposition [293]; whereas, the iCVD process
involves only deposition and no etching. Figure 4-3, for example, shows the cross-sectional SEM
images of silicon trenches that have been coated using either iCVD (image on the left) or pulsed-
a) b)
81
plasma polymerization[294]. The trench coated with the pulsed-plasma technique appears to have
a non-uniform coating layer, the likely cause being the competition between the etching and
deposition processes; on the other hand, the trench coated using iCVD has a uniform polymer
coating layer along the top, bottom, and sidewalls of the trench.
Figure 4-3 Silicon trenches coated using a) iCVD versus b) pulsed plasma CVD [294].
In the iCVD process, the deposition rate of the polymer film and the molecular weight of the
polymer chains are proportional to PM/PSat where PM is the partial pressure and PSat is the saturation
pressure of the monomer. The iCVD process is, typically, run at values of 0.4 < P M/PSat< 0.7 in
order to attain enhanced monomer adsorption on the surface, while at the same time also preventing
condensation. The value of PM can be increased by either increasing the molar flow-rate of the
monomer into the reactor or by increasing the total pressure of the reactor. The saturation pressure
corresponding to stage temperature T can be approximated using the Clausius-Clapeyron equation:
PSat=Aexp(-∆HVap/RT). Therefore, the value of PSat can be decreased by decreasing the stage
temperature. Figure 4-4a, for example, shows a plot of the deposition rate of poly(1H,1H,2H,2H-
perfluorodecyl acrylate) (PPFDA) onto a silicon wafer as a function of the substrate’s temperature
at a constant reactor pressure of 100 mTorr and a constant monomer flow-rate of 0.31 sccm.
82
Figure 4-4 Deposition rate of PPFDA onto silicon wafers as a function of a) substrate temperature and b)
P M/P Sat [282].
It can be seen that the deposition rate increases with decreasing substrate temperature. The
maximum deposition rate for this series of experiments (144 nm/min) was achieved at the lowest
substrate temperature used (39
o
C). Figure 4-4b shows a plot of the deposition rate as a function
of PM/Psat at a constant reactor pressure of 100 mTorr and a constant substrate temperature of 44
o
C. In these experiments, the monomer flow-rate was varied from 0.13 sccm to 0.67 sccm while
the monomer saturation pressure was kept constant by maintaining the substrate’s temperature.
Again, it can be seen that the deposition rate increases with P M/Psat. The maximum deposition rate
for this series of experiments (400 nm/min) was achieved at the highest P M/Psat ratio used. The
molecular weights of the resulting coatings were measured by dissolving the PPFDA in fluorinated
solvent (1,1,1,3,3,3-hexafluoro-2-propanol) and then using gel permeation chromatography
(GPC). Table 4-1 shows the weight-averaged molecular weight, M w, and the polydispersity of the
samples. It was found that similarly to the deposition rate, the M w also increases with decreasing
substrate temperature and increasing monomer flow-rate. The polydispersity ranged between (1.9
- 2.6) with no observable trend with respect to the substrate temperature or the monomer flow-rate.
83
This range in polydispersity has also been observed for the iCVD polymerization of other acrylates
and is typical of free-radical polymerizations [295].
Table 4-1 Weight Average Molecular Weight From GPC Measurements ([282])
Substrate Temperature Series
Substrate Temperature (
o
C) PM/Psat Mw Polydispersity
44 0.38 177300 2.27
49 0.27 114300 2.63
54 0.20 42800 2.01
Monomer Partial Pressure Series
Substrate Temperature (
o
C) PM/Psat Mw Polydispersity
39 0.50 127800 1.89
39 0.45 109300 1.91
39 0.38 91100 1.91
4.2 Previous Studies
Though the proposed process may have broader applications for the preparation of a variety of
other inorganic films (e.g., diamonds, graphene, etc.) via the pre-ceramic polymer pyrolysis route,
our current focus is nanoporous SiC membranes, which together with other types of inorganic
membranes, have recently received much attention, due to their high permeability and
thermochemical stability and resistance to harsh environments as well as due to their applications
as novel hydrogen sensors and storage media [238, 259, 260, 296]. We have been at the forefront
of efforts for the fabrication and characterization of these SiC nanoporous membranes, and of their
use in both conventional and reactive separation applications of important fluid mixtures.
84
Our efforts in this area are motivated by the growing interest in the H 2 economy, which necessitates
the development of robust nanoporous films (to be used as membranes and sensors), storage
materials (e.g., SiC nanotubes) as well as mechanically strong, highly permeable membranes for
gas separation especially in high-temperature processes (>800C) related to H 2 production such as
in membrane reactors (MRs) for steam reforming of methane (SRM), a process which supplies
nearly 50% of the world-wide demands for hydrogen, as noted in the Introduction of this Thesis.
SiC is a promising material for such applications due to its many unique properties, such as high
thermal conductivity [297], thermal shock resistance [256], biocompatibility [298], resistance in
acidic and alkali environments [241], chemical inertness (e.g., towards steam, H 2S, NH3, HCl of
concern for H2 production from biomass and coal), and high mechanical strength [299, 300]. Other
inorganic membranes, such as ceramic (e.g., alumina, silica, and zeolite) and metal (Pd, Ag, and
their alloys) membranes also have a number of important potential industrial applications but have,
so far, proven unstable in high-temperature reactive applications in which steam and H 2S are
present. Supported CMS membranes (CMSM), previously developed by our group [273, 274, 301-
303], and currently being commercialized [136] by M&PT, exhibit many similarly desirable
properties as SiC membranes (e.g., inertness towards H 2S), and they show good potential for a
variety of commercial uses (e.g., H2/CH4, propane/propylene separations, etc.) as well; they are
unstable, however, in the presence of O2/steam at temperatures >300
o
C, typically encountered in
reactive separations for H2 production.
In our previous and recent studies, we used two different approaches to prepare SiC membranes.
One involves the use of high-temperature CVD [238], and the other is based on the pyrolysis of
polymeric pre-ceramic precursors [238, 259, 260, 296]. These are asymmetric membranes, made
from a relatively thick (1-2 mm) macroporous SiC support that provides mechanical strength to
85
the composite material during practical applications. (Efforts to improve on the properties of SiC
substrates are ongoing in our group [268, 304], as they are important in determining the membrane
properties and cost; alumina substrates, particularly γ-alumina [243], used by most other groups
[250, 252] have different thermal expansion coefficients than the nanoporous SiC film, and are,
therefore, prone to cracking during thermal cycling.) The preparation techniques deposit a thin
nanoporous SiC film on the support that endows the membranes with high selectivity for the
separation of fluid mixtures. The high-temperature CVD approach prepares membranes (using tri-
propyl-silane or TPS as a precursor) with a He permeance ranging from 8.1×10
−8
to 1.7×10
−6
mol
m
−2
s
−1
Pa
−1
with a He/N2 ideal selectivity ranging from 4 to >100 [238]. Prior to our research,
there were a handful of studies using CVD/CVI to prepare porous SiC membranes [235, 305, 306].
None of these studies, however, resulted in defect-free nanoporous SiC membranes. The H 2-
selective SiC membranes exhibit excellent thermal stability at 500 C for >1000 h and are also
stable in high-pressure and temperature steam. However, the preparation procedure involves
multiple steps, which makes it costly and requires treatment at high temperature (1000
o
C), which
places a great burden on glass end-seals and membrane housing during the CVD process. More
importantly, the quality of the final product is difficult to predict and control, and from a
manufacturing standpoint, the advantage of the on-line control of the CVD technique is, therefore,
lost. The iCVD process, experimented here, shows great promise to overcome the many processing
challenges facing the high-temperature CVD technique.
4.3 Fabrication of Disk-shaped Nanoporous SiC Membranes
4.3.1 Research Approach
In this experimental study, the iCVD process was used to deposit silicon-containing polymer films
onto macroporous SiC flat-disk substrates (the choice of SiC is because other types of substrates,
86
particularly γ-alumina used in most investigations have different thermal expansion coefficients
than the nanoporous SiC film deposited, and are, therefore, prone to cracking during thermal
cycling). Three types of supports with average porosities listed in Table 4-2 were used for these
experiments. These substrates are made of pure 0.6 µm (HSC-059N, from Superior Graphite Co.,
Chicago, IL), pure 80 nm
1
(from US Research Nanomaterials Inc.), and 50% 0.6 µm/50% 80 nm
β-SiC via the pressureless sintering of the powder. The use of appropriate size of SiC powder to
prepare the supports is the key in the preparation of defect-free membranes. Before using all these
substrates in iCVD we deposited via slip-casting a top layer prior to deposition in order to prevent
polymer film infiltration. The approach for the deposition of the slip-casted layer is described in
greater detail in Chapter 3.
1
Silicon Carbide (SiC) Nanopowder Purity: 99+%, Silicon Carbide (SiC) Nanopowder Average Particle size (APS): <80nm;
Purity Free Si Free C
99+% 0.24% 0.76%
87
Table 4-2 Average Porosities of different support material.
Support content Average porosity
pure 80 nm SiC 0.44
pure 0.6 µm SiC 0.42
50% 0.6 µm/50% 80 nm SiC 0.46
As noted above, the iCVD process proceeds via a classical free radical polymerization mechanism
(see Figure 4-1), and has been successfully utilized in recent years by Gupta et al. [307-310] to
prepare variety of polymeric films whose functionality depends on the nature of the monomer
precursor. We are not aware, however, of any other group using the same technique to prepare
asymmetric, nanoporous inorganic membranes, despite the promise the technique shows for
preparing such materials (and to the best of our knowledge, the pyrolysis of iCVD films has never
been explored before, either). The objective of this project, therefore, was to use iCVD to prepare,
for the first time , inorganic membranes via the pyrolysis of pre-ceramic polymer films deposited
on macroporous SiC substrates.
In this study, we used iCVD to deposit a silicon-containing pre-ceramic polymer, poly(1,3,5-
tetravinyl-1,3,5-tetramethylcyclotetrasiloxane), which then acts as precursor for the preparation,
upon pyrolysis, of amorphous Si xOyCz nanoporous membrane films onto macroporous SiC
substrates. The molecular structure of the corresponding monomer can be seen in Figure 4-5. Prior
to the pyrolysis of the polymer deposited on the SiC substrates, we used the DRIFTS technique to
study the controlled-temperature pyrolysis of the aforementioned pre-ceramic polymers coated on
the surface of alpha alumina powder. These studies, described below, helped to identify the proper
pyrolysis temperature of the polymer films deposited on the macroporous SiC substrates and to
assure the production of microporous Si xOyCz ceramic films on these supports.
88
After the deposition of the pre-ceramic polymer onto all SiC substrates with different composition,
already coated with a single slip-casting layer, via iCVD, they were pyrolyzed in a tubular furnace
to form the desired ceramic material. The resulting membranes were then tested in the membrane
permeation apparatus to measure their single-gas permeance and ideal separation factors of He and
H2 over Ar. SiC microporous materials, as previously noted, show potential to find important
applications in the design and synthesis of fuels and chemicals as separation and reactive media
(e.g., adsorbents, membranes and catalysts, etc.), and sensors. The fabrication process involves a
series of tailored reactions via which the original precursor converts into the final ceramic. Our
efforts in this area are motivated by the growing interest in the H2 economy, which necessitates
the development of robust materials for use in the high-temperature and pressure processes related
to H2 production.
Figure 4-5 Chemical structures of silicon-containing monomer.
4.3.2 Pyrolysis of Pre-ceramic Polymer
As previously noted, in order to identify the appropriate pyrolysis conditions for the pre-ceramic
polymer generated by the iCVD process we used the FTIR technique. For that the polymer was
deposited on DX type alpha-alumina powder with average particle diameter of 1 µm, and the
89
composite material was then pyrolyzed under well-controlled conditions in order to investigate the
change in its molecular structure at different pyrolysis temperatures. Though the preparation of the
nanoporous membranes involves the pyrolysis of polymer films on SiC substrates, the alumina
powder was chosen for the fundamental pyrolysis investigations due to very high absorbance of
radiation by the SiC powder in the mid-IR region (400-4000 cm
-1
) and the resulting distortion in
the IR spectrum.
We have mostly carried out the fundamental pyrolysis studies by first depositing the pre-ceramic
polymer on the surface of alumina powder and then pyrolyzing it inside the DRIFTS cell, while
monitoring in real time the IR spectra for the composite material for temperatures as high as the
maximum temperature the cell can attain. We have also studied the polymer pyrolysis for
temperatures higher than what the DRIFTS cell is allowed to operate at, by pyrolyzing the alumina
powder/polymer composite material in a tubular furnace at different temperatures and then
transferring the resulting material in the DRIFTS cell and generating its IR spectra.
Once the alumina powder is coated with pre-ceramic polymer poly(1,3,5-tetravinyl-1,3,5-
tetramethylcyclotetrasiloxane), it is then mixed with a non-absorbing matrix, namely Barium
Fluoride (BaF2) with a melting point of 1368 ℃ (the commonly utilized KBr material melts around
734 ℃, which makes it not a good candidate for use at temperatures as high as 1000 ℃).
BaF2shows almost no absorbance in the mid-IR region and is used to dilute the sample and thereby
minimize the relative contribution from surface reflections. The distance the light travels inside
the sample before it is scattered or reflected back from the surface increases with the particle size,
thus reducing the amount of light reappearing from the sample. If the sample particles are much
larger than a few tens of microns in size, then many mid-IR bonds become totally absorbing.
Therefore, usually a finely ground mixture of sample powder and a non-adsorbing IR matrix (2-5
90
µm average diameter) is used in DRIFT spectroscopy. The powder mixture to be tested is loaded
into the sample cup of a FTIR spectrophotometer (Nicolet iS10, Thermo Scientific). The DRIFTS
assembly consists of a Collector II (from Spectra-Tech), the reaction chamber, and a power box
equipped with a temperature controller. The Collector II is mounted onto the FTIR instrument
together with the reaction chamber and the heating assembly and is connected to the power box,
see Figures. 4-6a and 4-6b. The chamber is basically an enclosed dome that allows you to control
the environmental conditions of the sample. This highly versatile chamber allows the exposure of
the sample to simulated reaction conditions. When used with the COLLECTOR II, the High
Temperature/High Pressure Chamber can reach temperatures up to ~900-1000 °C at pressures
close to ambient pressure or lower temperatures of up to 400 °C at higher pressures up to 1500 psi,
since the safe operating temperature of the cell decreases (See Figure 4-7) as the pressure increases.
The chamber dome features a unique zinc selenide (ZnSe) window, which is non-hygroscopic, has
low absorption in the red end of the visible spectrum, and is ideally suited to withstand these
conditions. The sample cup is located inside the dome. A ceramic heater equipped with a
thermocouple is in intimate contact with the sample for direct sample temperature measurement.
There is a coil around the dome for water circulation to keep the dome from being overheated.
There is an inlet and outlet in the chamber for the pyrolysis gas, Ar in our case, to flow in and out
of the chamber. The heating process is controlled by a temperature controller and the IR spectrum
is collected at a certain temperature once it is set by the controller. In this study, in-situ pyrolysis
of the sample is carried out under a constant flow of Ar at 30 psi, up to a maximum temperature
of 1000 ℃, via the use of DRIFT spectroscopy technique.
91
a)
b)
Figure 4-6 a) Experimental setup for pyrolysis of the pre-ceramic polymer poly(1,3,5-tetravinyl-1,3,5-
tetramethylcyclotetrasiloxane) inside the DRIFTS cell, b) Collector II with the reaction chamber and
heating assembly
92
Figure 4-7 The pressure/temperature chart of the chamber.
As noted above, for pyrolysis temperatures above the operating range of the DRIFTS cell the pre-
ceramic polymer was pyrolyzed in a tubular furnace and the pyrolyzed material transferred to the
DRIFTS cell for further studies. For this, the polymer-coated powder was spread on the surface of
an alumina plate and placed inside the tubular furnace. The sample was then heated first from room
temperature to 200
o
C at a rate of 2
o
C/min, where it was kept for 1 h, then to 400
o
C, where it
dwelled for 1 h, then to 750
o
C, where it was kept for another 1h, and finally to a pre-selected
maximum temperature (e.g., 1100
o
C) where it was kept again for an additional 2 h. Subsequently,
the sample was cooled down to room temperature in flowing Ar with a cooling rate of 3
o
C/min.
Three different maximum pyrolysis temperatures of 750
o
C, 900
o
C, and 1100
o
C were used to
analyze the effect of pyrolysis temperature on the chemical properties of the coated material. The
pyrolyzed sample was then analyzed using the Collector II and FTIR spectroscopy, the results of
which are provided in the result and discussion section.
93
4.3.3 Preparation of the Disk-Shaped SiC Substrate for iCVD
Three types of supports with average porosities listed in Table 4.2 were used for use in the iCVD
experiments. These supports were made via the pressureless sintering of pure 0.6 µm (HSC-059N
β-SiC powder, from Superior Graphite Co., Chicago, IL), pure <80 nm β-SiC powder (from US
Research Nanomaterials Inc.), and a mixture of 50% 0.6 µm/50% 80 nm β-SiC powders.
In order to prepare SiC disk-shaped porous supports, the β-SiC powders were first mixed together
at a proper weight ratio. Boron carbide (0.1wt%) and liquid phenolic resin (4wt%) was then added
to the mixture of powder as sintering aids, using acetone as the dispersing medium. The resulting
slurry of powders and sintering aids was then fully mixed in an ultra-sonicator for 1 h, and
subsequently dried in a fume-hood for 3 days. Afterwards, the dried slurry was thoroughly ground
in a mortar, and small amounts of oleic acid and toluene were added to it as pressing aids. These
powders were then pressed into the 3 mm thick (1.2 cm in diameter) disk-shaped green support
samples with a pressure of 6.68 MPa for 2 min. They were subsequently placed in a high-
temperature graphite furnace (Thermal Technology, Inc., Model 1000-3060-FP20), where they
were heated (3
o
C /min) in flowing He to a pre-set sintering temperature of 1900
o
C, where they
stayed for 3 h, and then cooled down (6
o
C /min) to room temperature. The support disks used in
membrane preparation in this study were treated in flowing synthetic air at 450
o
C (with the
purpose of oxidizing any potential carbon contaminants present), sonicated several times in
acetone, and then dried prior to membrane film deposition. A slip-casting step was used for
deposition of a mid-layer scaffold, which was employed to condition the support surface to make
it amenable to further SiC precursor film deposition and reduce the infiltration of the depositing
pre-ceramic polymer during iCVD process. For that, we utilized a solution that consists of 10 wt.%
AHPCS (SMP-10, Starfire Systems, Inc.) and 90 wt.% hexane (HX0299-5, EMD Chemicals). In
94
the solution 1.25 wt.% polystyrene (PS, GPC grade, M w=2500, Scientific Polymers Products, Inc.)
was added together with SiC particles (5 wt.%), which were obtained by separating the (100-200
nm, as verified by scanning electron microscopy (SEM) [268]) fraction of the 0.6 µm -SiC
powder via a sequential precipitation/separation approach beginning with a slurry of the original
SiC powder in acetone [258], and separating (specifically, 7g of the 0.6 µm SiC powder was mixed
with 200 ml of acetone in a beaker, with the top 50 ml of the 200 ml solution being separated) and
drying the lighter particles. For slip-casting, the slurry was first sonicated for 1 min and the support
disks were held in it for 12 s, and then drawn out of it at a speed of 0.28 mm/s. The coated disk
supports were heated in flowing Ar in a tube furnace (Lindberg/Blue, Model STF55433C) at a rate
of 2
o
C/min, first to 200
o
C, where they were kept for 1 h, then to 400
o
C, where they were also
kept for 1 h, and finally to 750
o
C, where they were kept for an additional 2 h. Subsequently, they
were cooled down to room temperature in flowing Ar with a cooling rate of 3
o
C/min. The reason
for the relatively slow heating (and holding at 200
o
C) is that prior studies [239, 240, 259, 260]
indicate that using such heating rates and treatment at lower temperatures result, generally, in better
cross-linked amorphous SiC materials.
Helium porosity measurements were carried out for the three substrates with different initial
powder compositions, the results of which are shown in Table 4-2. For such measurements, the
helium gas is fed into a reference chamber that has the volume of V R at pressure P1. Then a valve
connecting the two chambers is opened to let the helium gas fill another chamber with volume of
Vc occupied with a solid (non-porous) stainless steel samples with total solid volume of V s. Once
the equilibrium pressure is reached in the two chambers the new pressure, P 2, is recorded. The
following equations apply:
P1V1=P2V2 @ constant T=300K (4-1)
95
P1VR=P2(VR+Vc-Vs) (4-2)
And rearranging and rewriting the equation in terms of Vg, the linear equation becomes as follows:
Vs=-VR×(P1-P2)/P2+ Vc (4-3)
The reference volume, VR, and the volume of the second chamber holding the solid stainless steel
samples, Vc, are obtained using multiple measured data points for V s vs. (P1-P2)/P2, where P2 is
the equilibrium pressure in both the reference and the sample chambers after the valve is opened.
Once the values of VR and Vc are calculated using linear regression method, the second chamber
is loaded with the SiC disk samples with a total volume of V d. The same pressure measurement
process is repeated again for SiC sample disks. Since the slope, -V R, and the intercept of the linear
equation, Vc, are known from the previous step, one can calculate the new values of V s for sample
disks using the previously obtained linear equation for each pressure ratio measurement ((P 1-
P2)/P2) data point, and to calculate the arithmetic average porosity, ε
av
, of the sample disks using
the equation below:
ε
i
=
V
d
-V
s
V
d
(4-4)
ε
av
=
∑ ε
i
n
i=1
n
(4-5)
As can be seen from Figure 4-8, representing the helium Compressibility Factor (Z) Vs. Pressure (P,
psi/atm) for various temperatures up to 300K, at the measuring pressure (~98 psi) and temperature
(~300K), the compressibility factor of helium, Z, is very close to 1; or in other words, deviation
from ideal gas behavior is negligible; therefore, it can be assumed that the helium gas filling the
96
chamber behaves pretty much like an ideal gas and as a result, one can use the ideal gas equation
of state, PV=nRT, for helium to calculate for the porosity of the support samples.
Figure 4-8 Helium Compressibility Factor (Z) Vs. Pressure (P, psi/atm) for various temperatures up to 300K
[311].
4.3.4 iCVD Process and Preparation of Nano-porous SiC Membrane
Subsequent to the preparation of the disk-shaped SiC substrates, a number of them were placed
inside the iCVD reactor. In order to improve the heat transfer from the bottom of the reactor to the
substrate, the bottom of each sample was covered with aluminum foil. This way, the hope was that
97
heat would transfer faster resulting in in a lower temperature gradient across the sample with its
temperatures being closer to that of the reactor surface, which is cooled from the bottom through
circulation of water to adjust the deposition rate of the polymer. For all samples, 1,3,5-tetravinyl-
1,3,5-tetramethylcyclotetrasiloxane monomer was used together with butyl peroxide as the
initiator to deposit a thin polymeric film on the surface of the SiC substrates. The polymer film
formation during iCVD occurred through a classical free-radical polymerization mechanism of the
vinyl monomers. The monomer and initiator molecules were adsorbed onto the macroporous
substrate and polymerization was initiated and propagated via a low-energy input. The rate of
polymer deposition was estimated using a silicon wafer reference material under identical
deposition conditions. For that purpose, an interferometer, which employs wave interference to
make precise measurement of the length of displacement in terms of wavelength, was used. The
measurement was then converted into thickness of deposited film using Fast Fourier Transform
equation based on the time, number of periods, and the refractive index of the coating as it is being
deposited.
For all samples coated via iCVD, including both the alpha-alumina powder sample for FTIR
pyrolysis analysis and the SiC substrates, a deposition rate of 4 nm/min was achieved at the
monomer flow rate of 0.14 sccm and monomer line temperature of around 65 ºC. The system
pressure was set to 175 mtorr, the filament temperature was 250 ºC and the substrate temperature
was approximately 35 ºC. Subsequent to the iCVD process, the coated substrates were heated
(pyrolyzed) in a tubular furnace to form the desired ceramic membranes. The maximum
temperature of the pyrolysis process was selected based on the results obtained from the IR
spectroscopy analysis of the coated alpha-alumina powder pyrolyzed in the tube furnace.
Specifically, the heating process involved a series of temperature increments from room
98
temperature to 200
o
C at a rate of 2
o
C/min, where the samples were kept for 1 h, then to 400
o
C,
where they dwelled for 1 h, then to 750
o
C, where they dwelled again for another 1h, and finally
to 1100
o
C where they were kept for an additional 2 h. Subsequently, the sample was cooled down
to room temperature in flowing Ar with a cooling rate of 3
o
C/min.
4.3.5 SiC Membrane Permeation Test
The resulting membranes were tested individually using He, H 2, and Ar as test gases to measure
the permeance of each as well as the ideal separation factor (i.e., the ratio of He and H 2 permeances
over that of Ar). The permeation characteristics of the resulting membranes were tested using the
set-up shown in Figure 4-9, which was also previously utilized for the tubular membranes (See
Sec. 3.2).
Figure 4-9 Membrane permeance test set-up [312].
99
The Disk shaped membranes were placed in between the two half-cells of the permeation-test unit
(metal Rings) sealed by red silicone O-rings of the same outside diameter. He, H 2, and Ar were
fed at a certain flow rate through the bottom cell, and the gases permeated exited from the top cell.
During the experiments, the pressure drop across the membrane was maintained constant. The
pressure in the permeate-side was measured by a pressure gauge (OMEGA Engineering,
DPG1000B-15G) and was maintained atmospheric. The pressure in the feed-side was adjusted
with the aid of regulators of the upstream gas cylinders and a needle valve on the reject tube. A
differential pressure transducer was used to measure the pressure across the feed and permeate
sides. The flow rate of the gas exiting the permeate side was measured using bubble-flow meters
of different volumes. For each membrane several duplicate measurements at a certain temperature
and pressure drop across the membrane were carried out in order to minimize the errors. The
permeance of each membrane was then determined by dividing the flow of gas (mol/s) by the
pressure drop (in Pa) and the membrane area (m
2
) through which the transport occurred. The ideal
separation factors of He and H2 over Ar were then calculated as the ratio of the two corresponding
measured permeances. The permeance test results of disk membranes with different support
compositions fabricated via the use of iCVD technique have been reported in the result and
discussion section of this chapter.
4.3.6 Results and Discussion
4.3.6.1 FTIR Study of Poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane) Pyrolysis
Process
The DX type alpha-alumina powder was coated with a thin film of poly(1,3,5-tetravinyl-1,3,5-
tetramethylcyclotetrasiloxane) using the procedure described in Sec. 4.3.4. IR spectroscopy was
performed for the analysis of the sample. Figure 4-10 shows the IR absorbance spectrum for the
uncoated DX type alpha-alumina powder. Figure 4-11 shows the IR absorbance spectrum of the
100
alumina powder coated with poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane). For both
cases, the samples were diluted using non-absorbing KBr matrix.
Figure 4-10 The IR absorbance spectrum of alpha-alumina powder (DX Type (1µm particle size), Electron
Microscopy Sciences), diluted with KBr.
101
Figure 4-11 The IR absorbance spectrum of alpha alumina powder coated with pre-ceramic polymer
poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane), diluted with KBr.
Previous studies on dehydroxylation and high temperature phase formation in sol-gel processes
has shown that if the coordination is octahedral (AlO 6), the Al-O stretching and bending modes
are expected in the region 500-750 cm
-1
and 330-450 cm
-1
, respectively [18-20]. However, a
tetrahedral coordination (AlO4) is expected to give stretching modes in the narrow range 750-850
cm
-1
and bending modes between 250 and 320 cm
-1
[313]. The two bands with peaks at 590 and
442 cm
-1
in Figure 4-11 are attributed to the existing alpha-alumina phase [313].
The IR spectroscopy analysis of the alpha-alumina powder coated with poly(1,3,5-tetravinyl-1,3,5-
tetramethylcyclotetrasiloxane) revealed the following peaks:
Si-O-Si The very intense and broad band appearing at 1095–1089 cm
-1
and the shoulder at
around 1200 cm
-1
are attributed to the transversal optical (TO) and longitudinal optical (LO) modes
Si-O-Si
Stretching
vibration
Si-CH3
Stretching
vibration
Si-CH3
Stretching
vibration
C-H
Stretching
vibration
of methyl
groups
H-O-H
Bending
vibration
Alpha-
alumina
phase
Si-O-Si
Bending
vibration
H-O-H
Bending
vibration
102
of the Si-O-Si asymmetric stretching vibrations, respectively. The symmetric stretching vibrations
of Si-O-Si, on the other hand, appear at 800 cm
-1
and its bending mode appears at 469–467 cm
-1
[314, 315].
C-H stretching vibration of methyl groups at 2960 cm
-1
Si-CH3 stretching vibrations at 1400 cm
-1
and ~1250 cm
-1
H-O-H bending of water at 1600 cm
-1
The samples coated with the pre-ceramic polymer were then pyrolyzed outside the DRIFTS cell
in a tube furnace (Lindberg/Blue, Model STF55433C) at three different temperatures of 750, 900,
and 1100
o
C (see Figure 4-12). As mentioned earlier, the heating process involved series of
temperature increments from room temperature to 200
o
C at a rate of 2
o
C/min, where the samples
were kept for 1 h, then to 400
o
C, where they dwelled for 1 h, then to 750
o
C, where they dwelled
again for another 1h, and finally to selected temperature (750, 900, and 1100
o
C) where they were
kept for an additional 2 h. The samples were then cooled down to room temperature in flowing Ar
with a cooling rate of 3
o
C/min (for the case of 750
o
C as the maximum pyrolysis temperature,
samples were kept at 750
o
C for 2 h and were subsequently cooled down to room temperature with
a cooling rate of 3
o
C/min). As can be seen from Figure 4-12, the Si-CH3 stretching vibrations at
1400 cm
-1
and ~1250 cm
-1
have started vanishing at 750
o
C and they no longer exist at 900
o
C. The
C-H stretching vibration of methyl groups at 2960 cm
-1
started to deform near 750
o
C. The intense
and broad Si-O-Si stretching vibration in 1095–1089 cm
-1
; however, started to deform at
temperatures as high as 1100
o
C, and the expected Si-O-C grouping showed a band with high
intensity near 1100cm
-1
and bands of variable intensity exist in the range of 620-840 cm
-1
[316,
317].
103
Figure 4-12 The IR absorbance spectrum of alpha alumina powder coated with pre-ceramic polymer
poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane) and pyrolyzed outside the DRIFT, diluted with
KBr.
104
The IR spectroscopy analysis of the alpha alumina powder coated with the pre-ceramic polymer
was also carried out inside a Diffuse Reflectance Infrared Fourier Transform (DRIFT)
Spectroscopy cell and the real time change of the IR absornbance spectrum was recored as the
temperature was raised up to the maximum rated temperature of 1000
o
C for the DRIFT cell. The
corresponding results are shown in Figure 4-13 for various temperatures. Similar behavior to that
of pyrolysis outside the DRIFTS cell was also seen here when the sample powder was pyrolyzed
inside the DRIFTS cell. Unfortunately, the limitations regarding the maximum application
temperature of the DRIFT cell did not allow us to go to higher temperatures. Figure 4-14 Shows a
comparison between the IR absorbance spectra of alpha alumina powder coated with pre-ceramic
polymer poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane) pyrolyzed at room temperature
(a) and 1000 ºC (b). Again, as can be seen from the figure, the Si-CH 3 stretching vibrations at 1400
cm
-1
and ~1250 cm
-1
no longer exist at 1000
o
C. The C-H stretching vibration of methyl groups at
2960 cm
-1
has been deformed. The intense and broad Si-O-Si stretching vibration in 1095–1089
cm
-1
has started to deform at and the expected Si-O-C group bands have started to appear near
1100cm
-1
and in the range of 620-840 cm
-1
. It should be noted that for the case of pyrolysis inside
the drift cell, BaF2 was used as the non-absorbing matrix to dilute the sample since the more often
used KBr matrix could not resist higher temperatures and would melt at 734 ºC.
105
Figure 4-13 The change in IR absorbance spectrum of alpha alumina powder coated with pre-ceramic
polymer poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane), as being pyrolyzed at different
temperatures from room temperature to 1000 ºC, diluted with BaF 2 matrix.
Room Temperature
600 ºC
700 ºC
750 ºC
800 ºC
900 ºC
950 ºC
1000 ºC
106
a)
b)
Figure 4-14 The IR absorbance spectra of alpha alumina powder coated with pre-ceramic polymer
poly(1,3,5-tetravinyl-1,3,5-tetramethylcyclotetrasiloxane); a) room temperature, b) pyrolyzed at 1000 ºC,
both diluted with BaF 2 matrix.
107
4.3.6.2 The Permeation Test Results of SiC Disk Membranes Prepared via iCVD Method.
The permeance test result of the SiC membranes prepared via iCVD and the corresponding ideal
separation factors of He and H2 over Ar are listed in Table 4-3 below. As can be seen from the
results, membranes composed of 50% of particles with an average diameter less than 80 nm and
50% of particles with average diameter of 0.6 μm showed higher He and H2 permeances, 2.74E-
07 mol/m
2
.Pa.s and 1.89E-07 mol/m
2
.Pa.s, respectively as well as higher ideal He over Ar
separation factor of 25.15 and H2 over Ar separation factor of 17.39, while the membranes made
of pure <80 nm particles showed almost an order of magnitude smaller permeances (4.19E-08
mol/m
2
.Pa.s) compared to the other two sets of membranes. Despite a higher He permeance of
7.45E-07 mol/m
2
.Pa.s, the membranes made of pure 0.6 μm SiC particles showed less ideal
separation factors for both He and H2 over Ar. It should be noted that the results shown in Table
4-3 for each set of membranes with different composition of support are the reported average
permeances and ideal separation factors of three different samples prepared under the same
condition.
Table 4-3 He and H2 permeance results and ideal separation factors of He and H2 over Ar for Slip-casted
SiC support coated via iCVD, pyrolyzed at 1100ºC and tested at 35psig and 473K.
Slip-Casted SiC Support+Polymer Coating via iCVD (+ 4wt% Phenolic Resin and 0.1wt%
Boron Carbide Used as Sintering Aid)
Composition of the
Support
He permeance
(mol/m
2
.Pa.s)
H2 permeance
(mol/m
2
.Pa.s)
He/Ar
Separation
Factor
H2/Ar
Separation
Factor
100% <80 nm 4.19E-08 2.25E-08 20.7 11.0
100% 0.6 7.45E-07 4.81E-07 16.4 10.6
50%(<80nm)/50%(0.6 ) 2.74E-07 1.89E-07 25.1 17.4
108
4.3.7 Conclusions
Based on the obtained results for the disk-shaped SiC membranes prepared via the use of iCVD
technique. It was revealed, so far, that supports with a large average pore size and/or wide PSD
require an intermediate layer to make them appropriate for iCVD due to a high infiltration of the
depositing polymer into the substrate. The use of smaller size of particles in preparation of the
support for the membrane (<80nm), on the other hand, improved the separation factor but showed
relatively lower gas permeances. It is suggested that an optimum configuration of particles for the
support may be achieved by using combination of particles with different sizes as the membranes
fabricated via the use of 50% <80nm particles and 50% 0.6μm have shown to have higher
separation factors while maintaining a high gas permeance compared to supports and
corresponding membranes made of pure <80nm SiC particles. More studies are yet to be done to
investigate more in depth the parameters that play a key role in determining the final quality of
such membrane, such as the iCVD conditions, the type of monomer used for the polymerization
and deposition of the polymer (other types of monomers such as diphenyldivinylsilane or
dimethyldivinylsilane are suggested for further investigations in the future) as well as the structure
of the intermediate layer.
109
5. Suggestion for Future Work
In terms of our feasibility study on steam reforming of methane using landfill gas as a feed to the
reformer, the results of the study, “proof of concept” research are promising; however, it is still
early to begin commercialization or marketing. Significant additional development is required to
better understand the process and how it should be optimized. It is important, for example, that
further process development continues with a larger, turbocharged, lean-burn engine. This will,
then, allow a direct comparison of engine performance and NO x emissions using both the raw LFG
and the LFG/syngas mixtures that did not prove feasible in the present study. The testing should
also focus on optimizing waste heat utilization to maximize energy recovery. Preliminary estimates
are that the technology has the potential to improve energy efficiency by more than 10% from
waste heat recuperation. The break-point between the extent of reforming for increasing the energy
content of the fuel and turbocharger performance when operating on (reformate/biogas) mixtures
needs, however, to be studied. Given the added costs of carrying out field-testing with a larger
engine, in a future study, the focus should be on optimizing the heat recovery, turbocharger
performance, engine kW output, engine heat rate, and NO x emissions via process simulations first,
and then field-testing of the most promising process alternatives. It should be pointed out, that
just 1% reduction in fuel use in a large engine (via efficiency improvement) is significant. Having
the potential to decrease fuel use by as much as 10% would be a huge advancement for the “waste-
to-energy” field.
In terms of preparation of tubular SiC membranes, the future work needs to address the impact of
surface roughness and the pore size distribution of these supports, as from our own experience
they appear to be crucial factors in determining whether one is able to prepare crack-free, high-
quality SiC nonporous membranes. The characteristics of the slip-casting layer and its impact on
110
the subsequent deposition steps must be better understood as well. The impact of the quantity and
particle size distribution of the fine SiC powders utilized, as well as the effect of pore formers (like
PS) and other additives employed must be further investigated and better understood. In addition,
the interplay between the characteristics of the support and intermediate (slip-casting) layer need
to be studied in greater depth and detail. We know for example, from our own studies, so far, that
supports with a large average pore size and/or wide PSD require thick intermediate layers to make
them appropriate for further top-layer deposition and high-quality membrane preparation. This
then means that the higher permeance of such supports is more negatively impacted (from the
deposition of the intermediate layer) than the permeance of other supports with a smaller average
pore size and a narrower PSD (and a correspondingly lower starting permeance).
In terms of the deposition of the top nanoporous and highly permselective SiC layers, an area
where substantial improvements are needed is in the reproducibility in making these membranes.
Though we are able to make membranes with excellent permeation characteristics, the fraction of
such membranes is still quite small among a batch of membranes made and which show inferior
characteristics, hovering typically 20% - 30%. This is an unacceptably low success rate, and if the
SiC membranes proposed here are to eventually find commercial application, reproducibility in
their preparation must be significantly improved. A good starting point here would be a better
fundamental understanding of the reasons that cause so many of the membranes prepared under
identical conditions with others to fail. This should be the key focus of any future studies in this
area, in our opinion.
Membrane separation is a newly developed technology for hydrogen purification. Membrane
reactors that employ reaction and separation at the same time in one single unit, have attracted
considerable attention nowadays, especially for hydrogen production via the steam reforming of
111
methane. As stated earlier in the introduction section (see Sec. 1), The combination of the reaction
with selective permeation of hydrogen through the membrane increases the conversion of the
reaction at constant temperature significantly for the case of equilibrium-limited dehydrogenation
reaction. Lower operating temperature can be achieved while maintaining the conversion of the
dehydrogenation reaction in the palladium membrane reactor as the equilibrium shifts toward the
hydrogen-producing side. SiC membranes are potential candidates to be used in membrane
reactors as they show higher stability towards high temperatures as well as harsh environment such
as in the presence of H2S or steam in case of using landfill gas as feed for steam reforming of
methane in a membrane reactor. Therefore, further studies need to be done regarding the use of
membranes, particularly, silicon carbide membranes for important reactions such as the steam
reforming of methane, which was the subject of our study, as well as optimization of flow
conditions and membrane quality in terms of stability, gas permeance and selectivity in order to
achieve higher yields of H2 with reasonably high purities at the lowest possible energy cost.
112
Appendix
Cost Feasibility Evaluation
As part of this preliminary, “proof-of-concept” study, a cost feasibility evaluation for an 848
SCMH (500 SCFM) “biogas-to-energy” project was performed to verify whether it would be
financially viable. The choice of a small-size, 848 SCMH (500 SCFM) biogas-to-energy project
is based on the premise that the technology will be more appealing to owners/operators of landfills
at their declining stage of biogas production, during which period the biogas’ quality is also likely
to be more marginal. The results of this preliminary evaluation (shown in Table S1) are for building
a new, “grass-roots” system. Two technologies are compared in this Table. The conventional
technology, which requires the use of a SCR system to allow an engine to operate while satisfying
environmental regulations, and the proposed CR system which is envisioned capable to operate
without needing such a SCR system. It should be noted, however, that, the most likely scenario of
initial adaptations of the technology will be as retrofit systems, for landfills with existing biogas-
to-energy facilities, for which biogas quality has declined to the point where the economics of
operation of the existing systems are marginal. In such a circumstance, the choice will be between
abandoning the existing system (and flaring the biogas, instead) or retrofitting it with the CR
technology to prolong its operating life-time. Though flaring such gas is presently still permissible,
the development of technologies like the CR system proposed here that permit the beneficial use
of such a resource for electricity production, may “spur” new regulations that discourage the
continuation of present flaring practices.
113
Table A-1 Cost estimation of a 848 SCMH (500 SCFM) biogas-to-energy project
Process Parameters
Flow Rate 500 SCFM ~848 Nm3/hr
Catalyst Loading Gas Hourly Space Velocity (GHSV) 2000 hr-1 GHSV < 5,000 hr-1
Total Reactor Volume 2.12 m3 for R1=0.5
Loaded Density 900 kg/m3
Total Reactor Weight 1910.25 kg
Reformer Volume 0.198 m3 4" I.D. 8' heated length
Reformer Tubes 11 #
Cost Estimation
w/ Reformer w/o Reformer Basis
Pre-treatment $ 100,000 $ 100,000 Activated carbon treatment
Reformer tubes $ 225,000 $ - Engineer's estimate
Heat Exchangers, pumps, blowers $ 105,000 $ - Engineer's estimate
Gas Engine and interconnect $ 2,250,000 $ 2,250,000 LMOP cost-Web V2.1
Engineering/Construction $ 150,150 $ 107,250 Engineer's estimate
Capital Cost $ 2,830,150 $ 2,457,250 Sum of above
Post Combustion NOx Control (per yr)* $ - $ 130,759
O&M Cost (fixed+variable) $ 169,670 $ 121,193 LMOP cost-Web V2.1
Cap + 15yr NOx Control +15yr O&M $ 5,375,200 $ 6,236,530
Annually Cost $ 358,347 $ 415,769
Estimated Annual Revenue ($0.08/kWh)** $ 1,114,272 $ 1,051,200
*: Based on AQMD-Draft Tech Assessment ($10,900/ton)
**: Estimated green power price for small power production nominal 1,500 kw engine.
For an 848 SCMH (500 SCFM) project, the up-front capital cost of the system with reformer is
estimated to be around $2.83 million, which is higher (~$0.37 million) than the cost of the
conventional system without a reformer. However, the O&M cost (including NO X control) of the
system with reformer is lower when compared with its counterpart. The process with reformer
114
does not need to utilize post-combustion NOX control techniques (e.g., SCR) and eventually has
less annual cost ($358,347) compared with the one without reformer ($415,769) (the assumption
is made here that optimal design of the proposed CR system will ultimatelly eliminate the need to
use a SCR post-treatment system. The SCR cost data per ton of generated NO X for the related cost
estimations were obtained from the Air Pollution Control Technology Fact Sheet for SCR
technology of the US Environmental Protection Agency (EPA)). Even if this turns out to be too
optimistic of an assumption, and the CR system still requires a SCR system, it is likely that both
the CAPEX and O&M costs for such a system will be substantially lower than those for the
conventional system). The key advantage of the CR system is the gain in engine efficiency
resulting from the use of the hydrogen-rich fuel. As a result, for the 848 SCMH (500 SCFM)
project the CR process generates more revenue ($1,114,272) than its conventional counterpart
($1,051,200) without waste heat recuperation. The estimated net annual revenue (annual revenue
minus annual cost) for the CR process is $ 755,925 (with the payback period for the proposed
process being 2.7 years), while that for the conventional process is $ 635,431). It should be pointed
out that estimating the costs/revenues of a new, and not yet fully developed technology, and
comparing them against a mature technology is an “in-exact science”. It is expected, for example,
that once the new technology is fully developed, costs will be reduced. A key potential “draw” of
the proposed CR system for the intended application (closed or minimally-staffed landfill sites) is
that it is passive hence, except for minimal equipment maintenance (e.g., cleaning the heat
exchangers), there is no additional operation and maintenance costs.
115
List of Equipment
Generator…A Champion Power Equipment 7,500/9,375 W gasoline-powered generator, Figure
A-1a, was used to power-up the system and a smaller DuroMax XP4400EH 3500/4400 W, 7hp,
Dual Fuel gasoline/propane generator, Figure A-1b was used for experimental purposes, whereby
either a reformate gas or a blend of reformate and LFG that contained hydrogen would be used to
run the generator and monitor the NOX emissions in the exhaust for different loads on the engine.
a) b)
Figure A-1 a) The generator used to power-up the system, b) The generator used for the experiments.
Compressor/Back-Pressure Regulator…A compressor driven by a 1 hp motor was used to
pressurize the gas. The motor itself was driven by a variable frequency drive (VFD) to be able to
adjust the compressor speed. Using a back-pressure regulator, the pressure of the system was then
adjusted within the range of 1.15-1.49 bar (2-7 psig), and the surplus gas was recycled into the
inlet of the blower that fed the flare stack 1.
116
Figure A-2 The compressor driven by a 1 hp electrical motor.
Adsorbent Columns…The landfill gas contains small amounts (less than 1%) of NMOC,
including siloxanes, and trace amounts of various inorganic compounds, some of which have
strong, pungent odors such as hydrogen sulfide and are potentially poisonous for the reforming
catalysts used. Four 10.16 cm (4 in) Dia. PVC tanks filled with activated carbon and silica gel
were, therefore, used as adsorbent columns to remove those contaminants. Two of these PVC tanks
were installed upstream of the compressor to remove most of the contaminates and the moisture
contained in the LFG that could potentially damage the compressor; the other two PVC tanks were
installed downstream of the compressor in order to remove the remaining contaminants and the
extra moisture condensed after the pressurized flow coming out of the compressor discharge was
cooled down. The position of the four installed four adsorbent columns can be seen in Figure 2-2.
Boiler…A Boiler 20.32 cm (8 in) in diameter was used to generate steam at the desired pressure
up to the maximum designed pressure of 2.04 bar (15 psig). A 121 liter (32 US gal.) tank was filled
with de-ionized water to be used in the boiler during the experiments. A level switch was used to
keep the water level in the boiler constant. The pressure of the boiler was monitored through the
use of a pressure transducer that provides a (4 – 20 mA) output signal to the control system that
will be described later in this section. A portion of the heat required to generate steam in the boiler
117
was provided by two 1500 W electrical heaters, one of which was running and controlled by a
relay until the boiler reached a given pressure of the steam, and the other was controlled by the
programmable logic controller (PLC) – see below-- through a silicon-controlled rectifier (SiCR)
based on the defined pressure set-point. The other portion of the heat came from the circulation of
the water in the boiler through a heat exchanger that used the exhaust heat of the power generator
to heat the water, as well as by using the heat of the reformer outlet passing through a stainless
steel coil inside the boiler.
Reformer…A, 60.96 cm (24 in) long stainless steel pipe (7.62 cm (3 in) ID, 8.89 cm (3.5 in) OD,
Schedule 40) was used to build the packed-bed reactor to be used to reform the LFG into syngas
through the steam reforming and water-gas shift reactions. The bottom half of the reformer was
packed with two different sizes of stainless steel balls and served as the pre-heater section. The top
of the reformer was packed with Pt/Al2O3 catalyst pellets (JM-R44 PGM) and served as the
reaction zone. The properties of the catalyst by Johnson Matthey Inc. are listed in Table 1. Two
K-type thermocouples were inserted in a thermo-well made from a 6.35 mm (1/4 in) stainless steel
tubing and inserted axially in the reactor, their tips being within 15.24 cm (6 in) and 30.48 cm (12
in) from the bottom of the reformer. One additional thermocouple was inserted from the top of the
reformer, its tip placed exactly at the exit of the reformer. The reformer was heated from the outside
using four high-temperature ceramic heaters. The two bottom heaters were installed in parallel and
controlled by the system control through one SiCR; the two top heaters, installed in parallel again,
were controlled by the system control through another separate SiCR. The temperatures of the
reformer were monitored and recorded in real time by the system control panel. A flange was used
for sealing the bottom-end of the reformer. The flange plate was welded to a tee union used for
both the gas inlet connection and the thermocouple insertion point. The top-end was welded to a
118
pipe tee and used for both the reformer outlet using a stainless steel compression fitting and as a
thermocouple insertion point. The outer surface of the reformer was completely insulated using
high-temperature insulation. A schematic view of the reformer can be seen in Figure A-3.
Figure A-3 Schematic view of the 60.96 cm (24”) long stainless steel reformer.
Heat Exchanger…A cross-flow-type heat exchanger was designed and fabricated to preheat the
gas going into the reformer after being mixed with the steam, as well as to heat the circulating
water between the boiler and the heat exchanger using the exhaust gas of the power generator the
temperature of which varied from 300 to 340℃, see Figure A-4a for a schematic. An air-fin cooler
with a fan that would blow air into the fins was used to cool-down the reformate gas after passing
through the coil inside the boiler. The air-fin cooler mainly removed the latent heat and condensed
the water vapor in the reformate gas, which was then drained using a manual valve on the heat
exchanger and a water knock-out system, see Figure A-4b.
a)
Exhaust Gas
119
b)
Figure A-4 a) Cross-flow heat exchanger b) Air-fin cooler with a fan.
Sensors and Analyzers…A single-passage sensor block was designed and fabricated to hold five
sensors, two of which were 0-100% methane and carbon dioxide infrared-type sensors that
provided 0.2-4VDC output signal to the voltage input of the PLC, and the remaining three were 0-
25% oxygen sensor, 0-5000 ppmv nitrogen oxide sensor, and 0-200 ppm v nitrogen dioxide sensor;
each sensor came with an analog transmission board that converted the voltage output of the sensor
to the 4-20 mA output signal for the analog input of the system control unit. A portable hydrogen
analyzer was used to measure the concentration of hydrogen by directly connecting it to the system
outlet at the system pressure of 1.35-1.7 bar (5-10 psig) at an adjusted flow rate of 141.5 cc/min
(0.3 SCFH) to the analyzer. The analyzer was designed to measure the concentration of H 2 in CO2.
The real composition of hydrogen in the gas mixture which was mainly composed of H 2, CO, CO2,
CH4, N2, and O2 was then estimated using the measured data out the analyzer. Before going into
the hydrogen analyzer, the remaining moisture in the gas mixture was removed by passing it
through a water knock-out system and a column of calcium chloride. A separate thermoelectric
Peltier refrigeration cooling system, Figure A-5, was designed for the sensor block to cool the
sample gas, separate the remaining water vapor in a column of calcium chloride, and reheat it
before it flowed through the sensor passage. The calcium chloride would be replaced when it
eventually dissolved in its own crystal lattice water.
120
Figure A-5 Thermoelectric Peltier refrigeration cooling system.
Programmable Logic Controller…A Direct Logic 205 Koyo Programmable
Logic Controller (PLC), see Figure A-6, was used to receive signals and control different elements
in the system through sending an output signal. The input to the PLC was either a voltage input
such as the thermocouple inputs or the sensors voltage signals (e.g., methane and carbon dioxide
sensors), or analog input (4-20 mA) signals (e.g., pressure transducers, differential pressure
transducers), and sensors with analog transmission board (e.g., nitrogen oxide, nitrogen dioxide,
and oxygen sensors). The output of the PLC was either a relay output, which is mainly an on/off
switch for any device in the system controlled by the PLC (e.g., the high-temperature pump for
the water circulation, or one of the electrical heaters in the boiler) or an analog output signal (e.g.,
the input signals to the SiCR that would power-up the ceramic heaters on the reformer and one of
the electrical heaters in the boiler). A Human Machine Interface (HMI) was used to monitor and
control the process and the different elements of the system. Figure A-7 shows different HMI
window screens through which various elements were either monitored, such as the temperatures
and pressures, or controlled, such as the electrical heaters and pumps. Different parameters such
as the PID loop parameters for the boiler electrical heater and the reformer ceramic heater were
Upper Block
Lower Block
Cold Gas outlet
Gas
Fan
Dry Gas
Gas Outlet
CaCl 2 Water
Absorber Column
Thermoelectric Peltier
121
also set using the HMI. In addition to the PLC unit the system control panel housed the
5/12/24VDC power supplies, temperature controllers (TC), and all sorts of connection wires,
terminals, and circuit breakers, see Figure A-8. The whole unit is powered by an 800 W
Uninterruptible Power System (UPS) to avoid any kind of power failure and disturbance. The
P&ID of the overall system is shown in Figure A-9.
122
Figure A-6 The PLC unit with the total number of 8 cards connected to the HMI through Ethernet card/cable
and a Network Switch.
Figure A-7 The HMI window panels.
123
Figure A-8 The system control panel.
124
Figure A-9 P&ID drawing of the system.
125
References:
[1] M. S. Fan, A. Z. Abdullah, and S. Bhatia, "Catalytic technology for carbon dioxide
reforming of methane to synthesis gas," ChemCatChem, vol. 1, pp. 192-208, 2009.
[2] U. DoE, "Waste Heat Reduction and Recovery for Improving Furnace Efficiency
Productivity and Emissions Performance," US Department of Energy: Washington, DC,
USA, 2004.
[3] R. Broeker, "Combined-cycle cogeneration to power oil refinery," Mech. Eng.;(United
States), vol. 108, 1986.
[4] A. Schurz, L. Andersson, and H. Storm, "12 MW gas turbine cogeneration plant fired
with refinery gases," Chartered Mechanical Engineer, vol. 29, pp. 63-6, 1982.
[5] Y. S. Najjar and M. B. Habeebullah, "Energy conservation in the refinery by utilizing
reformed fuel gas and furnace flue gases," Heat Recovery Systems and CHP, vol. 11, pp.
517-521, 1991.
[6] A. O. Abdulrahman, D. Huisingh, and W. Hafkamp, "Sustainability improvements in
Egypt's oil & gas industry by implementation of flare gas recovery," Journal of Cleaner
Production, vol. 98, pp. 116-122, 2015.
[7] M. Rahimpour, Z. Jamshidnejad, S. Jokar, G. Karimi, A. Ghorbani, and A. Mohammadi,
"A comparative study of three different methods for flare gas recovery of Asalooye Gas
Refinery," Journal of Natural Gas Science and Engineering, vol. 4, pp. 17-28, 2012.
[8] M. R. Rahimpour and S. M. Jokar, "Feasibility of flare gas reformation to practical
energy in Farashband gas refinery: no gas flaring," Journal of hazardous materials, vol.
209, pp. 204-217, 2012.
[9] M. Saidi, F. Siavashi, and M. Rahimpour, "Application of solid oxide fuel cell for flare
gas recovery as a new approach; a case study for Asalouyeh gas processing plant, Iran,"
Journal of Natural Gas Science and Engineering, vol. 17, pp. 13-25, 2014.
[10] R. N. Hassanlouei, R. Pelalak, and A. Daraei, "Wettability study in CO 2 capture from flue
gas using nano porous membrane contactors," International Journal of Greenhouse Gas
Control, vol. 16, pp. 233-240, 2013.
[11] X. He, C. Fu, and M.-B. Hägg, "Membrane system design and process feasibility analysis
for CO 2 capture from flue gas with a fixed-site-carrier membrane," Chemical
Engineering Journal, vol. 268, pp. 1-9, 2015.
[12] R. Khalilpour, K. Mumford, H. Zhai, A. Abbas, G. Stevens, and E. S. Rubin,
"Membrane-based carbon capture from flue gas: a review," Journal of Cleaner
Production, vol. 103, pp. 286-300, 2015.
[13] S. R. Reijerkerk, R. Jordana, K. Nijmeijer, and M. Wessling, "Highly hydrophilic,
rubbery membranes for CO2 capture and dehydration of flue gas," International Journal
of Greenhouse Gas Control, vol. 5, pp. 26-36, 2011.
[14] W.-H. Chen, M.-R. Lin, T.-S. Leu, and S.-W. Du, "An evaluation of hydrogen production
from the perspective of using blast furnace gas and coke oven gas as feedstocks,"
international journal of hydrogen energy, vol. 36, pp. 11727-11737, 2011.
[15] M. Halmann and A. Steinfeld, "Hydrogen production and CO 2 fixation by flue-gas
treatment using methane tri-reforming or coke/coal gasification combined with lime
carbonation," international journal of hydrogen energy, vol. 34, pp. 8061-8066, 2009.
[16] R. Liberatore, M. Lanchi, G. Caputo, C. Felici, A. Giaconia, S. Sau, et al., "Hydrogen
production by flue gas through sulfur–iodine thermochemical process: Economic and
126
energy evaluation," international journal of hydrogen energy, vol. 37, pp. 8939-8953,
2012.
[17] J. R. Rostrup-Nielsen and T. Rostrup-Nielsen, "Large-scale hydrogen production,"
Cattech, vol. 6, pp. 150-159, 2002.
[18] S. M. Jokar, M. R. Rahimpour, A. Shariati, A. Iulianelli, G. Bagnato, A. Vita, et al.,
"Pure Hydrogen Production in Membrane Reactor with Mixed Reforming Reaction by
Utilizing Waste Gas: A Case Study," Processes, vol. 4, p. 33, 2016.
[19] J.-H. Wee, "Applications of proton exchange membrane fuel cell systems," Renewable
and sustainable energy reviews, vol. 11, pp. 1720-1738, 2007.
[20] S. Ahmed and M. Krumpelt, "Hydrogen from hydrocarbon fuels for fuel cells,"
International journal of hydrogen energy, vol. 26, pp. 291-301, 2001.
[21] G. Rabenstein and V. Hacker, "Hydrogen for fuel cells from ethanol by steam-reforming,
partial-oxidation and combined auto-thermal reforming: a thermodynamic analysis,"
Journal of Power Sources, vol. 185, pp. 1293-1304, 2008.
[22] Y.-S. Seo, A. Shirley, and S. Kolaczkowski, "Evaluation of thermodynamically
favourable operating conditions for production of hydrogen in three different reforming
technologies," Journal of Power sources, vol. 108, pp. 213-225, 2002.
[23] F. Gallucci, A. Comite, G. Capannelli, and A. Basile, "Steam reforming of methane in a
membrane reactor: an industrial case study," Industrial & engineering chemistry research,
vol. 45, pp. 2994-3000, 2006.
[24] C. M. Kalamaras and A. M. Efstathiou, "Hydrogen production technologies: current state
and future developments," in Conference Papers in Science, 2013.
[25] N. Muradov and T. Veziroǧlu, "From hydrocarbon to hydrogen–carbon to hydrogen
economy," International Journal of Hydrogen Energy, vol. 30, pp. 225-237, 2005.
[26] M. Balat and M. Balat, "Political, economic and environmental impacts of biomass-based
hydrogen," International journal of hydrogen energy, vol. 34, pp. 3589-3603, 2009.
[27] A. Iulianelli, S. Liguori, J. Wilcox, and A. Basile, "Advances on methane steam
reforming to produce hydrogen through membrane reactors technology: A review,"
Catalysis Reviews, vol. 58, pp. 1-35, 2016.
[28] J. R. Rostrup-Nielsen, "Production of synthesis gas," Catalysis today, vol. 18, pp. 305-
324, 1993.
[29] A. Brunetti, E. Drioli, and G. Barbieri, "Medium/high temperature water gas shift
reaction in a Pd–Ag membrane reactor: an experimental investigation," RSC Advances,
vol. 2, pp. 226-233, 2012.
[30] C. Rhodes, B. P. Williams, F. King, and G. J. Hutchings, "Promotion of Fe 3O4/Cr2O3
high temperature water gas shift catalyst," catalysis communications, vol. 3, pp. 381-384,
2002.
[31] L. Cornaglia, J. Múnera, and E. Lombardo, "Recent advances in catalysts, palladium
alloys and high temperature WGS membrane reactors: a review," International Journal of
Hydrogen Energy, vol. 40, pp. 3423-3437, 2015.
[32] D. Mendes, A. Mendes, L. Madeira, A. Iulianelli, J. Sousa, and A. Basile, "The water‐gas
shift reaction: from conventional catalytic systems to Pd‐based membrane reactors—a
review," Asia‐Pacific Journal of Chemical Engineering, vol. 5, pp. 111-137, 2010.
[33] T. de Freitas Silva, C. G. M. Reis, A. F. Lucrédio, E. M. Assaf, and J. M. Assaf,
"Hydrogen production from oxidative reforming of methane on Ni/γ-Al 2 O3 catalysts:
127
Effect of support promotion with La, La–Ce and La–Zr," Fuel Processing Technology,
vol. 127, pp. 97-104, 2014.
[34] M. A. Nieva, M. M. Villaverde, A. Monzón, T. F. Garetto, and A. J. Marchi, "Steam-
methane reforming at low temperature on nickel-based catalysts," Chemical Engineering
Journal, vol. 235, pp. 158-166, 2014.
[35] S. Palma, L. Bobadilla, A. Corrales, S. Ivanova, F. Romero-Sarria, M. Centeno, et al.,
"Effect of gold on a NiLaO 3 perovskite catalyst for methane steam reforming," Applied
Catalysis B: Environmental, vol. 144, pp. 846-854, 2014.
[36] V. Palma, M. Miccio, A. Ricca, E. Meloni, and P. Ciambelli, "Monolithic catalysts for
methane steam reforming intensification: Experimental and numerical investigations,"
Fuel, vol. 138, pp. 80-90, 2014.
[37] J. Zhang, T. Zhang, X. Zhang, W. Liu, H. Liu, J. Qiu, et al., "New synthesis strategies for
Ni/Al 2 O 3-Sil-1 core–shell catalysts for steam reforming of methane," Catalysis Today,
vol. 236, pp. 34-40, 2014.
[38] S. D. Angeli, G. Monteleone, A. Giaconia, and A. A. Lemonidou, "State-of-the-art
catalysts for CH 4 steam reforming at low temperature," International journal of
hydrogen energy, vol. 39, pp. 1979-1997, 2014.
[39] D. Homsi, S. Aouad, C. Gennequin, A. Aboukaïs, and E. Abi-Aad, "A highly reactive
and stable Ru/Co6− x Mgx Al2 catalyst for hydrogen production via methane steam
reforming," international journal of hydrogen energy, vol. 39, pp. 10101-10107, 2014.
[40] D. Mei, V.-A. Glezakou, V. Lebarbier, L. Kovarik, H. Wan, K. O. Albrecht, et al.,
"Highly active and stable MgAl2 O4-supported Rh and Ir catalysts for methane steam
reforming: A combined experimental and theoretical study," Journal of Catalysis, vol.
316, pp. 11-23, 2014.
[41] P. S. Roy, N.-K. Park, and K. Kim, "Metal foam-supported Pd–Rh catalyst for steam
methane reforming and its application to SOFC fuel processing," International journal of
hydrogen energy, vol. 39, pp. 4299-4310, 2014.
[42] M. M. Halmann, Chemical Fixation of Carbon DioxideMethods for Recycling CO 2 into
Useful Products: CRC press, 1993.
[43] Y. Cui, H. Zhang, H. Xu, and W. Li, "The CO 2 reforming of CH4 over Ni/La2O3/α-Al2 O3
catalysts: the effect of La 2O3 contents on the kinetic performance," Applied Catalysis A:
General, vol. 331, pp. 60-69, 2007.
[44] Y. Kathiraser, U. Oemar, E. T. Saw, Z. Li, and S. Kawi, "Kinetic and mechanistic aspects
for CO2 reforming of methane over Ni based catalysts," Chemical Engineering Journal,
vol. 278, pp. 62-78, 2015.
[45] J. Rostrupnielsen and J. B. Hansen, "CO2-reforming of methane over transition metals,"
Journal of Catalysis, vol. 144, pp. 38-49, 1993.
[46] E. Ruckenstein and Y. H. Hu, "The effect of precursor and preparation conditions of
MgO on the CO2 reforming of CH4 over NiO/MgO catalysts," Applied Catalysis A:
General, vol. 154, pp. 185-205, 1997.
[47] Z. Zhang and X. E. Verykios, "A stable and active nickel-based catalyst for carbon
dioxide reforming of methane to synthesis gas," Journal of the Chemical Society,
Chemical Communications, pp. 71-72, 1995.
[48] Z. Zhang and X. E. Verykios, "Carbon dioxide reforming of methane to synthesis gas
over Ni/La2O3 catalysts," Applied Catalysis A: General, vol. 138, pp. 109-133, 1996.
128
[49] M. Abashar, "Coupling of steam and dry reforming of methane in catalytic fluidized bed
membrane reactors," International Journal of Hydrogen Energy, vol. 29, pp. 799-808,
2004.
[50] G. Froment, "Production of synthesis gas by steam-and CO 2-reforming of natural gas,"
Journal of Molecular Catalysis A: Chemical, vol. 163, pp. 147-156, 2000.
[51] C.-l. Li, Y.-l. Fu, G.-z. Bian, Y.-n. Xie, T.-d. Hu, and J. Zhang, "Effect of Steam in CO 2
Reforming of CH4over a Ni/CeO2–ZrO2–Al2O3 Catalyst," Kinetics and catalysis, vol. 45,
pp. 679-683, 2004.
[52] J.-W. Snoeck, G. Froment, and M. Fowles, "Kinetic evaluation of carbon formation in
steam/CO2-natural gas reformers. Influence of the catalyst activity and alkalinity,"
International Journal of Chemical Reactor Engineering, vol. 1, 2003.
[53] V. R. Choudhary and A. M. Rajput, "Simultaneous carbon dioxide and steam reforming
of methane to syngas over NiO-CaO catalyst," Industrial & engineering chemistry
research, vol. 35, pp. 3934-3939, 1996.
[54] M. Hegarty, A. O'connor, and J. Ross, "Syngas production from natural gas using ZrO 2-
supported metals," Catalysis Today, vol. 42, pp. 225-232, 1998.
[55] Ş. Özkara-Aydınoğlu, "Thermodynamic equilibrium analysis of combined carbon dioxide
reforming with steam reforming of methane to synthesis gas," international journal of
hydrogen energy, vol. 35, pp. 12821-12828, 2010.
[56] A. Iulianelli, S. Liguori, Y. Huang, and A. Basile, "Model biogas steam reforming in a
thin Pd-supported membrane reactor to generate clean hydrogen for fuel cells," Journal of
Power Sources, vol. 273, pp. 25-32, 2015.
[57] K. Y. Koo, S.-h. Lee, U. H. Jung, H.-S. Roh, and W. L. Yoon, "Syngas production via
combined steam and carbon dioxide reforming of methane over Ni–Ce/MgAl 2O4
catalysts with enhanced coke resistance," Fuel Processing Technology, vol. 119, pp. 151-
157, 2014.
[58] A. A. Lemonidou and I. A. Vasalos, "Carbon dioxide reforming of methane over 5 wt.%
Ni/CaO-Al2 O3 catalyst," Applied Catalysis A: General, vol. 228, pp. 227-235, 2002.
[59] T. Boger and A. K. Heibel, "Heat transfer in conductive monolith structures," Chemical
engineering science, vol. 60, pp. 1823-1835, 2005.
[60] V. R. Choudhary and K. C. Mondal, "CO 2 reforming of methane combined with steam
reforming or partial oxidation of methane to syngas over NdCoO 3 perovskite-type mixed
metal-oxide catalyst," Applied Energy, vol. 83, pp. 1024-1032, 2006.
[61] F. B. Noronha, A. Shamsi, C. Taylor, E. C. Fendley, S. Stagg-Williams, and D. E.
Resasco, "Catalytic performance of Pt/ZrO2 and Pt/Ce-ZrO2 catalysts on CO 2 reforming
of CH4 coupled with steam reforming or under high pressure," Catalysis letters, vol. 90,
pp. 13-21, 2003.
[62] M. Sheng, H. Yang, D. R. Cahela, and B. J. Tatarchuk, "Novel catalyst structures with
enhanced heat transfer characteristics," Journal of Catalysis, vol. 281, pp. 254-262, 2011.
[63] Y. Hu and E. Ruckenstein, "An optimum NiO content in the CO 2 reforming of CH4 with
NiO/MgO solid solution catalysts," Catalysis letters, vol. 36, pp. 145-149, 1996.
[64] E. Ruckenstein and H. Wang, "Carbon deposition and catalytic deactivation during CO 2
reforming of CH4 over Co/γ-Al2O3 catalysts," Journal of Catalysis, vol. 205, pp. 289-293,
2002.
129
[65] Å. Slagtern, U. Olsbye, R. Blom, I. M. Dahl, and H. Fjellvåg, "In situ XRD
characterization of La Ni Al O model catalysts for CO 2 reforming of methane," Applied
Catalysis A: General, vol. 145, pp. 375-388, 1996.
[66] G. Valderrama, M. R. Goldwasser, C. U. de Navarro, J. M. Tatibouët, J. Barrault, C.
Batiot-Dupeyrat, et al., "Dry reforming of methane over Ni perovskite type oxides,"
Catalysis Today, vol. 107, pp. 785-791, 2005.
[67] J. W. Bae, A. R. Kim, S.-C. Baek, and K.-W. Jun, "The role of CeO 2–ZrO2 distribution
on the Ni/MgAl2O4 catalyst during the combined steam and CO 2 reforming of methane,"
Reaction Kinetics, Mechanisms and Catalysis, vol. 104, pp. 377-388, 2011.
[68] M. Danilova, Z. Fedorova, V. Zaikovskii, A. Porsin, V. Kirillov, and T. Krieger, "Porous
nickel-based catalysts for combined steam and carbon dioxide reforming of methane,"
Applied Catalysis B: Environmental, vol. 147, pp. 858-863, 2014.
[69] L. Li, R. W. Borry, and E. Iglesia, "Design and optimization of catalysts and membrane
reactors for the non-oxidative conversion of methane," Chemical engineering science,
vol. 57, pp. 4595-4604, 2002.
[70] Y. Chen, Y. Wang, H. Xu, and G. Xiong, "Efficient production of hydrogen from natural
gas steam reforming in palladium membrane reactor," Applied Catalysis B:
Environmental, vol. 81, pp. 283-294, 2008.
[71] V. Choudhary, B. Uphade, and A. Mamman, "Simultaneous steam and CO 2 reforming of
methane to syngas over NiO/MgO/SA-5205 in presence and absence of oxygen," Applied
Catalysis A: General, vol. 168, pp. 33-46, 1998.
[72] Z.-W. Liu, H.-S. Roh, and K.-W. Jun, "Important factors on carbon dioxide reforming of
methane over nickel-based catalysts," Journal of Industrial and Engineering Chemistry,
vol. 9, pp. 753-761, 2003.
[73] J. Rostrup-Nielsen, L. Christiansen, and J.-H. B. Hansen, "Activity of steam reforming
catalysts: role and assessment," Applied Catalysis, vol. 43, pp. 287-303, 1988.
[74] J. R. Rostrup-Nielsen, J. Sehested, and J. K. Nørskov, "Hydrogen and synthesis gas by
steam-and C0 2 reforming," Advances in catalysis, vol. 47, pp. 65-139, 2002.
[75] I. H. Son, S. J. Lee, A. Soon, H.-S. Roh, and H. Lee, "Steam treatment on Ni/γ-Al 2O3 for
enhanced carbon resistance in combined steam and carbon dioxide reforming of
methane," Applied Catalysis B: Environmental, vol. 134, pp. 103-109, 2013.
[76] J. Wei and E. Iglesia, "Isotopic and kinetic assessment of the mechanism of reactions of
CH4 with CO2 or H2O to form synthesis gas and carbon on nickel catalysts," Journal of
Catalysis, vol. 224, pp. 370-383, 2004.
[77] K. Liu, C. Song, and V. Subramani, Hydrogen and syngas production and purification
technologies: Wiley Online Library, 2010.
[78] M. M. Barroso-Quiroga and A. E. Castro-Luna, "Catalytic activity and effect of modifiers
on Ni-based catalysts for the dry reforming of methane," International Journal of
Hydrogen Energy, vol. 35, pp. 6052-6056, 2010.
[79] M. Bradford and M. Vannice, "CO 2 reforming of CH4," Catalysis Reviews, vol. 41, pp. 1-
42, 1999.
[80] C. E. Gigola, M. S. Moreno, I. Costilla, and M. D. Sánchez, "Characterization of Pd–CeO
x interaction on α-Al2O3 support," Applied surface science, vol. 254, pp. 325-329, 2007.
[81] K. Hashimoto, S. Watase, and N. Toukai, "Reforming of methane with carbon dioxide
over a catalyst consisting of ruthenium metal and cerium oxide supported on mordenite,"
Catalysis letters, vol. 80, pp. 147-152, 2002.
130
[82] J. F. Múnera, S. Irusta, L. M. Cornaglia, E. A. Lombardo, D. V. Cesar, and M. Schmal,
"Kinetics and reaction pathway of the CO2 reforming of methane on Rh supported on
lanthanum-based solid," Journal of Catalysis, vol. 245, pp. 25-34, 2007.
[83] K. Nagaoka, K. Seshan, K.-i. Aika, and J. A. Lercher, "Carbon deposition during carbon
dioxide reforming of methane—comparison between Pt/Al 2O3 and Pt/ZrO2," Journal of
Catalysis, vol. 197, pp. 34-42, 2001.
[84] F. Pompeo, N. N. Nichio, M. M. Souza, D. V. Cesar, O. A. Ferretti, and M. Schmal,
"Study of Ni and Pt catalysts supported on α-Al2O3 and ZrO2 applied in methane
reforming with CO2," Applied Catalysis A: General, vol. 316, pp. 175-183, 2007.
[85] S. Adhikari and S. Fernando, "Hydrogen membrane separation techniques," Industrial &
Engineering Chemistry Research, vol. 45, pp. 875-881, 2006.
[86] T. Tsuru, K. Yamaguchi, T. Yoshioka, and M. Asaeda, "Methane steam reforming by
microporous catalytic membrane reactors," AIChE journal, vol. 50, pp. 2794-2805, 2004.
[87] V. M. Gryaznov, M. Ermilova, N. Orekhova, and G. Tereschenko, "Reactors with metal
and metal-containing membranes," in Structured Catalysts and Reactors, ed:
Taylor&Francis London, NY, 2005, p. 579p.
[88] H. Hsieh, Inorganic membranes for separation and reaction vol. 3: Elsevier, 1996.
[89] S. W. Nam, S. P. Yoon, H. Y. Ha, S.-A. Hong, and A. P. Maganyuk, "Methane steam
reforming in a Pd-Ru membrane reactor," Korean Journal of Chemical Engineering, vol.
17, pp. 288-291, 2000.
[90] J. G. S. Marcano and T. T. Tsotsis, Catalytic membranes and membrane reactors: Wiley-
VCH Verlag GmbH, 2002.
[91] S. Uemiya, N. Sato, H. Ando, T. Matsuda, and E. Kikuchi, "Steam reforming of methane
in a hydrogen-permeable membrane reactor," Applied catalysis, vol. 67, pp. 223-230,
1990.
[92] A. Shigarov, V. Meshcheryakov, and V. Kirillov, "Use of Pd membranes in catalytic
reactors for steam methane reforming for pure hydrogen production," Theoretical
Foundations of Chemical Engineering, vol. 45, pp. 595-609, 2011.
[93] A. Shigarov and V. Kirillov, "Modeling of membrane reactor for steam methane
reforming: from granular to structured catalysts," Theoretical Foundations of Chemical
Engineering, vol. 46, pp. 97-107, 2012.
[94] W. Allen, P. Irving, and W. Thomson, "Microreactor systems for hydrogen generation
and oxidative coupling of methane," in IMRET, 2000, pp. 5-9.
[95] F. García-García, M. Rahman, I. González-Jiménez, and K. Li, "Catalytic hollow fibre
membrane micro-reactor: high purity H2 production by WGS reaction," Catalysis today,
vol. 171, pp. 281-289, 2011.
[96] F. García-García, M. Rahman, B. Kingsbury, and K. Li, "A novel catalytic membrane
microreactor for COx free H2 production," Catalysis Communications, vol. 12, pp. 161-
164, 2010.
[97] A. Gil, Z. Wu, D. Chadwick, and K. Li, "Novel Catalytic Membrane Micro-reactors for
CO 2 Capture Via pre-combustion Decarbonization Route," Procedia Engineering, vol.
44, pp. 1315-1316, 2012.
[98] C. Juergen, "Catalysis in micro-structured membrane reactors with nano-designed
membranes," Chinese Journal of Catalysis, vol. 29, pp. 1169-1177, 2008.
131
[99] S. V. Karnik, M. K. Hatalis, and M. V. Kothare, "Towards a palladium micro-membrane
for the water gas shift reaction: Microfabrication approach and hydrogen purification
results," Journal of Microelectromechanical Systems, vol. 12, pp. 93-100, 2003.
[100] A. Mejdell, M. Jøndahl, T. Peters, R. Bredesen, and H. Venvik, "Experimental
investigation of a microchannel membrane configuration with a 1.4 μm Pd/Ag23wt.%
membrane—Effects of flow and pressure," Journal of Membrane Science, vol. 327, pp. 6-
10, 2009.
[101] M. Rahman, F. García-García, M. I. Hatim, B. Kingsbury, and K. Li, "Development of a
catalytic hollow fibre membrane micro-reactor for high purity H 2 production," Journal
of Membrane Science, vol. 368, pp. 116-123, 2011.
[102] P. Seelam, M. Huuhtanen, R. Keiski, and A. Basile, "Microreactors and membrane
microreactors: fabrication and applications," Handbook of Membrane reactors.
Woodhead Publishing Limited, pp. 188-235, 2013.
[103] A. P. Singh, S. Singh, S. Ganguly, and A. V. Patwardhan, "Steam reforming of methane
and methanol in simulated macro & micro-scale membrane reactors: Selective separation
of hydrogen for optimum conversion," Journal of Natural Gas Science and Engineering,
vol. 18, pp. 286-295, 2014.
[104] M. Chai, M. Machida, K. Eguchi, and H. Arai, "Promotion of hydrogen permeation on
metal-dispersed alumina membranes and its application to a membrane reactor for
methane steam reforming," Applied Catalysis A: General, vol. 110, pp. 239-250, 1994.
[105] T. Tsotsis, A. Champagnie, S. Vasileiadis, Z. Ziaka, and R. Minet, "The enhancement of
reaction yield through the use of high temperature membrane reactors," Separation
science and technology, vol. 28, pp. 397-422, 1993.
[106] T. Tsuru, T. Tsuge, S. Kubota, K. Yoshida, T. Yoshioka, and M. Asaeda, "Catalytic
membrane reaction for methane steam reforming using porous silica membranes,"
Separation science and technology, vol. 36, pp. 3721-3736, 2001.
[107] K. Yoshida, Y. Hirano, H. Fujii, T. Tsuru, and M. Asaeda, "An application of silica-
zirconia membrane for hydrogen separation to membrane reactor," Kagaku Kogaku
Ronbunshu, vol. 27, pp. 657-660, 2001.
[108] A. Prabhu, A. Liu, L. Lovell, and S. Oyama, "Modeling of the methane reforming
reaction in hydrogen selective membrane reactors," Journal of membrane science, vol.
177, pp. 83-95, 2000.
[109] A. K. Prabhu and S. T. Oyama, "Highly hydrogen selective ceramic membranes:
application to the transformation of greenhouse gases," Journal of membrane science,
vol. 176, pp. 233-248, 2000.
[110] S. Araki, N. Hino, T. Mori, and S. Hikazudani, "Autothermal reforming of biogas over a
monolithic catalyst," Journal of Natural Gas Chemistry, vol. 19, pp. 477-481, 2010.
[111] L. B. Braga, J. L. Silveira, M. E. Da Silva, C. E. Tuna, E. B. Machin, and D. T. Pedroso,
"Hydrogen production by biogas steam reforming: A technical, economic and ecological
analysis," Renewable and Sustainable Energy Reviews, vol. 28, pp. 166-173, 2013.
[112] A. Brunetti, Y. Sun, A. Caravella, E. Drioli, and G. Barbieri, "Process Intensification for
greenhouse gas separation from biogas: More efficient process schemes based on
membrane-integrated systems," International Journal of Greenhouse Gas Control, vol. 35,
pp. 18-29, 2015.
132
[113] Y. N. Chun, Y. C. Yang, and K. Yoshikawa, "Hydrogen generation from biogas
reforming using a gliding arc plasma-catalyst reformer," Catalysis today, vol. 148, pp.
283-289, 2009.
[114] A. L. da Silva, L. F. P. Dick, and I. L. Müller, "Performance of a PEMFC system
integrated with a biogas chemical looping reforming processor: a theoretical analysis and
comparison with other fuel processors (steam reforming, partial oxidation and auto-
thermal reforming)," international journal of hydrogen energy, vol. 37, pp. 6580-6600,
2012.
[115] A. Effendi, K. Hellgardt, Z.-G. Zhang, and T. Yoshida, "Optimising H 2 production from
model biogas via combined steam reforming and CO shift reactions," Fuel, vol. 84, pp.
869-874, 2005.
[116] A. Galvagno, V. Chiodo, F. Urbani, and F. Freni, "Biogas as hydrogen source for fuel
cell applications," International Journal of hydrogen energy, vol. 38, pp. 3913-3920,
2013.
[117] U. Izquierdo, V. Barrio, N. Lago, J. Requies, J. Cambra, M. Güemez, et al., "Biogas
steam and oxidative reforming processes for synthesis gas and hydrogen production in
conventional and microreactor reaction systems," international journal of hydrogen
energy, vol. 37, pp. 13829-13842, 2012.
[118] C. Lau, A. Tsolakis, and M. Wyszynski, "Biogas upgrade to syn-gas (H 2–CO) via dry and
oxidative reforming," International Journal of Hydrogen Energy, vol. 36, pp. 397-404,
2011.
[119] K.-H. Lin, H.-F. Chang, and A. C.-C. Chang, "Biogas reforming for hydrogen production
over mesoporous Ni 2x Ce 1− x O2 catalysts," International Journal of Hydrogen Energy,
vol. 37, pp. 15696-15703, 2012.
[120] A. F. Lucredio, J. M. Assaf, and E. M. Assaf, "Reforming of a model biogas on Ni and
Rh–Ni catalysts: Effect of adding La," Fuel Processing Technology, vol. 102, pp. 124-
131, 2012.
[121] N. Muradov and F. Smith, "Thermocatalytic conversion of landfill gas and biogas to
alternative transportation fuels," Energy & Fuels, vol. 22, pp. 2053-2060, 2008.
[122] T. Ohkubo, Y. Hideshima, and Y. Shudo, "Estimation of hydrogen output from a full-
scale plant for production of hydrogen from biogas," International journal of hydrogen
energy, vol. 35, pp. 13021-13027, 2010.
[123] M. Pöschl, S. Ward, and P. Owende, "Evaluation of energy efficiency of various biogas
production and utilization pathways," Applied Energy, vol. 87, pp. 3305-3321, 2010.
[124] T. Sato, T. Suzuki, M. Aketa, Y. Ishiyama, K. Mimura, and N. Itoh, "Steam reforming of
biogas mixtures with a palladium membrane reactor system," Chemical Engineering
Science, vol. 65, pp. 451-457, 2010.
[125] H. J. Alves, C. B. Junior, R. R. Niklevicz, E. P. Frigo, M. S. Frigo, and C. H. Coimbra-
Araujo, "Overview of hydrogen production technologies from biogas and the applications
in fuel cells," international journal of hydrogen energy, vol. 38, pp. 5215-5225, 2013.
[126] S. M. Correa, "A Review of NOx Formation Under Gas-turbine Combustion Conditions,"
Combustion Science and Technology, vol. 87, pp. 329-362, 1993.
[127] R. Steele, A. Jarrett, P. C. Malte, J. Tonouchi, and D. Nicol, "Variables affecting NOx
formation in lean-premixed combustion," in ASME 1995 International Gas Turbine and
Aeroengine Congress and Exposition, 1995, pp. V003T06A018-V003T06A018.
133
[128] J.-Y. Ren, W. Qin, F. Egolfopoulos, H. Mak, and T. Tsotsis, "Methane reforming and its
potential effect on the efficiency and pollutant emissions of lean methane–air
combustion," Chemical Engineering Science, vol. 56, pp. 1541-1549, 2001.
[129] J.-Y. Ren, F. Egolfopoulos, and T. Tsotsis, "NO x emission control of lean methane-air
combustion with addition of methane reforming products," Combustion Science and
Technology, vol. 174, pp. 181-205, 2002.
[130] Y. S. Sanusi, M. A. Habib, and E. M. Mokheimer, "Experimental study on the effect of
hydrogen enrichment of methane on the stability and emission of nonpremixed swirl
stabilized combustor," Journal of Energy Resources Technology, vol. 137, p. 032203,
2015.
[131] C. Park, C. Kim, Y. Choi, S. Won, and Y. Moriyoshi, "The influences of hydrogen on the
performance and emission characteristics of a heavy duty natural gas engine,"
International journal of hydrogen energy, vol. 36, pp. 3739-3745, 2011.
[132] L. García, "Hydrogen production by steam reforming of natural gas and other
nonrenewable feedstocks," Compendium of Hydrogen Energy, pp. 83-107, 2015.
[133] D. B. Levin, L. Pitt, and M. Love, "Biohydrogen production: prospects and limitations to
practical application," International journal of hydrogen energy, vol. 29, pp. 173-185,
2004.
[134] S. Soltani, M. Sahimi, and T. Tsotsis, "Catalytic Membrane Reactors: A Brief
Overview," Encyclopedia of Membrane Science and Technology, 2013.
[135] Y.-J. Wu, P. Li, J.-G. Yu, A. F. Cunha, and A. E. Rodrigues, "Progress on sorption-
enhanced reaction process for hydrogen production," Reviews in Chemical Engineering,
vol. 32, pp. 271-303, 2016.
[136] P. K. Liu, M. Sahimi, and T. T. Tsotsis, "Process intensification in hydrogen production
from coal and biomass via the use of membrane-based reactive separations," Current
Opinion in Chemical Engineering, vol. 1, pp. 342-351, 2012.
[137] S. R. Kersten, W. Prins, B. Van der Drift, and W. P. van Swaaij, "Principles of a novel
multistage circulating fluidized bed reactor for biomass gasification," Chemical
Engineering Science, vol. 58, pp. 725-731, 2003.
[138] Z. Wu, C. Wu, H. Huang, S. Zheng, and X. Dai, "Test results and operation performance
analysis of a 1-MW biomass gasification electric power generation system," Energy &
fuels, vol. 17, pp. 619-624, 2003.
[139] M. L. de Souza-Santos, "A study of thermochemically recuperative power generation
systems using natural gas," Fuel, vol. 76, pp. 593-601, 1997.
[140] S. Harvey, "Analysis of a reheat gas turbine cycle with chemical recuperation using
Aspen," Energy conversion and management, vol. 38, pp. 1671-1679, 1997.
[141] D. Meyers and J. Kubesh, "The hybrid rich-burn/lean-burn engine," Journal of
engineering for gas turbines and power, vol. 119, pp. 243-249, 1997.
[142] C. Jeong, T. Kim, K. Lee, S. Song, and K. M. Chun, "Generating efficiency and
emissions of a spark-ignition gas engine generator fuelled with biogas–hydrogen blends,"
International Journal of hydrogen energy, vol. 34, pp. 9620-9627, 2009.
[143] S. H. Yoon and C. S. Lee, "Experimental investigation on the combustion and exhaust
emission characteristics of biogas–biodiesel dual-fuel combustion in a CI engine," Fuel
processing technology, vol. 92, pp. 992-1000, 2011.
[144] J. R. Rostrup-Nielsen, "Syngas in perspective," Catalysis today, vol. 71, pp. 243-247,
2002.
134
[145] M. Twigg, Catalyst Handbook, 2nd ed. London, UK: Manson Publishing, 1996.
[146] J. Rodriguez, E. Romeo, J. Fierro, J. Santamaria, and A. Monzon, "Deactivation by
coking and poisoning of spinel-type Ni catalysts," Catalysis today, vol. 37, pp. 255-265,
1997.
[147] H. S. Bengaard, J. K. Nørskov, J. Sehested, B. Clausen, L. Nielsen, A. Molenbroek, et al.,
"Steam reforming and graphite formation on Ni catalysts," Journal of Catalysis, vol. 209,
pp. 365-384, 2002.
[148] K. Urasaki, Y. Sekine, S. Kawabe, E. Kikuchi, and M. Matsukata, "Catalytic activities
and coking resistance of Ni/perovskites in steam reforming of methane," Applied
Catalysis A: General, vol. 286, pp. 23-29, 2005.
[149] S. Sangsong, M. Phongaksorn, S. Tungkamani, T. Sornchamni, and R. Chuvaree, "Dry
Methane Reforming Performance of Ni-based Catalyst Coated onto Stainless Steel
Substrate," Energy Procedia, vol. 79, pp. 137-142, 2015.
[150] S. J. Kho ET, Amal R, "Ni/TiO2 for low temperature steam reforming of methane," Che.
Eng. Sci., vol. 140, p. 10, 2016.
[151] A. L. Boehman and O. L. Corre, "Combustion of syngas in internal combustion engines,"
Combustion Science and Technology, vol. 180, pp. 1193-1206, 2008.
[152] C.-J. Sung and C. K. Law, "Fundamental combustion properties of H2/CO mixtures:
ignition and flame propagation at elevated pressures," Combustion Science and
Technology, vol. 180, pp. 1097-1116, 2008.
[153] C. Olm, I. G. Zsély, T. Varga, H. J. Curran, and T. Turányi, "Comparison of the
performance of several recent syngas combustion mechanisms," Combustion and Flame,
vol. 162, pp. 1793-1812, 2015.
[154] J.-Y. Ren, W. Qin, F. Egolfopoulos, and T. Tsotsis, "Strain-rate effects on hydrogen-
enhanced lean premixed combustion," Combustion and Flame, vol. 124, pp. 717-720,
2001.
[155] M. Emadi, D. Karkow, T. Salameh, A. Gohil, and A. Ratner, "Flame structure changes
resulting from hydrogen-enrichment and pressurization for low-swirl premixed methane–
air flames," international journal of hydrogen energy, vol. 37, pp. 10397-10404, 2012.
[156] J. Fu, C. Tang, W. Jin, L. D. Thi, Z. Huang, and Y. Zhang, "Study on laminar flame
speed and flame structure of syngas with varied compositions using OH-PLIF and
spectrograph," international journal of hydrogen energy, vol. 38, pp. 1636-1643, 2013.
[157] P. Griebel, E. Boschek, and P. Jansohn, "Lean blowout limits and NO x emissions of
turbulent, lean premixed, hydrogen-enriched methane/air flames at high pressure,"
Journal of engineering for gas turbines and power, vol. 129, pp. 404-410, 2007.
[158] F. Y. Hagos, A. R. A. Aziz, and S. A. Sulaiman, "Syngas (H 2/CO) in a spark-ignition
direct-injection engine. Part 1: Combustion, performance and emissions comparison with
CNG," International Journal of Hydrogen Energy, vol. 39, pp. 17884-17895, 2014.
[159] I. Hraiech, J.-C. Sautet, S. Yon, and A. Mhimid, "COMBUSTION OF HYTHANE
DILUTED WITH CO2," Thermal Science, vol. 19, pp. 1-10, 2015.
[160] M. Ilbas, A. Crayford, I. Yılmaz, P. Bowen, and N. Syred, "Laminar-burning velocities of
hydrogen–air and hydrogen–methane–air mixtures: an experimental study," International
Journal of Hydrogen Energy, vol. 31, pp. 1768-1779, 2006.
[161] A. E. Khalil and A. K. Gupta, "Hydrogen addition effects on high intensity distributed
combustion," Applied energy, vol. 104, pp. 71-78, 2013.
135
[162] G. Lim, S. Lee, C. Park, Y. Choi, and C. Kim, "Effect of ignition timing retard strategy
on NOx reduction in hydrogen-compressed natural gas blend engine with increased
compression ratio," International Journal of Hydrogen Energy, vol. 39, pp. 2399-2408,
2014.
[163] D. Littlejohn, A. J. Majeski, S. Tonse, C. Castaldini, and R. K. Cheng, "Laboratory
investigation of an ultralow NO x premixed combustion concept for industrial boilers,"
Proceedings of the Combustion Institute, vol. 29, pp. 1115-1121, 2002.
[164] J. Liu, X. Zhang, T. Wang, X. Hou, J. Zhang, and S. Zheng, "Numerical study of the
chemical, thermal and diffusion effects of H2 and CO addition on the laminar flame
speeds of methane–air mixture," International Journal of Hydrogen Energy, vol. 40, pp.
8475-8483, 2015.
[165] C. J. Mordaunt and W. C. Pierce, "Design and preliminary results of an atmospheric-
pressure model gas turbine combustor utilizing varying CO 2 doping concentration in CH
4 to emulate biogas combustion," Fuel, vol. 124, pp. 258-268, 2014.
[166] F. Moreno, J. Arroyo, M. Muñoz, and C. Monné, "Combustion analysis of a spark
ignition engine fueled with gaseous blends containing hydrogen," international journal of
hydrogen energy, vol. 37, pp. 13564-13573, 2012.
[167] C. Park, S. Park, Y. Lee, C. Kim, S. Lee, and Y. Moriyoshi, "Performance and emission
characteristics of a SI engine fueled by low calorific biogas blended with hydrogen,"
International Journal of hydrogen energy, vol. 36, pp. 10080-10088, 2011.
[168] E. Porpatham, A. Ramesh, and B. Nagalingam, "Effect of hydrogen addition on the
performance of a biogas fuelled spark ignition engine," International Journal of Hydrogen
Energy, vol. 32, pp. 2057-2065, 2007.
[169] M. D. Ruter, D. B. Olsen, M. V. Scotto, and M. A. Perna, "NO x reduction from a large
bore natural gas engine via reformed natural gas prechamber fueling optimization," Fuel,
vol. 91, pp. 298-306, 2012.
[170] P. Sayad, A. Scho ̈ nborn, and J. Klingmann, "Experimental investigations of the lean
blowout limit of different syngas mixtures in an atmospheric, premixed, variable-swirl
burner," Energy & Fuels, vol. 27, pp. 2783-2793, 2013.
[171] R. Schefer, "Hydrogen enrichment for improved lean flame stability," International
Journal of Hydrogen Energy, vol. 28, pp. 1131-1141, 2003.
[172] S. B. Shrestha and G. Karim, "Hydrogen as an additive to methane for spark ignition
engine applications," International Journal of Hydrogen Energy, vol. 24, pp. 577-586,
1999.
[173] S. B. Shrestha and G. Narayanan, "Landfill gas with hydrogen addition–A fuel for SI
engines," Fuel, vol. 87, pp. 3616-3626, 2008.
[174] F. Delattin, G. Di Lorenzo, S. Rizzo, S. Bram, and J. De Ruyck, "Combustion of syngas
in a pressurized microturbine-like combustor: Experimental results," Applied Energy,
vol. 87, pp. 1441-1452, 2010.
[175] F. Delattin, A. Rabhiou, S. Bram, J. D. Ruyck, R. Orbay, J. Klingmann, et al., "A
comparison between the combustion of natural gas and partially reformed natural gas in
an atmospheric lean premixed turbine-type combustor," Combustion Science and
Technology, vol. 180, pp. 1478-1501, 2008.
[176] F. Coppens, J. De Ruyck, and A. Konnov, "Effects of hydrogen enrichment on adiabatic
burning velocity and NO formation in methane+ air flames," Experimental Thermal and
Fluid Science, vol. 31, pp. 437-444, 2007.
136
[177] K. B. Fackler, M. Karalus, I. Novosselov, J. Kramlich, P. Malte, and S. Vijlee, "NO x
Behavior for Lean-Premixed Combustion of Alternative Gaseous Fuels," Journal of
Engineering for Gas Turbines and Power, vol. 138, p. 041504, 2016.
[178] K. B. Fackler, M. F. Karalus, I. V. Novosselov, J. C. Kramlich, and P. C. Malte,
"Experimental and numerical study of NOx formation from the lean premixed combustion
of CH4 mixed with CO2 and N2," Journal of Engineering for Gas Turbines and Power,
vol. 133, p. 121502, 2011.
[179] S. Göke, M. Füri, G. Bourque, B. Bobusch, K. Göckeler, O. Krüger, et al., "Influence of
steam dilution on the combustion of natural gas and hydrogen in premixed and rich-
quench-lean combustors," Fuel processing technology, vol. 107, pp. 14-22, 2013.
[180] M. C. Lee, S. B. Seo, J. Yoon, M. Kim, and Y. Yoon, "Experimental study on the effect
of N2, CO2, and steam dilution on the combustion performance of H 2 and CO synthetic
gas in an industrial gas turbine," Fuel, vol. 102, pp. 431-438, 2012.
[181] S. Gauthier, A. Nicolle, and D. Baillis, "Investigation of the flame structure and nitrogen
oxides formation in lean porous premixed combustion of natural gas/hydrogen blends,"
international journal of hydrogen energy, vol. 33, pp. 4893-4905, 2008.
[182] D. E. Giles, S. Som, and S. K. Aggarwal, "NO x emission characteristics of counterflow
syngas diffusion flames with airstream dilution," Fuel, vol. 85, pp. 1729-1742, 2006.
[183] G. M. Watson, J. D. Munzar, and J. M. Bergthorson, "NO formation in model syngas and
biogas blends," Fuel, vol. 124, pp. 113-124, 2014.
[184] S. Alavandi and A. Agrawal, "Experimental Study of Combustion of Hydrogen–
Syngas/Methane Fuel Mixtures in a Porous Burner," International Journal of Hydrogen
Energy, vol. 33, pp. 1407-1415, 2008.
[185] C. E. Arrieta and A. A. Amell, "Combustion Analysis of an Equimolar Mixture of
Methane and Syngas in a Surface-stabilized Combustion Burner for Household
Appliances," Fuel, vol. 137, pp. 11-20, 2014.
[186] J. Arroyo, F. Moreno, M. Muñoz, and C. Monné, "Efficiency and emissions of a spark
ignition engine fueled with synthetic gases obtained from catalytic decomposition of
biogas," International Journal of Hydrogen Energy, vol. 38, pp. 3784-3792, 2013.
[187] J. Arroyo, F. Moreno, M. Muñoz, C. Monné, and N. Bernal, "Combustion behavior of a
spark ignition engine fueled with synthetic gases derived from biogas," Fuel, vol. 117,
pp. 50-58, 2014.
[188] U. Azimov, E. Tomita, N. Kawahara, and Y. Harada, "Effect of syngas composition on
combustion and exhaust emission characteristics in a pilot-ignited dual-fuel engine
operated in PREMIER combustion mode," international journal of hydrogen energy, vol.
36, pp. 11985-11996, 2011.
[189] W.-T. CHEN, "Effects of CO addition on the lean premixed CH 4/Air flame," Acta
Physico-Chimica Sinica, vol. 26, pp. 1481-1487, 2010.
[190] L. Chen, S. Shiga, and M. Araki, "Combustion characteristics of an SI engine fueled with
H2–CO blended fuel and diluted by CO2," international journal of hydrogen energy, vol.
37, pp. 14632-14639, 2012.
[191] T. Cheng, Y.-C. Chang, Y.-C. Chao, G.-B. Chen, Y.-H. Li, and C.-Y. Wu, "An
experimental and numerical study on characteristics of laminar premixed H 2/CO/CH
4/air flames," International journal of hydrogen energy, vol. 36, pp. 13207-13217, 2011.
[192] E. R. Hawkes and J. H. Chen, "Direct numerical simulation of hydrogen-enriched lean
premixed methane–air flames," Combustion and Flame, vol. 138, pp. 242-258, 2004.
137
[193] Y. He, Z. Wang, L. Yang, R. Whiddon, Z. Li, J. Zhou, et al., "Investigation of laminar
flame speeds of typical syngas using laser based Bunsen method and kinetic simulation,"
Fuel, vol. 95, pp. 206-213, 2012.
[194] E. Hu, Z. Huang, J. Zheng, Q. Li, and J. He, "Numerical study on laminar burning
velocity and NO formation of premixed methane–hydrogen–air flames," International
journal of hydrogen energy, vol. 34, pp. 6545-6557, 2009.
[195] Z. Huang, Y. Zhang, K. Zeng, B. Liu, Q. Wang, and D. Jiang, "Measurements of laminar
burning velocities for natural gas–hydrogen–air mixtures," Combustion and Flame, vol.
146, pp. 302-311, 2006.
[196] O. Laget, S. Richard, D. Serrano, and D. Soleri, "Combining experimental and numerical
investigations to explore the potential of downsized engines operating with
methane/hydrogen blends," international journal of hydrogen energy, vol. 37, pp. 11514-
11530, 2012.
[197] C. Liu, B. Yan, G. Chen, and X.-S. Bai, "Structures and burning velocity of biomass
derived gas flames," International Journal of hydrogen energy, vol. 35, pp. 542-555,
2010.
[198] C. Mandilas, M. Ormsby, C. Sheppard, and R. Woolley, "Effects of hydrogen addition on
laminar and turbulent premixed methane and iso-octane–air flames," Proceedings of the
Combustion Institute, vol. 31, pp. 1443-1450, 2007.
[199] T. Tahtouh, F. Halter, E. Samson, and C. Mounaïm-Rousselle, "Effects of hydrogen
addition and nitrogen dilution on the laminar flame characteristics of premixed methane–
air flames," International Journal of Hydrogen Energy, vol. 34, pp. 8329-8338, 2009.
[200] J. Wang, Z. Huang, H. Kobayashi, and Y. Ogami, "Laminar burning velocities and flame
characteristics of CO–H2–CO2–O2 mixtures," international journal of hydrogen energy,
vol. 37, pp. 19158-19167, 2012.
[201] Z. Wang, W. Weng, Y. He, Z. Li, and K. Cen, "Effect of H 2/CO ratio and N2/CO2
dilution rate on laminar burning velocity of syngas investigated by direct measurement
and simulation," Fuel, vol. 141, pp. 285-292, 2015.
[202] P. Zahedi and K. Yousefi, "Effects of pressure and carbon dioxide, hydrogen and
nitrogen concentration on laminar burning velocities and NO formation of methane-air
mixtures," Journal of Mechanical Science and Technology, vol. 28, pp. 377-386, 2014.
[203] Y. Zhang, W. Shen, M. Fan, H. Zhang, and S. Li, "Laminar flame speed studies of lean
premixed H 2/CO/air flames," Combustion and Flame, vol. 161, pp. 2492-2495, 2014.
[204] E. Hu, J. Fu, L. Pan, X. Jiang, Z. Huang, and Y. Zhang, "Experimental and numerical
study on the effect of composition on laminar burning velocities of H 2/CO/N2/CO2/air
mixtures," international journal of hydrogen energy, vol. 37, pp. 18509-18519, 2012.
[205] Y. Cheng, C. Tang, and Z. Huang, "Kinetic analysis of H 2 addition effect on the laminar
flame parameters of the C1–C4 n-alkane-air mixtures: From one step overall assumption
to detailed reaction mechanism," International Journal of Hydrogen Energy, vol. 40, pp.
703-718, 2015.
[206] F. Halter, C. Chauveau, N. Djebaili-Chaumeix, and I. Gökalp, "Characterization of the
effects of pressure and hydrogen concentration on laminar burning velocities of methane–
hydrogen–air mixtures," Proceedings of the Combustion Institute, vol. 30, pp. 201-208,
2005.
138
[207] F. H. Coppens and A. A. Konnov, "The effects of enrichment by H 2 on propagation
speeds in adiabatic flat and cellular premixed flames of CH 4+ O2+ CO2," Fuel, vol. 87,
pp. 2866-2870, 2008.
[208] A. K. Das, K. Kumar, and C.-J. Sung, "Laminar flame speeds of moist syngas mixtures,"
Combustion and Flame, vol. 158, pp. 345-353, 2011.
[209] D. Singh, T. Nishiie, S. Tanvir, and L. Qiao, "An experimental and kinetic study of
syngas/air combustion at elevated temperatures and the effect of water addition," Fuel,
vol. 94, pp. 448-456, 2012.
[210] D. G. Pugh, A. P. Crayford, P. J. Bowen, and M. Al-Naama, "Parametric investigation of
water loading on heavily carbonaceous syngases," Combustion and Flame, vol. 164, pp.
126-136, 2016.
[211] Y. He, Z. Wang, W. Weng, Y. Zhu, J. Zhou, and K. Cen, "Effects of CO content on
laminar burning velocity of typical syngas by heat flux method and kinetic modeling,"
International Journal of Hydrogen Energy, vol. 39, pp. 9534-9544, 2014.
[212] N. Hinton and R. Stone, "Laminar burning velocity measurements of methane and carbon
dioxide mixtures (biogas) over wide ranging temperatures and pressures," Fuel, vol. 116,
pp. 743-750, 2014.
[213] M. A. Delmas and V. C. Burbano, "The drivers of greenwashing," California
Management Review, vol. 54, pp. 64-87, 2011.
[214] V. R. Kishore, M. Ravi, and A. Ray, "Adiabatic burning velocity and cellular flame
characteristics of H2–CO–CO2–air mixtures," Combustion and Flame, vol. 158, pp. 2149-
2164, 2011.
[215] D. Lapalme and P. Seers, "Influence of CO 2, CH4, and initial temperature on H2/CO
laminar flame speed," International Journal of Hydrogen Energy, vol. 39, pp. 3477-3486,
2014.
[216] Y. Zhang, W. Shen, H. Zhang, Y. Wu, and J. Lu, "Effects of inert dilution on the
propagation and extinction of lean premixed syngas/air flames," Fuel, vol. 157, pp. 115-
121, 2015.
[217] C. Prathap, A. Ray, and M. Ravi, "Effects of dilution with carbon dioxide on the laminar
burning velocity and flame stability of H2–CO mixtures at atmospheric condition,"
Combustion and Flame, vol. 159, pp. 482-492, 2012.
[218] C. De Smet, M. De Croon, R. Berger, G. Marin, and J. Schouten, "Design of adiabatic
fixed-bed reactors for the partial oxidation of methane to synthesis gas. Application to
production of methanol and hydrogen-for-fuel-cells," Chemical Engineering Science, vol.
56, pp. 4849-4861, 2001.
[219] T. Numaguchi and K. Kikuchi, "Intrinsic kinetics and design simulation in a complex
reaction network; steam-methane reforming," Chemical engineering science, vol. 43, pp.
2295-2301, 1988.
[220] B. Fayyaz, A. Harale, B.-G. Park, P. K. Liu, M. Sahimi, and T. T. Tsotsis, "Design
aspects of hybrid adsorbent-membrane reactors for hydrogen production," Industrial &
engineering chemistry research, vol. 44, pp. 9398-9408, 2005.
[221] C. Wilke, "A viscosity equation for gas mixtures," The journal of chemical physics, vol.
18, pp. 517-519, 1950.
[222] D. W. Green and R. H. Perry, Perry's chemical engineers' handbook, 8th ed. New York,
Chicago, San Francisco, Lisbon, London, Madrid, Mexico City, Milan, New Delhi, San
Juan, Seoul, Singapore, Sydney, Toronto: McGraw-Hill, 2008.
139
[223] J. R. Elliott and C. T. Lira, Introductory Chemical Engineering Thermodynamics, 2nd
ed.: Prentice Hall, 2012.
[224] M. Mbodji, J.-M. Commenge, L. Falk, D. Di Marco, F. Rossignol, L. Prost, et al., "Steam
methane reforming reaction process intensification by using a millistructured reactor:
Experimental setup and model validation for global kinetic reaction rate estimation,"
Chemical engineering journal, vol. 207, pp. 871-884, 2012.
[225] S. Dabir, "Biogas Reforming: Conventional and Reactive Separation Processes and the
Preparation and Charcterization of Related Materials," PhD PhD Dissertation, Mork
Family Department of Chemical Engineering and Materials Science, University of
Southern California, 2016.
[226] H. Zhen, Z. Wei, C. Leung, C. Cheung, and Z. Huang, "Emission of impinging biogas/air
premixed flame with hydrogen enrichment," international journal of hydrogen energy,
vol. 41, pp. 2087-2095, 2016.
[227] F. Christodoulou and A. Megaritis, "The effect of reformer gas mixture on the
performance and emissions of an HSDI diesel engine," international journal of hydrogen
energy, vol. 39, pp. 9798-9808, 2014.
[228] F. Christodoulou and A. Megaritis, "Experimental investigation of the effects of
simultaneous hydrogen and nitrogen addition on the emissions and combustion of a diesel
engine," international journal of hydrogen energy, vol. 39, pp. 2692-2702, 2014.
[229] R. Bhave, Inorganic Membranes Synthesis, Characteristics and Applications: Synthesis,
characteristics, and applications: Springer Science & Business Media, 2012.
[230] L. Cot, C. Guizard, A. Julbe, and A. Larbot, "Preparation and Application of Inorganic
Membranes," in Membrane Processes in Separation and Purification, ed: Springer, 1994,
pp. 431-442.
[231] G. Liu, Y. Liu, H. Shi, and Y. Qian, "Application of inorganic membranes in the alkali
recovery process," Desalination, vol. 169, pp. 193-205, 2004.
[232] X. Tan and K. Li, Inorganic Membrane Reactors: Fundamentals and Applications: John
Wiley & Sons, 2015.
[233] D. White, R. Labayru, H. Obermoller, and X. Deryde, "The application of inorganic
membranes to electrodialysis processes," Desalination, vol. 84, pp. 177-188, 1991.
[234] F. Gallucci, E. Fernandez, P. Corengia, and M. van Sint Annaland, "Recent advances on
membranes and membrane reactors for hydrogen production," Chemical Engineering
Science, vol. 92, pp. 40-66, 2013.
[235] Y. Takeda, N. Shibata, and Y. Kubo, "SiC Coating on Porous Gamma-Al 2O3 Using
Alternative-Supply CVI Method," Journal of the Ceramic Society of Japan, vol. 109, pp.
305-309, 2001.
[236] X. Pages, V. Rouessac, D. Cot, G. Nabias, and J. Durand, "Gas permeation of PECVD
membranes inside alumina substrate tubes," Separation and purification technology, vol.
25, pp. 399-406, 2001.
[237] B.-K. Sea, K. Ando, K. Kusakabe, and S. Morooka, "Separation of hydrogen from steam
using a SiC-based membrane formed by chemical vapor deposition of triisopropylsilane,"
Journal of membrane science, vol. 146, pp. 73-82, 1998.
[238] R. J. Ciora, B. Fayyaz, P. K. Liu, V. Suwanmethanond, R. Mallada, M. Sahimi, et al.,
"Preparation and reactive applications of nanoporous silicon carbide membranes,"
Chemical engineering science, vol. 59, pp. 4957-4965, 2004.
140
[239] L. V. Interrante, C. Whitmarsh, and W. Sherwood, "Fabrication of SiC matrix composites
by liquid phase infiltration with a polymeric precursor," in MRS Proceedings, 1994, p.
139.
[240] L. V. Interrante, C. Whitmarsh, W. Sherwood, H.-J. Wu, R. Lewis, and G. Maciel, "High
yield polycarbosilane precursors to stoichiometric SiC. Synthesis, pyrolysis and
application," in MRS Proceedings, 1994, p. 595.
[241] Z. Li, K. Kusakabe, and S. Morooka, "Preparation of thermostable amorphous Si-C-O
membrane and its application to gas separation at elevated temperature," Journal of
membrane science, vol. 118, pp. 159-168, 1996.
[242] Z. Li, K. Kusakabe, and S. Morooka, "Pore structure and permeance of amorphous Si-CO
membranes with high durability at elevated temperature," Separation science and
technology, vol. 32, pp. 1233-1254, 1997.
[243] M. H. Suh, W.-T. Kwon, E. B. Kim, S.-R. Kim, S. Y. Bae, D. J. Choi, et al., "H-2
permeable nanoporous SiC membrane for an IGCC application," J. Ceram. Process. Res,
vol. 10, pp. 359-363, 2009.
[244] L. L. Lee and D. S. Tsai, "A Hydrogen‐Permselective Silicon Oxycarbide Membrane
Derived from Polydimethylsilane," Journal of the American Ceramic Society, vol. 82, pp.
2796-2800, 1999.
[245] L.-L. Lee and D.-S. Tsai, "Synthesis and permeation properties of silicon-carbon-based
inorganic membrane for gas separation," Industrial & engineering chemistry research,
vol. 40, pp. 612-616, 2001.
[246] C.-L. Chen and D.-S. Tsai, "Permeation properties of microporous membranes prepared
via coating of evaporated polydimethylsilane," Journal of membrane science, vol. 237,
pp. 163-165, 2004.
[247] R. A. Wach, M. Sugimoto, A. Idesaki, and M. Yoshikawa, "Molecular sieve SiC-based
membrane for hydrogen separation produced by radiation curing of preceramic
polymers," Materials Science and Engineering: B, vol. 140, pp. 81-89, 2007.
[248] R. A. Wach, M. Sugimoto, and M. Yoshikawa, "Formation of silicone carbide membrane
by radiation curing of polycarbosilane and polyvinylsilane and its gas separation up to
250 C," Journal of the American Ceramic Society, vol. 90, pp. 275-278, 2007.
[249] K. Kita, N. Kondo, Y. Izutsu, and H. Kita, "Investigation of the properties of SiC
membrane on alumina by using polycarbosilane," Materials Letters, vol. 75, pp. 134-136,
2012.
[250] H. Suda, H. Yamauchi, Y. Uchimaru, I. Fujiwara, and K. Haraya, "Preparation and gas
permeation properties of silicon carbide-based inorganic membranes for hydrogen
separation," Desalination, vol. 193, pp. 252-255, 2006.
[251] H. Suda, H. Yamauchi, Y. Uchimaru, I. Fujiwara, and K. Haraya, "Structural evolution
during conversion of polycarbosilane precursor into silicon carbide-based microporous
membranes," Journal of the Ceramic Society of Japan, vol. 114, p. 6, 2006.
[252] T. Nagano, K. Sato, T. Saitoh, and Y. Iwamoto, "Gas permeation properties of
amorphous SiC membranes synthesized from polycarbosilane without oxygen-curing
process," Journal of the Ceramic Society of Japan, vol. 114, p. 6, 2006.
[253] T. Nagano, K. Sato, S. Fujisaki, T. Saitoh, and Y. Iwamoto, "Helium‐Permselective
Amorphous SiC Membrane Modified by Chemical Vapor Infiltration," Soft Materials,
vol. 4, pp. 109-122, 2007.
141
[254] Z. Y. Li, K. Kusakabe, and S. Morooka, "Pore structure and permeance of amorphous Si-
C-O membranes with high durability at elevated temperature," Separation Science and
Technology, vol. 32, pp. 1233-1254, 1997 1997.
[255] Y. Takeda, K. Nakamura, K. Maeda, and Y. Matsushita, "Effects of Elemental Additives
on Electrical Resistivity of Silicon Carbide Ceramics," Journal of the American Ceramic
Society, vol. 70, pp. C-266-C-267, 1987.
[256] K. Schulz and M. Durst, "Advantages of an integrated system for hot gas filtration using
rigid ceramic elements," Filtration & separation, vol. 31, pp. 25-28, 1994.
[257] S. H. Kenawy and W. M. N. Nour, "Microstructural evaluation of thermally fatigued SiC-
reinforced Al2O3/ZrO2 matrix composites," Journal of Materials Science, vol. 40, pp.
3789-3793, 2005/07/01 2005.
[258] B. Elyassi, W. Deng, M. Sahimi, and T. T. Tsotsis, "On the use of porous and nonporous
fillers in the fabrication of silicon carbide membranes," Industrial & Engineering
Chemistry Research, vol. 52, pp. 10269-10275, 2013.
[259] B. Elyassi, M. Sahimi, and T. T. Tsotsis, "Silicon carbide membranes for gas separation
applications," Journal of membrane science, vol. 288, pp. 290-297, 2007.
[260] B. Elyassi, M. Sahimi, and T. T. Tsotsis, "A novel sacrificial interlayer-based method for
the preparation of silicon carbide membranes," Journal of Membrane Science, vol. 316,
pp. 73-79, 2008.
[261] A. Takeyama, M. Sugimoto, and M. Yoshikawa, "Gas permeation property of SiC
membrane using curing of polymer precursor film by electron beam irradiation in helium
atmosphere," Materials transactions, vol. 52, pp. 1276-1280, 2011.
[262] B. Elyassi, M. Sahimi, and T. T. Tsotsis, "Inorganic Membranes," in Encyclopedia of
Chemical Processing. vol. 1, S. Lee, Ed., ed: Taylor & Francis Group, LLC, 2009, pp. 1-
16.
[263] V. Suwanmethanond, E. Goo, P. K. T. Liu, G. Johnston, M. Sahimi, and T. T. Tsotsis,
"Porous silicon carbide sintered substrates for high-temperature membranes," Industrial
& Engineering Chemistry Research, vol. 39, pp. 3264-3271, Sep 2000.
[264] K. König, V. Boffa, B. Buchbjerg, A. Farsi, M. L. Christensen, G. Magnacca, et al.,
"One-step deposition of ultrafiltration SiC membranes on macroporous SiC supports,"
Journal of Membrane Science, vol. 472, pp. 232-240, 2014.
[265] M. Facciotti, V. Boffa, G. Magnacca, L. B. Jørgensen, P. K. Kristensen, A. Farsi, et al.,
"Deposition of thin ultrafiltration membranes on commercial SiC microfiltration tubes,"
Ceramics International, vol. 40, pp. 3277-3285, 2014.
[266] S. Samingprai, S. Tantayanon, and Y. H. Ma, "Chromium oxide intermetallic diffusion
barrier for palladium membrane supported on porous stainless steel," Journal of
Membrane Science, vol. 347, pp. 8-16, Feb 2010.
[267] A. E. Ayturk, I. P. Mardilovich, E. E. Engwall, and Y. H. Ma, "Synthesis of composite
Pd-porous stainless steel (PSS) membranes with a Pd/Ag intermetallic diffusion barrier,"
Journal of Membrane Science, vol. 285, pp. 385-394, Nov 2006.
[268] W. Deng, X. Yu, M. Sahimi, and T. T. Tsotsis, "Highly permeable porous silicon carbide
support tubes for the preparation of nanoporous inorganic membranes," Journal of
Membrane Science, vol. 451, pp. 192-204, 2014.
[269] Y. Zhou, M. Fukushima, H. Miyazaki, Y.-i. Yoshizawa, K. Hirao, Y. Iwamoto, et al.,
"Preparation and characterization of tubular porous silicon carbide membrane supports,"
Journal of Membrane Science, vol. 369, pp. 112-118, Mar 1 2011.
142
[270] J. J. Burke, A. Gorum, and R. Katz, "Ceramics for high-performance applications,"
Brook Hill Publishing Co., Chestnut Hill, MA1974.
[271] J. Hojo, "Sintering behavior of ultrafine silicon carbide powder," in Silicon Carbide
Ceramics—1, ed: Springer, 1991, pp. 149-168.
[272] H. Tanaka, "Sintering of silicon carbide," in Silicon Carbide Ceramics—1, ed: Springer,
1991, pp. 213-238.
[273] M. Abdollahi, J. Yu, H. T. Hwang, P. K. Liu, R. Ciora, M. Sahimi, et al., "Process
intensification in hydrogen production from biomass-derived syngas," Industrial &
Engineering Chemistry Research, vol. 49, pp. 10986-10993, 2010.
[274] M. Abdollahi, J. Yu, P. K. Liu, R. Ciora, M. Sahimi, and T. T. Tsotsis, "Hydrogen
production from coal-derived syngas using a catalytic membrane reactor based process,"
Journal of Membrane Science, vol. 363, pp. 160-169, 2010.
[275] K. S. Rothenberger, A. V. Cugini, B. H. Howard, R. P. Killmeyer, M. V. Ciocco, B. D.
Morreale, et al., "High pressure hydrogen permeance of porous stainless steel coated with
a thin palladium film via electroless plating," Journal of membrane science, vol. 244, pp.
55-68, 2004.
[276] C.-C. Chang, C.-L. Pai, W.-C. Chen, and S. A. Jenekhe, "Spin coating of conjugated
polymers for electronic and optoelectronic applications," Thin solid films, vol. 479, pp.
254-260, 2005.
[277] D. B. Hall, P. Underhill, and J. M. Torkelson, "Spin coating of thin and ultrathin polymer
films," Polymer Engineering & Science, vol. 38, pp. 2039-2045, 1998.
[278] S. L. Clark, E. S. Handy, M. F. Rubner, and P. T. Hammond, "Creating Microstructures
of Luminescent Organic Thin Films Using Layer‐by‐Layer Assembly," Advanced
Materials, vol. 11, pp. 1031-1035, 1999.
[279] Y. Shimazaki, M. Mitsuishi, S. Ito, and M. Yamamoto, "Preparation of the layer-by-layer
deposited ultrathin film based on the charge-transfer interaction," Langmuir, vol. 13, pp.
1385-1387, 1997.
[280] C. Basceri, S. Streiffer, A. I. Kingon, and R. Waser, "The dielectric response as a
function of temperature and film thickness of fiber-textured (Ba, Sr) TiO 3 thin films
grown by chemical vapor deposition," Journal of Applied Physics, vol. 82, pp. 2497-
2504, 1997.
[281] W. Shan, T. Schmidt, X. Yang, S. Hwang, J. Song, and B. Goldenberg, "Temperature
dependence of interband transitions in GaN grown by metalorganic chemical vapor
deposition," Applied physics letters, vol. 66, pp. 985-987, 1995.
[282] M. Gupta and K. K. Gleason, "Initiated chemical vapor deposition of poly (1H, 1H, 2H,
2H-perfluorodecyl acrylate) thin films," Langmuir, vol. 22, pp. 10047-10052, 2006.
[283] Y. Mao and K. K. Gleason, "Hot filament chemical vapor deposition of poly (glycidyl
methacrylate) thin films using tert-butyl peroxide as an initiator," Langmuir, vol. 20, pp.
2484-2488, 2004.
[284] K. Chan and K. K. Gleason, "Initiated chemical vapor deposition of linear and cross-
linked poly (2-hydroxyethyl methacrylate) for use as thin-film hydrogels," Langmuir, vol.
21, pp. 8930-8939, 2005.
[285] M. Gupta and K. K. Gleason, "Surface modification of high aspect ratio structures with
fluoropolymer coatings using chemical vapor deposition," Thin Solid Films, vol. 517, pp.
3547-3550, 2009.
143
[286] M. Gupta, V. Kapur, N. M. Pinkerton, and K. K. Gleason, "Initiated chemical vapor
deposition (iCVD) of conformal polymeric nanocoatings for the surface modification of
high-aspect-ratio pores," Chemistry of Materials, vol. 20, pp. 1646-1651, 2008.
[287] K. K. Lau and K. K. Gleason, "Particle Surface Design using an All‐Dry Encapsulation
Method," Advanced Materials, vol. 18, pp. 1972-1977, 2006.
[288] W. O'shaughnessy, S. Murthy, D. Edell, and K. Gleason, "Stable biopassive insulation
synthesized by initiated chemical vapor deposition of poly (1, 3, 5-
trivinyltrimethylcyclotrisiloxane)," Biomacromolecules, vol. 8, pp. 2564-2570, 2007.
[289] M. E. Alf, P. D. Godfrin, T. A. Hatton, and K. K. Gleason, "Sharp hydrophilicity
switching and conformality on nanostructured surfaces prepared via initiated chemical
vapor deposition (iCVD) of a novel thermally responsive copolymer," Macromolecular
rapid communications, vol. 31, pp. 2166-2172, 2010.
[290] M. Ma, Y. Mao, M. Gupta, K. K. Gleason, and G. C. Rutledge, "Superhydrophobic
fabrics produced by electrospinning and chemical vapor deposition," Macromolecules,
vol. 38, pp. 9742-9748, 2005.
[291] W. E. Tenhaeff and K. K. Gleason, "Initiated and oxidative chemical vapor deposition of
polymeric thin films: iCVD and oCVD," Advanced Functional Materials, vol. 18, pp.
979-992, 2008.
[292] S. Coulson, I. Woodward, J. Badyal, S. Brewer, and C. Willis, "Ultralow surface energy
plasma polymer films," Chemistry of materials, vol. 12, pp. 2031-2038, 2000.
[293] S. Morita, S. Hattori, and R. d'Agostino, "Plasma Deposition, Treatment, and Etching of
Polymers," ed: Academic Press, San Diego, 1990.
[294] S. H. Baxamusa and K. K. Gleason, "Initiated chemical vapor deposition of polymer
films on nonplanar substrates," Thin Solid Films, vol. 517, pp. 3536-3538, 2009.
[295] K. K. Lau and K. K. Gleason, "Initiated chemical vapor deposition (iCVD) of poly (alkyl
acrylates): an experimental study," Macromolecules, vol. 39, pp. 3688-3694, 2006.
[296] B. Elyassi, T. W. Kim, M. Sahimi, and T. T. Tsotsis, "Effect of polystyrene on the
morphology and physical properties of silicon carbide nanofibers," Materials Chemistry
and Physics, vol. 118, pp. 259-263, 2009.
[297] Y. Takeda, K. Nakamura, K. Maeda, and Y. Matsushita, "Effects of elemental additives
on electrical resistivity of silicon carbide ceramics," Journal of the American Ceramic
Society, vol. 70, 1987.
[298] A. Rosenbloom, D. Sipe, Y. Shishkin, Y. Ke, R. Devaty, and W. Choyke, "Nanoporous
SiC: A candidate semi-permeable material for biomedical applications," Biomedical
microdevices, vol. 6, pp. 261-267, 2004.
[299] S. Kenawy and W. Nour, "Microstructural evaluation of thermally fatigued SiC-
reinforced Al2O3/ZrO2 matrix composites," Journal of materials science, vol. 40, pp.
3789-3793, 2005.
[300] C. A. Zorman, A. J. Fleischman, A. S. Dewa, M. Mehregany, C. Jacob, S. Nishino, et al.,
"Epitaxial growth of 3C–SiC films on 4 in. diam (100) silicon wafers by atmospheric
pressure chemical vapor deposition," Journal of Applied Physics, vol. 78, pp. 5136-5138,
1995.
[301] M. G. Sedigh, M. Jahangiri, P. K. Liu, M. Sahimi, and T. T. Tsotsis, "Structural
characterization of polyetherimide‐based carbon molecular sieve membranes," AIChE
journal, vol. 46, pp. 2245-2255, 2000.
144
[302] M. G. Sedigh, W. J. Onstot, L. Xu, W. L. Peng, T. T. Tsotsis, and M. Sahimi,
"Experiments and simulation of transport and separation of gas mixtures in carbon
molecular sieve membranes," The Journal of Physical Chemistry A, vol. 102, pp. 8580-
8589, 1998.
[303] M. G. Sedigh, L. Xu, T. T. Tsotsis, and M. Sahimi, "Transport and morphological
characteristics of polyetherimide-based carbon molecular sieve membranes," Industrial &
engineering chemistry research, vol. 38, pp. 3367-3380, 1999.
[304] V. Suwanmethanond, E. Goo, P. K. Liu, G. Johnston, M. Sahimi, and T. T. Tsotsis,
"Porous silicon carbide sintered substrates for high-temperature membranes," Industrial
& engineering chemistry research, vol. 39, pp. 3264-3271, 2000.
[305] L.-S. Hong and H.-T. Lai, "Pore structure modification of alumina support by SiC-Si 3N4
nanoparticles prepared by the particle precipitation aided chemical vapor deposition,"
Industrial & engineering chemistry research, vol. 38, pp. 950-957, 1999.
[306] L. L. Lee and D. S. Tsai, "Silicon carbide membranes modified by chemical vapor
deposition using species of low sticking coefficients in a silane/acetylene reaction
system," Journal of the American Ceramic Society, vol. 81, pp. 159-165, 1998.
[307] B. Chen, S. Seidel, H. Hori, and M. Gupta, "Self-assembly of pillars modified with vapor
deposited polymer coatings," ACS applied materials & interfaces, vol. 3, pp. 4201-4205,
2011.
[308] R. J. Frank-Finney, P. D. Haller, and M. Gupta, "Ultrathin free-standing polymer films
deposited onto patterned ionic liquids and silicone oil," Macromolecules, vol. 45, pp.
165-170, 2011.
[309] P. Haller, R. Frank-Finney, and M. Gupta, "Vapor-phase free radical polymerization in
the presence of an ionic liquid," Macromolecules, vol. 44, pp. 2653-2659, 2011.
[310] C. T. Riche, B. C. Marin, N. Malmstadt, and M. Gupta, "Vapor deposition of cross-linked
fluoropolymer barrier coatings onto pre-assembled microfluidic devices," Lab on a Chip,
vol. 11, pp. 3049-3052, 2011.
[311] J. Jenson, W. Tuttle, R. Stewart, H. Brechna, and A. Prodell, "Selected cryogenic data
notebook. Brookhaven National Laboratory," ed, 1980.
[312] W. Deng, Fabrication of silicon carbide sintered supports and silicon carbide membranes:
University of Southern California, 2013.
[313] G. Krishna Priya, P. Padmaja, K. G. WARRIER, A. Damodaran, and G. Aruldhas,
"Dehydroxylation and high temperature phase formation in sol-gel boehmite
characterized by Fourier transform infrared spectroscopy," Journal of materials science
letters, vol. 16, pp. 1584-1587, 1997.
[314] R. Al-Oweini and H. El-Rassy, "Synthesis and characterization by FTIR spectroscopy of
silica aerogels prepared using several Si(OR)4 and R′′Si (OR′)3 precursors," Journal of
Molecular Structure, vol. 919, pp. 140-145, 2009.
[315] B. J. Saikia and G. Parthasarathy, "Fourier transform infrared spectroscopic
characterization of kaolinite from Assam and Meghalaya, Northeastern India," Journal of
Modern Physics, vol. 1, p. 206, 2010.
[316] T. Oh, "Correlation between potential barrier and FTIR spectra in SiOC film with the CO
bond of sp3 structure," Bull. Korean Chem. Soc, vol. 30, pp. 467-70, 2009.
[317] Y. Zhou, Organosilicon Chemistry: Special Lectures Presented at the International
Symposium on Organosilicon Chemistry: Elsevier, 2013.
Abstract (if available)
Abstract
The purpose of this study with the subject of biogas Reforming: Conventional and Reactive Separation Processes and the Preparation and Characterization of Related Materials was to determine the feasibility of employing the catalytic steam reforming of biogas to increase its energy content by converting the methane it contains into a hydrogen-rich syngas mixture
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Experimental studies and computer simulation of the preparation of nanoporous silicon-carbide membranes by chemical-vapor infiltration/chemical-vapor deposition techniques
PDF
Fabrication of nanoporous silicon carbide membranes for gas separation applications
PDF
Continuum and pore netwok modeling of preparation of silicon-carbide membranes by chemical-vapor deposition and chemical-vapor infiltration
PDF
Fabrication of silicon carbide sintered supports and silicon carbide membranes
PDF
Novel methods for landfill gas and biogas clean-up
PDF
A study of the application of membrane-based reactive separation to the carbon dioxide methanation
PDF
Process intensification in hydrogen production via membrane-based reactive separations
PDF
On the use of membrane reactors in biomass utilization
PDF
A hybrid adsorbent-membrane reactor (HAMR) system for hydrogen production
PDF
Controlling polymer film patterning, morphologies, and chemistry using vapor deposition
PDF
The use of carbon molecule sieve and Pd membranes for conventional and reactive applications
PDF
Silicon carbide ceramic membranes
PDF
Lab-scale and field-scale study of siloxane contaminants removal from landfill gas
PDF
Understanding structural changes of polymer films during vapor phase deposition
PDF
An integrated 'one-box' process for hydrogen production
PDF
A process-based molecular model of nano-porous silicon carbide membranes
PDF
Methanol synthesis in a contactor-type membrane reactor
PDF
The purification of contaminated air streams via recative separation techniques
PDF
Simultaneous monomer deposition and polymerization at low substrate temperatures for the formation of porous polymer membranes
PDF
Methanol synthesis in the membrane reactor
Asset Metadata
Creator
Dabir, Sasan
(author)
Core Title
Biogas reforming: conventional and reactive separation processes and the preparation and characterization of related materials
School
Viterbi School of Engineering
Degree
Doctor of Philosophy
Degree Program
Chemical Engineering
Publication Date
02/13/2017
Defense Date
01/19/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
biogas reforming,gas separation,initiated chemical vapor deposition,membrane reactors,NOₓ emissions,OAI-PMH Harvest,silicon carbide membrane
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Tsotsis, Theodore (
committee chair
), Gupta, Malancha (
committee member
), Hogen-Esch, Thieo E. (
committee member
)
Creator Email
sasan_dr@yahoo.com,sdabir@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-334797
Unique identifier
UC11258314
Identifier
etd-DabirSasan-5045.pdf (filename),usctheses-c40-334797 (legacy record id)
Legacy Identifier
etd-DabirSasan-5045.pdf
Dmrecord
334797
Document Type
Dissertation
Rights
Dabir, Sasan
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
biogas reforming
gas separation
initiated chemical vapor deposition
membrane reactors
NOₓ emissions
silicon carbide membrane