EPA- 600 / R- 97-093
September 1997
<&EPA Research and
Development
METHANE-STEAM REACTION
OVER NICKEL CATALYSTS
IN THE HYNCL PROCESS
united States
Environmental Protection
Agency
Prepared for
National Risk Management Research Laboratory
Prepared by
National Risk Management
Research Laboratory
Research Triangle Park, NC 27711
-------
KPA-600/R-97-093
September 1997
METHANE-STEAM REACTION OVER NICKEL CATALYSTS
IN THE HYNOL PROCESS
Prepared by:
Yuanji Dong and Jarek Karwowski
Acurex Environmental Corporation
4915 Prospectus Drive
P.O. Box 13109
Research Triangle Park, NC 27709
EPA Contract No.68-D4-0005, Work Assignments No.2-032 and 3-019
Project Officer: Robert H. Borgwardt
U.S. Environmental Protection Agency
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Atmospheric Protection Branch
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, D.C. 20460
-------
FOREWORD
The U, S. Environmental Protection Agency is charged by Congress with pro-
tecting the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions lead-
ing to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA1 s research
program is providing data and technical support for solving environmental pro-
blems today and building a science knowledge base necessary to manage our eco-
logical resources wisely, understand how pollutants affect our health, and pre-
vent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for
investigation of technological and management approaches for reducing risks
from threats to human health and the environment. The focus of the Laboratory's
research program is on methods for the prevention and control of pollution to air,
land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and groundwater; and prevention and
control of indoor air pollution. The goal of this research effort is to catalyze
development and implementation of innovative, cost-effective environmental
technologies; develop scientific and engineering information needed by EPA to
support regulatory and policy decisions; and provide technical support and infor-
mation transfer to ensure effective implementation of environmental regulations
and strategies.
This publication has been produced as part of the Laboratory's strategic long-
term research plan. It is published and made available by EPA's Office of Re-
search and Development to assist the user community and to link researchers
with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
EPA REVIEW NOTICE
This report has been peer and administratively reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information
Service, Springfield, Virginia 22161.
-------
TECHNICAL REPORT DATA ln
(Please read Instructions on the reverse before compl ||| ||| 11|||| || 1II11||| III
1. REPORT NO. 2.
EPA-600/R-97-093
3 III 111 II lllll III III llllllll III
PB98-100480 '
4. TITLE AND SUBTITLE
Me thane-Steam Reaction Over Nickel Catalysts in the
Hynol Process
5. REPORT DATE
September 1997
6. PERFORMING ORGANIZATION CODE
7. AUTHORiS)
Yuanji Dong and Jarek K arwow ski
8. PERFORMING ORGANIZATION REPORT NO.
9. performing organization name and address
Acurex Environmental Corporation
P. 0, Box 13100
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D4-0005, WAS 2-032 and
3-019
12. SPONSORING AGENCY name and address
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13 TYpE OF report AND PERIOD COVERED
Final; 10/95 - 4/97
14, SPONSORING AGENCY CODE
EPA/600/13
15. supplementary NOTES ^PPCD project officer is Robert II. Borgwardt, Mail Drop 63,
919/541-2336.
is.abstract rj-^g rep0rt discusses the reaction of methane-steam over nickel catalysts
in the Hynol process, a process that uses biomass and natural gas as feedstocks to
maximize methanol yields and minimize greenhouse gas emissions. EPA's APPCD
has established a laboratory in which to conduct experiments on the critical reactions
involved in the Hynol process. In this study, an integral fixed-bed reactor was used
to perform kinetic measurements for methane-steam reforming at simulated Hynol
operating conditions. The activity of a commercially available Ni- catalyst was eval-
uated. A kinetic model was developed for quantitatively interpreting the experimental
data. The intrinsic reaction rates at different temperaturejwere measured using
crushed catalyst pellets, resulting in an activation energy of 28 keal/mol. The ef-
fectiveness factor for the commercial catalyst pellets (16 mm in diameter and 10 mm
long) was determined and correlated as a function of reaction temperature. Experi-
mental results indicate that a steam-to-carbon ration of 2.5 is appropriate. The car-
bon monoxide and carbon dioxide in the feed gas were found to be insensitive to the
catalyst performance within the range of the experimental study. The hydrogen in the
feed gas helps catalysts remain in the reducing state and prevents carbon deposition.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.identifiers/open ended terms
c. COSATI Field/Group
Pollution Natural Gas
Catalysis Methane
Nickel Steam
Carbinols
Greenhouse Effect
Biomass
Pollution Control
Stationary Sources
Hynol Process
Methanol
13 B 2 ID
07D
07B
07C
04A
08A.06C
18, distribution statement
Release to Public
19. SECURITY CLASS (ThisReport/
Unclassified
21. NO. OF PAGES
90
20. SECURITY CLASS (Thispagef
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
ABSTRACT
The Hynol process, which uses biomass and natural gas as feedstocks to maximize
methanol yields and minimize greenhouse gas emissions, is under evaluation by APPCD of EPA.
A laboratory has been established to conduct experiments on the critical reactions involved in the
Hynol process. In this study, an integral fixed-bed reactor was used to perform kinetic
measurements for methane-steam reforming at simulated Hynol operating conditions. The
activity of a commercially available Ni-catalyst was evaluated. A kinetic model was developed
for quantitatively interpreting the experimental data. The intrinsic reaction rates at different
temperatures were measured using catalyst powders smaller than 0.1 mm, resulting in an
activation energy of 28 kcal/mol. The effectiveness factor for the commercial catalyst pellets
(16 mm in diameter and 10 mm long) was determined and correlated as a function of reaction
temperature. Experimental results indicate that a steam-to-carbon ratio of 2.5 is appropriate. The
carbon monoxide and carbon dioxide in the feed gas were found to be insensitive to the catalyst
performance within the range of the experimental study. The hydrogen in the feed gas helps
catalysts remain in the reducing state and prevents carbon deposition. The experimental results
showed that the catalyst activity dropped when the hydrogen partial pressure in the feed gas was
below a certain value. The size of a steam reformer suitable for the Hynol demonstration plant
was estimated. The study concludes that the commercial Ni-catalyst can be used for the
methane-steam pyrolysis of the Hynol process.
ii
-------
TABLE OF CONTENTS
Section Page
ABSTRACT . ii
LIST OF TABLES iv
LIST OF FIGURES v
NOTICE vi
ACKNOWLEDGMENTS vi
METRIC CONVERSION FACTORS vii
1.0 INTRODUCTION.... 1
2.0 PREVIOUS KINETIC STUDIES 5
3.0 EXPERIMENTAL...... 10
3.1. Experimental Apparatus 10
3.2. Experimental Procedures 13
3.3. Catalyst Properties 14
3.4. Feed Gas Composition and Flow Rates 16
4.0 KINETIC MODEL 18
5.0 RESULTS AND DISCUSSIONS 22
5.1. Blank Tests ....22
5.2. Catalyst Stabilization 23
5.3. Minimum Steam Ratio 23
5.4. Intrinsic Reaction Rate 25
5.5. Effective Activities of Commercial Catalyst Pellets 31
5.6. Effects of Feed Gas Composition 34
5.7. Demonstration Plant Sizing 37
6.0 QUALITY ASSURANCE 39
7.0 CONCLUSIONS AND RECOMMENDATIONS 44
8.0 REFERENCES 47
iii
-------
LIST OF TABLES
Table Page
1 Activation Energies Reported in Previous Work 49
2 Purities of the Gases Used 49
3 GC and Integrator Operating Conditions 50
4 Catalyst Particle Density 50
5 Catalyst Solid Density 51
6 Catalyst - Specific Surface Areas and Pore Sizes 51
7 Comparison of Catalyst Properties 51
8 Experimental Data Summary 52
9 Blank Test Results (PR-09) 54
10 Temperature Dependence of Catalyst Powders 54
11 Transport Properties of Gas Components 55
12 Effectiveness Factors of Catalyst Pellets 55
13 Data Quality Indicator Goals for Accuracy 56
14 Data Quality of GC Analysis (Gas Cylinder No.5) 57
15 Data Quality of GC Analysis (Gas Cylinder No.6) 58
IV
-------
LIST OF FIGURES
Figure Page
1 Hynol Process Flowsheet ,59
2 Schematic Diagram of the Steam Reforming Reactor 60
3 Flowsheet of the Experimental Equipment 61
4 Typical Printout of the GC Integrator 62
5 Example of SPR Data Sheets 63
6 Hynol Process Simulation Results 68
7 Stabilization of Catalyst Activity 69
8 Effects of Steam Ratio on Catalyst Activity 70
9 Effects of Temperature on Methane Conversion by Catalyst Powders 71
10 Arrhenius Plot for Catalyst Powders 72
11 Activities of Commercially Sized Catalyst Pellets 73
12 Effectiveness Factors of Catalyst Pellets 74
13 Comparison of Calculation Results with Experimental Methane Conversion Data .... 75
14 Effects of the CO Partial Pressure in the Feed Gas on Catalyst Activity 76
15 Effects of the CO, Partial Pressure in the Feed Gas on Catalyst Activity 77
16 Effects of the H2 Partial Pressure in the Feed Gas (R039) 78
17 Effects of the H, Partial Pressure in the Feed Gas (R041) 79
18 Effects of the H2 Partial Pressure in the Feed Gas (R042) 80
19 Constant Observed at Various Methane Feed Rates 81
20 Chauvenet's Criterion for Rejecting a Reading 82
v
-------
NOTICE
This report was submitted in fulfillment of 68-D4-0005. Work Assignments No. 2-032
and 3-019 under the sponsorship of the U.S. Environmental Protection Agency. This report
covers a period from October 1995 to March 1997, and work was completed as of March 1997.
acknowledgments
The help of Kevin Bruce, Project Manager, Acurex Environmental Corporation, with
management support and review of this report is greatly appreciated. Special thanks are
extended to David Proffitt and Bobby Sharpe of Acurex Environmental for design and
construction of the experimental facility.
vi
-------
METRIC CONVERSION FACTORS
Readers more familiar with the metric system may use the following factors to convert
the non-metric units used in this report:
1 in = 2.54 cm
1 ft = 30.48 cm
1 lb = 0.454 kg
1 psi = 6.89 kPa
1 atm = 101.3 kPa
vii
-------
SECTION 1.0
INTRODUCTION
Concerns regarding air quality, global climate change, and the national energy security
impacts of the intensive uses of gasoline in the transportation sector have raised interest in
alternative and renewable fuels.
Methanol has potential as an alternative fuel. It is a chemically simple liquid fuel,
compatible with the existing automotive refueling infrastructure. It can be made from a variety
of domestic feedstocks and is relatively inexpensive to produce. The combustion of methanol in
internal-combustion engines is very efficient and produces fewer toxic emissions than gasoline
(Motor Vehicle Emission Laboratory, 1989). In addition, methanol is the most viable onboard
hydrogen source for fuel cells, which are being considered as a replacement for internal-
combustion engines for road transportation. Fuel cells are more efficient than internal
combustion engines and produce no pollutants.
About 75 percent of commercial methanol production uses natural gas as feedstock. The
process includes steam reforming, methanol synthesis, and purification. Steam reforming
converts natural gas into the synthesis gas, a mixture of hydrogen and carbon monoxide, which is
then synthesized to methanol. Although steam reforming is conducted at 15 to 20 atm and 800 to
900 "C, methanol synthesis operates at 50 atm and 260°C. Both reactions require catalysts to
achieve high methane conversion and methanol yields. The overall reaction is expressed as:
1
-------
CH4 + H20 - CILOH + H2
(I)
As shown in Equation I, steam reforming produces excess hydrogen for methanol production,
which is usually used as combustion fuel.
Synthesis gas can also be produced by the steam gasification of coal or biomass. Because
the amount of C02 absorbed by photosynthesis during the growth of biomass is equal to the
amount released when it is finally used, the use of biomass as feedstock for methanol production
offers advantages in reducing greenhouse gas emissions. However, biomass contains insufficient
hydrogen, and the H2/CO ratio in the synthesis gas produced by biomass steam gasification is not
suitable for methanol synthesis. An additional step, water-gas shift reaction, is usually needed.
In the shift converter, part of the CO in the reforming product is further reacted with steam to
produce more H2. The excess CO, formed is removed before methanol synthesis. If CH0 4O0 6 is
used to represent a typical biomass composition, the overall reaction for this process is:
CH14O06 0biomass) + 0.7H20 - 0.7CH^OH + 0.3CO, (II)
Steam gasification is highly endothermic. The energy required for steam gasification is provided
by burning a portion of biomass with oxygen within the gasifier. The requirements for CO
shifting, C02 removal, and oxygen supply increase capital and operation costs. As a result,
methanol production from biomass has not been cost effective.
2
-------
The idea of using the excess hydrogen from natural gas steam reforming to gasify
biomass for methanol production has led to the invention of a new process: the Hynol process
(Steinberg and Dong, 1994). The Hynol process maximizes methanol yields by using biomass
and natural gas as co-feedstocks and combining biomass gasification and natural gas reforming
as consecutive steps, as illustrated by the following overall reaction:
CHl4O06 +0,9CHa + 1.3///} - 13CHMH (HI)
The process consists of three reaction steps: (1) gasification of biomass at 30 atm and 800°C
with the H ,-rich gas recycled from methanol synthesis, (2) steam reforming of the product gas
with an addition of natural gas feedstock at 25 atm and 950 to 1000°C, and (3) methanol
synthesis of the produced H2 and CO at 30 to 50 atm and 260 °C. The process flowsheet is
presented in Figure 1. Because biomass is gasified by the gas recycled from the methanol
synthesis step, which is enriched with the excess hydrogen, the overall yield of methanol from
biomass is increased. CO shifting and CO , removal, required for the steam gasification process,
are no longer necessary. The exothermic reaction of biomass with hydrogen eliminates the need
for expensive 02 plants or complicated external heating systems for gasification. The integrated
configuration, which uses the heat recovered from steam reforming to preheat the gasification
feed gas and generate steam for methane reforming, increases process thermal efficiency.
The Air Pollution Prevention and Control Division (APPCD) of the National Risk
Management Research Laboratory, U.S. Environmental Protection Agency (EPA), has conducted
3
-------
a theoretical evaluation of process options for the production of transportation fuel from biomass
and has concluded that the Ilynol process represents a promising technology for maximizing fuel
production with minimum greenhouse gas emissions (Borgwardt. 1995). Consequently, the
APPCD established a laboratory to further assess the process feasibility and to carry out
fundamental studies on the reactions that are crucial to the Hynol process. These studies are
intended to provide preliminary answers to questions regarding biomass gasification kinetics at
specific operating conditions of Hynol process and the conditions necessary for steam reforming
the gasification products, and to provide quantitative information to support the design and
operation of a bench-scale demonstration plant.
The study of biomass gasification under the Hynol process conditions was conducted
using a thermobalance reactor (TBR). The TBR testing studied the effects of biomass particle
size, residence time, reaction temperature, and feed gas composition on biomass gasification
rates. The kinetic model developed for the interpretation of experimental data was able to predict
gasification rates and biomass conversion at different operating conditions. The results of that
study have been published (Dong and Cole, 1996; Dong and Borgwardt, 1996).
The steam reforming reaction of the Hynol process is addressed in this report. Using a
fixed-bed reactor, the intrinsic reaction rates and the effectiveness factor for a commercially
available nickel catalyst were evaluated. The minimum steam-to-carbon ratio required to prevent
the carbon deposition on catalysts was determined for Hynol process conditions. The study also
investigated the effects of reaction temperature and feed gas composition on methane conversion
and reaction rates. This report summarizes these experimental results.
4
-------
SECTION 2.0
PREVIOUS KINETIC STUDIES
In conventional steam reforming, natural gas feedstock is desulfurized (usually to less
than 0.5 ppm II;S). mixed with steam, and preheated to between 425 and 550°C. This mixture is
fed to a reactor where it passes through an arrangement of externally fired tubes containing a
nickel catalyst. The process usually operates at 15 to 20 atm and 800 to 900°C, The space
velocities are on the order of 5000 to 8000 h"1. Steam-to-carbon ratios range from 2.5 to 3.5. In
a well-designed reformer, up to 95 percent of the methane is converted; the product gas usually
attains a composition-representative equilibrium concentration at a temperature 10 to 15°C
below the actual exit temperature of the catalyst bed. Nickel is the most widely used catalyst for
steam reforming. Because the activity of a catalyst is closely related to the available surface
area, nickel metal is usually dispersed on an alumina support to maximize the stable nickel
surface area available to the reactants.
The reaction of methane with steam is complex. Various reaction mechanisms have been
proposed. The commonly accepted mechanism for steam reforming is Reaction (IV) followed
by Reaction (V):
CH4 + H20 ¦' CO + 3H2
(IV)
-------
CO + H20 ** C02 + H2 (V)
Akers and Camp (1955) were probably the first researchers to conduct a kinetic study for
steam reforming using a nickel catalyst supported on diatomite. The catalyst pellets were 3,18
mm cylinders. The experimental results obtained at one atm and from 340 to 640 °C showed a
first-order reaction mechanism. The rate of reaction over the whole range of conversion is
directly proportional to the partial pressure of methane,
r =
dt '
k Pa,. W
where r = reaction rates, mol/h-g of catalyst. There was no dependence on other reactants. The
reaction rate constant, k, was reported to be represented by an Arrhenius expression:
E,
k = k exp (- —) (2)
0 RT
where the pre-exponential factor, k0 "=• 127 mol/g-h-atm; the activation energy, EA = 8,778
cal/mol; the gas law constant, R = 1.987 cal/mol-K; and the reaction temperature, T, is in K.
6
-------
However, Bodrov et al. (see Agnelli et al, 1987) found that, at atmospheric pressure, a
certain correlation exists between the partial pressures of CO, H2, and H20. Their influence
varies with temperature. At temperatures below 6000 C, the reaction rates are inhibited by the
presence of hydrogen:
For 400°C < T < 500°C
For 500°C < T < 60O°C
Pen,
r = k (4)
0.5
Ph2
At temperatures above 700 °C, the rate is also influenced by the presence of CO in the feed gas.
For 700 cC < T < 900 °C
Pch4
r = k
Ph2o (5)
1 + a + b pco
An activation energy, EA, of 19,400 cal/mol was obtained. The constants a and b were 0.5 and
2,0 atar1 at 800°C, and 0.2 and 0 atm ' at 900°C. respectively. Bodrov et al. concluded that
7
-------
water-gas shift equilibrium was always established (Ridler and Twigg, 1989). However, Gerhard
and Moe (1965) found that it was not at equilibrium.
Allen et al. (1975) measured the methane conversion at 640°C and pressures up to IB atm
over a Ni-catalyst and correlated it to a time factor (g eat-hr/mol) using a third-degree
polynomial. They proposed a reaction mechanism and concluded that the desorptions of CO and
C02 were controlling steps for steam reforming.
Munster and Grabke (1981) proposed that methane-steam reforming involved methane
decomposition followed by a reaction of adsorbed carbon with steam. The overall reaction was
controlled by the first reaction. Therefore, Equation (1) can be used to express the steam
reforming reaction rate over nickel catalysts. However, Munster and Grabke obtained an
activation energy of 38,000 cal/mol,
Agnelli et al. (1987) studied the dependence of the conversion of the methane-steam
reforming reaction on the partial pressures of methane, hydrogen, and water. They found that the
reaction is first order with respect to methane partial pressure, and that the influence of the
reactants and products (except methane) is small enough to allow the use of a first-order kinetic
equation. An activation energy of 41,650 cal/mol was reported. By comparing the ratios of the
pressures of products to reactants for both Reaction (IV) and Reaction (V), they concluded that
Reaction (V) was close to equilibrium.
Xu and Froment (1989) measured the intrinsic reaction rates of steam reforming at
temperatures up to 575 °C and pressures up to 15 atm. Their experimental results showed that
there were no significant internal diffusion limitations for catalysts sized from 0.17 to 0.25 mm.
8
-------
An activation energy of 57,000 cal/mol was reported. They proposed a more general kinetic
model, which assumed a triangular reaction scheme in methane, carbon monoxide, and carbon
dioxide. Xu and Froment claimed that Reaction (V) was always close to equilibrium at 10 to 15
atm but not at three to five arm.
Some of the reported activation energy data for steam reforming are compared in Table 1.
It can be seen that the data are scattered. By reviewing a number of kinetic studies, Ridler and
Twigg (1989) pointed out that the lack of appreciation of diffusion and heat transfer limitations is
the primary reason for discrepancy in the kinetic results. They compared the results of methane
steam reforming, methane cracking and its exchange with deuterium over nickel films and found
that, in each case, the activation energy (in the absence of different effects) is about 31 kcal/mol.
A literature survey was performed for kinetic studies of steam reforming (Dong, 1994).
The resulting report reviewed the various mechanisms and kinetics of steam reforming reaction,
and carbon deposition on catalysts, as well as possible uses of noble metal catalysts.
Previous studies have been limited mostly to relatively lower pressures and temperatures.
Small sizes of catalysts were used. In the Hynol process, however, steam reforming operates at
25 to 30 atm and 950 to 1000°C, conditions that are beyond the range of the previous studies.
The feed gas into steam reformers is also different: a mixture of gasification product gas with an
addition of natural gas feedstock, which contains 15 to 30 percent hydrogen as well as some CO
and C02. No prior measurements have been reported under such Hynol-specific operating
conditions. This study provides this important information for the Hynol process evaluation and
development.
9
-------
SECTION 3.0
EXPERIMENTAL
3.1 Experimental Apparatus
Figure 2 is a schematic diagram of the fixed-bed reactor used in this study. The reactor
design was based on a double-shell, balanced pressure system, which allows for high pressure
and high temperature operation. The reactor consists of a 2.09 cm inside-diameter reactor tube
and a 30.5-cm diameter stainless-steel, pressure-retaining vessel. The reactor tube was made of
310 stainless steel and was electrically polished to remove the nickel contained on the wall
surface of the reactor tube. The reactor was 120 cm long.
In the pressure-retaining vessel, a separately controlled, three-zone electrical heater
surrounded the reactor tube. Both the top and middle heater zones were 1.4 kW and 30.5 cm
long. The bottom heater zone was 0.7 kW and 15.2 cm long. The top heater zone served for
preheating the feed gas. The reaction temperature was controlled by the other two heating
elements.
The annular volume between the heater and the vessel body was densely filled with
Inswool® bulk insulation fibers to minimize heat loss. During testing, a constant nitrogen flow
of 0.1 SLPM entered the pressure-retaining vessel. It then joined with the process gas exiting
from the condenser to balance pressures between the reactor tube and the pressure-retaining
10
-------
vessel. At 800 °C and a gas flow rate of 10 SLPM in the reactor, the measured reactor pressure
was 25.17 atm when the vessel pressure was held at 25.07 atm, indicating a 0.1 atm of pressure
difference.
A perforated-plate catalyst support was used to position catalysts in the middle section of
the reactor. Catalysts were charged into the reactor through the top plate flange. For those tests
with a single commercially sized catalyst pellet. 45 ml and 65 ml of 3mm~diameter ceramic
Raschig rings were packed under and above the pellet to ensure a uniform gas flow through the
reactor. When crashed catalyst powders were used to measure the intrinsic reaction rates, the
Raschig rings served as a diluent mixed with the catalyst powders to improve the distribution of
heat load in the reactor.
Three K-type thermocouples were inserted from the bottom of the reactor through
separate 1/8-in stainless-steel tubing lances to measure the temperatures at the top, middle, and
bottom of the catalyst bed.
The flowsheet of the experimental equipment used in this study is shown in Figure 3,
The feed gas components — methane, hydrogen, and carbon monoxide — were supplied from
individual gas cylinders. The carbon dioxide was obtained from a custom-blended CG../IL
cylinder containing 30 percent carbon dioxide. The purity of each gas component is presented in
Table 2. The flow rates of these gas components were controlled separately by the mass flow
controllers, and then were blended in an on-line gas mixer to simulate the Hynol steam pvroiysis
conditions.
11
-------
A constant steam flow for the reforming reaction was generated by an electrically heated
steam generator in which distilled water was injected through an IIPLC metering pump and then
vaporized and mixed with the other feed gases. The steam generator was a 4.4-kW split tube
furnace-type heater with a 18-in long, 3-3/41.D. heated chamber. A 20-it long, 25-lum coil of
1/4 -in stainless steel tube was horizontally mounted in the heater to provide sufficient heat
transfer surface area for steam generation. The steam generator was controlled at 380°C.
The gas mixture from the steam generator was preheated to 400°C by the on-line heating
tapes and then was entered into the top of the reactor. In the top reactor heater zone, the gas
mixture was further heated to the required reaction temperature before contacting catalysts.
The product gas exiting the bottom of the reactor was cooled in a 52-in long vertical
water-cooled condenser followed by an ice-bath to remove moisture. Upon exiting the ice bath,
the gas temperature was well below 15°C. The water condensate collected in the 0.5-liter trap
was weighed every 30 minutes during testing to determine the moisture in the product gas.
The offgas, from which moisture has been removed by the condenser, was then
depressurized through a back-pressure regulator and vented to atmosphere. The offgas flow rate
was measured by an on-line dry gas meter.
The offgas composition was analyzed by a Hewlett Packard 5890 Series II gas
chromatograph (GC) equipped with a 30-ft-long HayeSep DB column and a thermal conductivity
detector (TCD). An automatic sampling valve, controlled by the integrator keyboard, took gas
samples every 30 minutes from the sampling port and injected them into the heated GC injection
port. A gas purifier, filled with sodium/calcium sulfate, was placed between the sampling port
12
-------
and the automatic valve to further remove moisture and impurities from gas samples. Helium
was used as a carrier gas to sweep samples into the column. The column separated the samples
into the following compounds: hydrogen, nitrogen, carbon monoxide, methane, and carbon
dioxide, which were detected by the TCD and recorded as peaks on the integrator. The integrator
calculated the composition of each compound based on its peak area and relative response factor.
Table 3 summarizes the operating conditions set for GC and integrator. A typical printout from
the GC integrator is shown in Figure 4.
Pressure and temperature measurement locations are shown in Figure 3.
A personal computer (PC) controlled the system. LabTech control software was used to
display, control, and record the pressure, temperatures, and flow rates. The data were logged
every 30 minutes into a set of specially designed Excel spreadsheets, which provided automatic
calculations for steam feed rates, water condensate rates, offgas flow rates, methane conversion,
as well as material balance checking. An example of the recorded data spreadsheets is provided
in Figure 5.
3.2 Experimental Procedures
An activation process is usually required when fresh catalysts are used. The purposes of
activation are to stabilize supported metal crystals and remove adventitious poison. The
techniques used for catalyst activation depend primarily on the nature of the catalyst, but also on
the process for which the catalyst is to be used. For \'i-catalyst used in methane-steam
reforming, Agnelli et al. (1987) suggested reducing catalysts by a 10 percent gaseous mixture of
Il2 and N2 at reaction temperature for 15 hours. In this study, a commercially recommended
13
-------
activation procedure for Ni-catalyst was employed (Ridler and Twigg, 1989). The procedure
involved passing a reducing gas containing an 8:1 ratio of steam to hydrogen through the catalyst
at 800 °C for six hours. The hydrogen feed rate was set to 1.2 SLPM in all the activation runs.
To start a test run, catalysts with known weights were first loaded into the reactor. The
system was pressurized with 10 SLPM of nitrogen through the catalyst bed and one SLPM of
balance nitrogen through the pressure-retaining vessel. After the system reached the desired
operating pressure, a SNOOP liquid leak detector was used to check for leaks. The reactor was
then heated up at a rate less than 200°C/h.
When the desired reaction temperature was reached, the balance nitrogen flow in the
pressure-retaining vessel was reduced to 0.1 SLPM. The gas flow in the catalyst bed was
switched from nitrogen to the feed gas mixture. The heater elements were carefully controlled to
maintain a consistent temperature at the top, middle, and bottom of the catalyst bed. After
stabilization, the experimental data were recorded at 30-minute intervals.
When the run was complete, the reactor was cooled down at a rate less than 200°C/h.
The system was then depressurized. Careful control of heating and cooling rates prevented
thermal damage to the heating elements and catalyst pellets.
3.3 Catalyst Properties
The commercial Ni-based catalyst used in this study was 16 mm in diameter and 10 mm
long. It was ring shaped with seven 3-mm diameter holes. The catalyst contained 15 percent
nickel, 25 percent magnesium oxide, and 60 percent aluminum oxide. Its melting point was
1400°C.
14
-------
The sphericity, 4>s, of the catalyst pellet was estimated based on the above given
geometric dimensions in the following equation:
4>s = (
surface of sphere
both of same volume
622.9 mm 2
- U.iy
(6)
surface of particle
1586.8 mm 2
An average particle density of 1,75 g/em3 for this type of catalyst pellet was determined
from the weight and volume measurements, as summarized in Table 4. The solid density was
estimated to be 4.65 g/cm3 based on the catalyst ingredients and illustrated in Table 5. The
porosity of the catalyst was then estimated as:
where e = porosity
pP = particle density, g/cm3
ps = solid density, g/cm3
The specific surface area and average pore size of the catalysts were measured with a
Micromeritics Surface Area Analyzer and calculated using the Brunauer-Emmett-Teller (BET)
method. The results for the fresh catalyst and the catalyst after 20 hours of operation at 25 atm
and 1000°C are provided in Table 6.
4.65
(7)
15
-------
Table 7 compares the catalyst properties in this study with those reported in the previous
publications.
3,4 Feed Gas Composition and Flow Rates
The feed gas composition used for the base case studies was chosen to simulate the
results of the Hynol process simulation, which were obtained using ASPEN PLUS as shown in
Figure 6 (Borgwardt, 1994), In this report, the Hynol feed gas always refers to the feed gas
mixture with the following mole ratios of components:
CO c°2 H2
- 0,266; = 0.12; —— = 1.06
TT /-* ZT
L- Jti . w Jri . C1 jfjf A
4 4 4
The steam-to-carbon ratio in the feed gas was a parameter to be investigated. The
nitrogen in the feed gas was neglected in the study because of its inert nature and the small
quantity involved.
The total feed-gas flow rate in the base case study was determined based on the following
considerations: (1) The gas velocity in the reactor under the reaction conditions should be high
enough to minimize the resistance to the gas-film mass transfer; (2) the maximum capacity of the
water-cooling condenser was 50 mL/min; (3) the maximum capacity of the trap for condensate
collection was 0.4 liters in 30 minutes or a rate of 13 mL/min; and (4) the methane feed rate
should be less than 5.6 SLPM to allow a single methane cylinder to supply methane for at least
16
-------
16 hours. As a result, the rates of feed gas components in the base case were set to CH4 = 3.0
SLPM, H2 = 3.181 SLPM, CO = 0.799 SLPM, and C02 = 0.359 SLPM.
An operating pressure of 25 atm was chosen for all the experimental runs based on the
consideration of a designated pressure of 30 atm for biomass gasification and an estimated
pressure drop of 5 atm from the exit of the gasifier to the exit of the steam reformer in the Hynol
process.
17
-------
SECTION 4,0
KINETIC MODEL
Most investigators of methane-steam reforming have agreed on a kinetics of first order
with respect to the partial pressure of methane. A constant gas flow throughout the catalyst bed
was assumed when they integrated the first order reaction rate expression to obtain the reaction
rate constants. The steam feed in the reforming reaction is usually in excess of that required to
suppress carbon formation on catalysts and improve reaction yields. Therefore, the first-order
scheme with respect to the methane partial pressure is considered appropriate and is adopted in
the kinetic model for the Hynol steam pyrolysis reaction. However, the assumption of constant
gas flow is valid only if a small amount of methane converted across the catalyst bed. As
Reaction (IV) shows, each mole of methane reacting with one mole of steam produces one mole
of carbon monoxide and three moles of hydrogen, resulting in two moles of net increase in the
total gas flow. In other words, the total mole flow rate increases as the process gas passes
through the catalyst bed. This increase in total gas flow rate can be significant if the conversion
is high. For this reason, the increase in total flow rate due to reaction was taken into account in
developing a quantitative kinetic model of the Hynol steam reforming reaction suitable for an
engineering application.
18
-------
If the reaction rate, r, is defined as the moles of methane converted by one gram of
catalyst per hour, the first order reaction rate can be expressed by:
dF F
1 cm , D , cm D /ox
r = - = k PrH, = k F (8)
i-, n h pp \ s
where FCH4 = methane flow rate, mol/h
k = rate constant for Reaction (IV), mol/g-h-atm
P = reactor pressure, atm
PCH4= methane partial pressure, atm
W = catalyst weight, g
Ft = total process gas rate, mol/h
It is convenient to express the reaction rate in terms of methane conversion. The methane
conversion is a measure of how far the steam reforming reaction has progressed and is defined as
the methane reacted divided by the methane fed to the reactor:
fch4 - FCm
Xc*4 = 0 (9)
4h4
where F°CH4 = methane mole feed rate, mol/h
Fum = methane rate after reaction, mol/h
19
-------
In the experiments, FCII4 at the exit of the reactor can be calculated from the ofleas flow
rate measured by the dry gas meter and the methane mole fraction in the offgas obtained from the
GC analysis.
Because the total gas flow rate, Fx, in Equation (8) increases as the reaction proceeds, it
should be a function of XCH4. From the stoichiometric relationship of Reaction (IV) and
Equation (9), the following equation can be obtained:
where FT°= initial total feed rate, moi/hr
Fcih''" initial methane feed rate, mol/hr
Substituting Equations (9) and (10) into Equation (8) results in a reaction rate expression in terms
of methane conversion as:
ft = f; + 2F"cm xcha
(10)
^ p
dW
(11)
Because the methane conversion is zero at the reactor inlet, the initial condition is:
at W = 0
(12)
20
-------
Integrating Equation (11) with the initial condition gives the methane conversion, as a
function of catalyst weight, W, by
X.
Fr + 2 Fcm k P W
(13)
CH4
CH 4
The above implicit equation must be solved iteratively. If the reaction rate constant is known,
the methane conversion for a given amount of catalyst can be calculated from Equation (13).
The equation can be rearranged to allow for the determination of the reaction rate constants from
methane conversion measurements:
Because the catalyst weight, system pressure, and total and methane feed rates are known, the
rate constant can be calculated from the methane conversion observed from the experiment.
The primary focus of this kinetic study was methane conversion over catalysts because
the water-gas shift reaction is faster than the methane-steam reaction and is close to equilibrium
at high pressures and temperatures. The overall rate of methane-steam reforming is therefore
considered to be controlled by Reaction (IV).
k
P W
[ 2 Fgff4 (F? + 2 Fciii) 1^(1 ^ch4^ ]
(14)
21
-------
SECTION 5.0
RESULTS AND DISCUSSIONS
A number of test runs have been made under the conditions of the steam reforming step
of the Hynol process. Table 8 summarizes all the data obtained in those runs and addresses all of
the variables considered potentially influential in the conversion of methane over catalysts at
Hynol conditions. The results from these investigations are separately discussed in detail in the
following sections.
5.1 Blank Tests
In order to ensure that the observed reaction results truly represented the catalyst
per formance, several blank tests were conducted to examine the catalytic effects of the reactor
tube wall on the methane conversion. In the tests, methane and steam were introduced into an
empty reactor. The feed gas was a mixture of 3.14 SLPM of methane and 5.13 SLPM of
nitrogen. The steam-to-carbon ratio was 3.3. The reactor operated at 25 atm and 950°C. The
typical experimental results are shown in Table 9. The difference between the methane flows
into and out of the reactor averaged 0.64 percent, indicating that the methane conversion is
negligible in the empty reactor tube.
22
-------
5.2 Catalyst Stabilization
It is important that catalysts be stabilized before kinetic data can be collected. In an early
kinetic study, Akers and Camp (1955) found that the activity of Ni-catalyst was constant; no
change in the product gas composition was observed over a period of 6.5 hours. However, the
deactivation of Ni-catalysts in the initial stage of use as a result of sintering was reported. The
reported times for catalyst stabilization differ significantly, ranging from 70 hours (Xu and
Froment, 1989) to less than 30 minutes (Rostrup-Nielsen, 1973). It is expected that the system
operating conditions will affect the catalyst stabilization period. Higher pressures, temperatures,
and steam-to-carbon ratios promote catalyst sintering and, thus, increase catalyst deactivation
rates. The stabilization time must be determined experimentally.
A test was conducted to examine the stabilization time of the catalyst used in this study.
The test was performed at 25 atm and 10G0°C with 0.8 g of crushed Ni-catalyst powders. The
Hynol feed gas composition was used. The steam-to-carbon ratio in the feed gas was 2.07.
Methane conversion was measured every 30 minutes over the entire operation period of 27 hours.
The observed variation in methane conversion with reaction time is presented in Figure 7. Under
Hynol operating conditions, the activity of nickel catalysts decreased rapidly in the first five
hours, after which a steady-state condition was reached. Thus, a minimum stabilization time of
four to five hours is considered necessary before kinetic data collection can be started.
5.3 Minimum Steam Ratio
Carbon deposition on catalysts reduces the catalyst activity and, therefore, must be
prevented. Carbon can be formed in the reactor in two basic ways. Carbon is formed in the
23
-------
catalyst pores when the following two reactions are thermodynamically favorable for a given
mixture of CO, CO,. H,, CH4 and 11,0
2 CO - C02 + C
(VI)
CO + h2 = h2o + c
(VII)
Carbon can also be formed as a result of thermal cracking by:
CH4 = 2 H2 + C
(VIII)
Carbon resulting from hydrocarbon cracking does not usually form in the inner pores of the
catalyst, but rather as soot-like deposits on the outside surface of the catalyst.
To avoid carbon formation, a high steam-to-carbon ratio in the feed gas must be
employed. The steam ratio in conventional steam reforming processes usually ranges from 2.5 to
The feed gas contains hydrogen in the Hynol steam pyrolysis reaction. This hydrogen
assists in preventing carbon formation and in maintaining the catalyst in a reducing state, A
relatively lower steam ratio requirement is thus possible for the Hynol process. Tests were
conducted to estimate the appropriate steam ratio under the Hynol operating conditions. The
tests operated at 25 atm and 950 °C with two g of Ni-catalyst powders. Methane conversion was
continuously measured at a steam ratio of 1.5 for 30 hours. No methane conversion loss was
3.5.
24
-------
observed during the test, but carbon deposition was found at the exit of the reactor. When the
steam ratio was greater than two, carbon formation was found to be eliminated. Therefore, the
minimum steam ratio is considered to be two at the test conditions.
The independence of the reaction rate upon the steam ratio above the minimum can be
illustrated by comparing the experimental results obtained at different steam ratios for a single
commercially sized catalyst pellet, as shown in Figure 8. It was noted from the figure that, at the
same reaction temperature, the reaction rate is zero order with respect to the steam ratio. Thus, a
steam ratio considerably higher than the minimum is undesirable because a higher steam ratio
requires more heat load. The appropriate range of steam ratios for the Hynol steam pyrolysis is
considered to be from 2 to 2.5.
5.4 Intrinsic Reaction Rate
Gaseous chemical reactions over a solid catalyst system involve mechanistic steps such as
external mass transfer, pore diffusion, chemisorption, and intrinsic reaction. The overall reaction
rate is controlled by the slowest mechanistic step. The intrinsic kinetics means that the measured
rate is free from pore diffusional influence or any other mass transfer influence. Therefore, the
intrinsic reaction rate must be measured under conditions showing negligible restrictions for
mass transfer and pore diffusion.
Pore diffusion resistance can be minimized by reducing catalyst size. Agnelli et al.
(1987) indicated that catalyst pellets should be crushed down to a maximum size between 0.30
and 0.42 mm to avoid internal diffusion effects. Xu and Froment (1989) suggested this critical
catalyst size should be between 0.17 and 0.25 mm. In this study, intrinsic kinetics was measured
25
-------
using powders smaller than 0,1 mm, crushed from commercial catalyst pellets. To ensure
representativeness, about 500 grams of catalyst pellets was ground and screened to obtain testing
samples.
The intrinsic reaction rates were measured at reaction temperatures ranging from 900 to
1000 °C. Two series of runs were conducted. In each test, a one-gram sample of Ni-catalyst
powders was charged into the reactor in which, ceramic rings act as diluents to improve heat
distribution. The Ilvnol feed gas composition then was simulated. The system pressure was kept
at 25 atm. The methane conversions observed at different temperatures were plotted as a
function of reaction temperature as shown in Figure 9. The plot displays a close agreement
between the conversion results obtained from the two samples of catalyst powders, indicating a
good repeatability of the measurements. The results also showed that methane conversion in
steam reforming increases with reaction temperature. The reaction rate constants at different
reaction temperatures were calculated from the conversion data using the kinetic model, Equation
(14), and are summarized in Table 10. The activation energy was then determined by Arrhenius
law. The reaction rate constants were plotted against the reciprocal of the absolute temperature
in Figure 10. The pre-exponential factor, k0, and the activation energy, EA, thus obtained were
65,552 mol/g-h-atm and 27.97 kcal/mol. The activation energy is close to the previously
reported 31 kcal/mol measured in the absence of diffusion effects (Ridler and Twigg, 1989).
The influence of the external gas film resistance on the intrinsic kinetics under the actual
experimental conditions was theoretically evaluated. The rate of methane transport from the bulk
gas phase to the catalyst external surface was calculated by:
-------
^CH4 kg Ap (PCH4 g P CH4,t)
(15)
where NCIW = moles of methane transferred from gas to solid, mole/s
Ap = catalyst external surface, cm2
kg = mass transfer coefficient between gas and solid, mol/s-cm2-atm
PCH4jg = methane partial pressure in the bulk phase, atm
PcH4,e = methane partial pressure at the catalyst external surface, atm
in which the mass transfer coefficient, kg, was estimated by the following correlations for a
packed catalyst bed at Re <190 (Froment and Bischoff, 1979):
1.66 Re ~051
(16)
with
(17)
Sc
P D
rm m
(18)
(19)
27
-------
where Sc = Schmidt dimensionless number
Re = Reynolds dimensionless number
Ft = Total gas mole flow, mol/s
P = System pressure, atm
jim = viscosity of gas mixture, g/cm-s
prn = density of gas mixture, g/em3
Dm = diffusivity of methane through the gas mixture, em2/s
U = gas velocity, cm/s
dp = catalyst diameter, cm
At = cross-sectional area of the reactor, cm2
Dt = diameter of the reactor, cm
The transport properties of the feed gas mixture were calculated from individual gas properties
using the methods discussed by Froment and Bisehoff (1979) as follows:
The density of each pure gas component, pi; under the reaction conditions was calculated
by:
- - 273*p (20)
22.4
-------
To calculate the viscosity of the gas mixture, the viscosities of pure gases at the critical
conditions, nCj, were first estimated by Equation (22) (Bird et aL 1960):
ne, - 7JM/^rcf (22)
where Pc>; and TCji are critical pressure and temperature of the i-th gas component. The viscosities
of the pure gases under the operating conditions were then calculated from:
V- = Vr (23)
where \xr was obtained as a function of Pr = P/Pc and Tr = T/Tc (Bird et aL, 1960). The viscosity
of the gas mixture was then obtained by:
" Pal
-- E
mix
/ = 1
(24)
in which
i u 1/2 m M m 2
$ = _L (i + [1 + A {-I) ] (25)
# Vj Pj M•
i
29
-------
The diffusivities of methane in other gas components at low pressure were calculated based on
Equation (26) (Bird et al., 1960):
D
cm-i
a(
) (P ri!AP )m(T -fn£(
' x c,CH4 c,t' K c,CHA c,i> v
5/12,
T
c.CHA cj
i
i 1/2
+_L>
MCH4 ^i
(26)
in which the constants a and b were given as:
for nonpolar gas-pairs,
a = 2.745xl0"4 and b-1.823
for H20 with a nonpolar gas,
a = 3.640xlO"4 and b = 2.334
The chart (Bird et al.. 1960) was then used to correct the diffusivities to the operating conditions.
The diffusivity of methane through the gas mixture was calculated using the following equation
provided by JFroment and Bischoff (1979):
P +2 P
_ T cm ___
m i i ii (27)
(P +3 P )+ - (P +P H P + (P -P 1 v'
n V H2 CH4' n v1 CO 1 CHA} ~ 1 C02 n ^J/20 f CHA'
cha -hi u cm-co ucm-coi u cm-mo
By substituting known data (DT - 2.667 cm, P - 25 atm, T = 1000°C, PCH4 = 4.79 atm, PH2 = 5.08
atm, Pco = 1.27 atm, Pco, = 0.57 atm, and PH20 = 13.29 atm) into the above equations, the
transport properties were calculated and are presented in Table 11.
30
-------
From these estimated transport properties, the Schmidt number and Reynolds number
were calculated to be 0.36 and 0,9. The film mass transfer coefficient, kg, of 0.0001 mol/s-em2-
atm for 0.1 mm diameter catalysts was then obtained by Equation (16).
The methane transfer rate through the film of the catalyst was then calculated by Equation
(15). If the mass transfer was a controlling step in this equation, the methane partial pressure at
the catalyst surface was taken to be zero. The rate of methane transport to one gram of catalyst
powders was
6W 6x1
NCB4 = KAPCCH4 = kg- ^,4-0.0001* *4-™ =0-164molfs (2g)
g cm4 gdpxpp 0.01x1.748 v '
It was then compared with the initial intrinsic rate at 1000°C, which was calculated as
J?
N - k —C—PW = 0.000283*—-—><25xl = 0.00135 ma lis (29)
'' F.j 15.67
It can be seen that, in this case, the overall reaction was limited by chemical reaction. Therefore,
the measurements are considered to be representative of intrinsic kinetics.
5.5 Effective Activities of Commercial Catalyst Pellets
Although catalyst powders can minimize pore diffusion restriction and provide high
catalytic activity, large pellets are always used in commercial reactors in order to reduce pressure
drops across the catalyst bed. It is thus important to know the effective activity of the
commercial-size catalyst pellets used in practical steam reformers. In this study, the overall
31
-------
reaction rates over commercial Ni-catalyst pellets were measured. Two series of tests were
carried out at 25 atm. One pellet of catalyst was charged for each series of tests, in which the
methane conversion was measured at various temperatures ranging from 900 to 1000°C. The
reaction rate constants obtained in the two series of tests increased with reaction temperature as
shown in Figure 11. This figure shows a difference in the reaction rate constants obtained from
the two catalyst pellets, indicating the possible existence of a variation in catalytic activity
among the pellets. The figure also shows that catalyst pellets had substantially lower reaction
rate constants than did powders, a result of the strong influence of pore diffusion within the large
pellets. The importance of the pore diffusion in the overall reaction rate can be measured by the
catalyst effectiveness factor (Froment and Bischoff, 1979), which is the ratio of the reaction rate
observed with pellets to the intrinsic reaction rate as expressed by:
(Observed reaction rate) _ kois
y| - —— - — - - VjUj
(Intrinsic reaction rate) k
The overall reaction rate of a catalyst pellet is thus equal to the intrinsic reaction rate multiplied
by the effectiveness factor at the same operating conditons.
The effectiveness factors for the Ni-catalyst pellets used in this study were estimated
from the experimental results of the pellets and the intrinsic kinetics discussed in Section 5.4.
Considering the activity deviation among the tested catalyst pellets, a statistical analysis was
conducted to calculate the effectiveness factors from observed reaction rate constants. The
averaged effectiveness factors at different reaction temperatures were plotted with error bars in
-------
Figure 12. The statistical analysis results were also summarized in Table 12, In the table, the
maximum and minimum values of the effectiveness factor are obtained with a confidence
interval of 95 percent. The results reflect a trend in which the catalyst effectiveness factor
decreased as reaction temperature increased. The chemical reaction gets faster at a higher
reaction temperature so that the restriction of pore diffusion becomes more dominant. The
catalyst effectiveness factors for the Ni-catalyst pellets are well within the range of 0.01 to 0,3
for a similar commercial catalyst reported by Ridler and Twigg (1989). The effectiveness factor
for the catalyst pellets used in this study was con-elated with reaction temperature in the
following expression:
for900°C
-------
The calculation results of the above equation were compared with the experimental data as
shown in Figure 13. Considering the difference in activity among the catalyst pellets used in the
tests, the agreement between the prediction and observation is sufficiently good.
The mechanical strength of the commercial Ni-catalyst used was examined. A single
3.431 -g catalyst pellet was loaded into the reactor. After 50 hours of testing under Hynol process
operating conditions (25 atm and 1000°C), the catalyst pellet was discharged for physical
examination. The pellet weighed 3.423 g after reaction and retained its original shape. No
cracking or any other damage was observed, indicating that the catalyst is applicable to an
operation temperature up to 1000°C. However, as mentioned previously, it is critical that the
heating and cooling rate of the reactor be controlled below 200°C/h to prevent the damage of
both catalyst and furnace caused by thermal shock.
5.6 Effects of Feed Gas Composition
In the Hynol process, the gas composition that is fed to a steam reformer occasionally
may vary slightly. This variation is caused by a fluctuation in the composition of biomass and
natural gas feedstocks or by the performance of the preceeding biomass gasifier. Thus, the
sensitivity of catalyst activity to these potential fluctuations must be assessed. To do so, the
dependence of the extent of methane conversion upon the partial pressures of carbon monoxide,
carbon dioxide, and hydrogen in the feed gas was investigated.
The effects of the CO partial pressure in the feed gas were investigated by varying the CO
feed rate while maintaining the other gas feed rates at the base study levels. The steam feed rate
was fixed at 10.4 SLPM. Two series of runs (R036 and R038) were conducted with a single
34
-------
pellet of Ni-catalyst. Test R036 was made at a temperature of 900°C and Test R038 was at
1000° C, In Test R038, the CO partial pressure in the feed gas was varied at four levels: 2.15,
1.13, 0.57, and zero atm, while the rates of other feed gases including steam were held constant.
Two levels of the CO partial pressure, 0.73 and zero atm, were investigated In Test R036. The
methane conversion was measured at these different CO pressure levels and was used to
calculate the corresponding reaction rate constants. The results are presented in Figure 14 as a
function of the CO partial pressure. It is noted that the reacton rate is nearly independent of the
CO content of the feed gas as CO's partial pressure varies from zero to 2.2 atm. Therefore, it is
concluded that the rate of the methane-steam reaction is unaffected by the CO content of the feed
gas within the range of experimental conditions used in this study.
The effects on catalyst activity of the CO , in the feed gas were investigated at two
reaction temperatures: 1000DC and 900°C. At 1000 :'C, the reaction rate at the Hynol operating
conditions was compared to the results achieved when the feed gas contained no CO,. A zero-
C02 condition was examined in two ways. In the first, steam was used as a make up gas to
maintain the partial pressure constants of other gas components. In the second, nitrogen was
used as a make-up gas. The reaction rate constants obtained under these two different CO,
conditions, compared in Figure 15, show no significant change in catalyst activity when the C02
feed was cut off completely. In the tests at 900 °C, the C02 feed rate was varied at three different
levels of partial pressure: 0.96, 0.48, and zero atm. The results at 900 °C shown in Figure 15 also
indicated that the reaction rate is nearly constant as the CO, partial pressure increased from zero
35
-------
to one atm. Therefore, it can be concluded that, as with carbon monoxide, changes in the C02
composition in the feed gas have no significant effects on catalyst activity.
The dependence of catalyst activity on the hydrogen in the feed gas is complicated.
Three series of tests (R039. R041, and R042) were conducted by varying the hydrogen feed rate
in different sequences at constant flow rates of the other feed gases to obtain the hydrogen partial
pressure between 1.7 to 5.5 atm. All three tests showed a sharp drop in catalyst activity when the
hydrogen partial pressure was below a certain level. Test R039 was made at 950°C. The change
in methane conversion with the hydrogen partial pressure over time is plotted in Figure 16. The
hydrogen partial pressure levels for each period of time are noted on the top of the conversion
curve. It is evident that the deactivation occurs when the hydrogen partial pressure was reduced
from 4.5 to 1.75 atm. In Test R041. the hydrogen feed rate varied periodically over this same
broad range. As shown in Figure 17, the catalyst activity resumed when the hydrogen partial
pressure was set back from a low level. No significant effects on catalyst activity were observed
when the hydrogen partial pressure was above the critical level. In Test R042, the hydrogen feed
rate was reduced from the Hynol condition to allow the reactor to operate at low hydrogen partial
pressure for more than 10 hours. The results presented in Figure 18 show that the catalyst
activity drops as the hydrogen feed rate is reduced but it cannot be recovered simply by resetting
the hydrogen feed back to a high level, indicating the possible influence of the time period under
low hydrogen partial pressures. The reason for substantial drop in methane conversion when
increasing the hydrogen partial pressure from 1.91 to 5.81 atm is not clear and needs further
investigation. The deactivation at a low hydrogen feed rate is caused by both carbon deposition
36
-------
on catalyst surfaces and an insufficient reducing environment for nickel metal. The critical level
of hydrogen partial pressure in the feed gas in the above experiments was around three to four
atm. but is expected to be a function of other experimental variables. The limited results
available in this type of investigation make a quantitative criterion difficult to determine.
Equation (13) indicates that the methane feed rate affects the methane conversion through
both the time factor, W/FCH4°, and the initial methane composition in the feed gas, FCH4"/FT. The
increase in methane feed rate increases the methane partial pressure in the feed gas. resulting in
an increased reaction rate, but a decrease in the residence time of the process gas in contact with
catalysts. The methane conversion is thus an overall result of these two opposite effects. If
Equation (14) holds true, the reaction rate constant calulated from the methane conversion results
obtained from the test runs with different methane feed rates should be constant. Thus, two
series of tests (R045 and R046) were conducted at various methane feed rates to verify the
kinetic model developed in this study. The methane feed rate in the tests was varied from three
to one SLPM, while the feed rates of the other gas components were kept constant. The reaction
temperature in the tests was 950°C. The methane conversion was measured and used to
calculate the reaction rate constants. The results in Figure 19 show that a constant kob is obtained
in spite of the change in methane feed rate, indicating that the kinetic model is valid.
5,7 Demonstration Plant Sizing
The reaction rate of the Ni-catalyst pellets measured in this study can be used to
determine the size of the steam reformer for the Hynol demonstration plant. The design
37
-------
specification of the demonstration plant (Acurex Environmental, 1995) provides the following
data for the gas feed rates:
H2 = 1860 mol/h; CO = 650 mol/h; C02 = 370 mol/h; CH4 = 1960 mol/h
For H20/C = 2,5, the water feed rate is 7450 mol/h, and the initial total gas flow rate is 12290
mol/h. The reactor was assumed to operate at 25 atm and 950 °C and reach 95 percent methane
conversion. The catalyst effectiveness factor at this condition is 0.117 from Equation (31). By
substituting all these data into Equation (32), the required catalyst weight was calculated as:
W = [ 2F*,,XrH, (Fr +2F.//4) ln(l X= 20 kg
£CCC, , 27970 Cm cm T Cm cm (33)
655S2exp( )Pt] „
1.9877*
Because the catalyst particle density is 1750 kg/m3, the catalyst bed volume is calculated to be 23
liters by assuming the voidage is 0.5. The reactor tube used in industrial steam reformers usually
has an inside diameter of 0.07 m and a length of 8 to 10 m. Therefore, the volume of a single
reactor tube is large enough for the demonstration steam reformer.
38
-------
SECTION 6.0
QUALITY ASSURANCE
The critical parameters involved in this study include catalyst weight, feed rates of gas
components, water feed rate, system pressure, reaction temperature, water condensate rate, and
offgas flow rate and composition. These parameters, as well as their associated data quality
indicator goals for accuracy, are listed in Table 13. This section discusses the quality controls of
these measurements.
Catalyst weight was used to calculate the time factor, W/FCH4, which was a parameter
affecting methane conversion. The amount of catalyst to be charged into the reactor was
weighed by an analytical balance (Mettler AT200). The balance was equipped with built-in
calibration weights. According to its manufacturer, the reproducibility is 0.04 mg when the
weight of a sample is less than 50 g. This is considered to be sufficiently accurate for catalyst
weight measurements.
Carbon and total material balance calculations require accurate measurements of the gas
flow rates. The flow rates of individual feed gas components were controlled by Brooks 5850E
Mass Flow Controllers (MFC). The accuracy of the MFCs was ± two percent. Because no
reliable heat capacity data at high pressures are available for these gas components, each MFC
was calibrated by a Gilibrator bubble flow meter using the real working gas under the actual
39
-------
operating pressure. The accuracy of the Gilibrator bubble flow meter is 0.5 percent. The meter
was installed downstream of the SPR back pressure regulator. The pressure and temperature of
the bubble meter were recorded and used to correct the readings to the standard conditions (one
atm and 25 °C). MFC measurements was estimated to be accurate to ±2.5 percent.
The water feed rate was controlled by the 11PLC metering pump and measured by a
graduated glass tube installed at the inlet of the pump. To measure the feed rate, the water flow
was switched from the water reservoir to the graduated tube through a three-way valve. The time
needed to remove 10 ml water from the graduated tube was then measured. The estimated
accuracy for water flow rate measurement was ± two percent.
The system pressure was controlled by a Grove Mity-Mite Back Pressure Regulator. The
pressure in the pressure vessel was measured by a Foxboro/ICT pressure transducer (Model No.
1124-155-A52-A), which was calibrated using a precise pressure gage (Ashcraft, Model No. 45-
1082 SX A2-02B-0-1500). The system pressure was maintained within ± 0.2 atm of nominal
values.
Three K-type thermocouples, one each at the top, middle, and bottom of the catalyst bed,
were used to measure reactor temperature. This temperature was controlled by adjusting the
temperature set points of the reactor heating elements. In all the tests, the reaction temperature
was controlled ± 5°C of the desired values.
The gas temperature at the exit of the ice-bath was well below 15 °C. At that temperature,
the water vapor pressure is 0.0168 atm, and the saturated moisture in the offgas at 25 atm should
be less than 0.07 mol percent
40
-------
The water condensate collected in the trap was discharged and weighed every 30 minutes.
A stop watch was used to determine the collection time. The condensate rate thus was equal to
the weight of water condensate divided by the collection time. The estimated accuracy of
condensate rate measurement was ± 2 percent.
The offgas flow rate was measured by the diy gas meter, which was calibrated by a wet
test meter. The accuracy of the dry gas meter was ±1.5 percent.
The GC was calibrated using six standard cylinders. The response factor for each gas
component was obtained from a linear regression of the peak areas with known gas
compositions. A one-point check was conducted daily before testing to make sure the GC was
working properly. Tables 14 and 15 illustrate the data quality of the GC analysis. In each table,
the repeated GC results are compared with the known compositions of a standard gas cylinder.
The tables include the uncertainties and relative uncertainties calculated from the repeated
measurements using Student's t test at a confidence level of 99 percent. The biggest uncertainty
in the tables is 1.85 percent for hydrogen measurement, indicating the repeatability of all
measurements was quite good. However, it can be seen from the tables that the mean of the
measurements for each gas component differs slightly from the actual composition of the
standard gas cylinders. These systematical deviations were corrected by the correction factors
shown in the tables, which were equal to the known composition divided by the mean of the
measurements.
After catalyst activity reached steady state in testing, the data logging usually continued
for a few hours to obtain a set of methane conversion data. In most cases, the data were quite
-------
stable. However, the observed data occasionally fluctuated. In these eases, a Chauvenet
criterion was used to evaluate the data quality and reject outliers (Holman and Gajda, 1984).
Based oil the Chauvenet criterion, any data point having a deviation greater than a maximum
acceptable value, expressed as a ratio of the deviation to the standard deviation, is considered a
false reading and should be eliminated from the data. The Chauvenet criterion is presented in
Figure 20 as a function of the total available data points. An example of this type of data
checking can be found in Figure 5. Page 4 of Figure 5 showed a standard deviation of 1.88,
calculated from all the methane conversion data. The Chauvenet criterion was 1.39, resulting in
a maximum acceptable deviation of 2.6. The deviation from the average of each data point was
listed in column "d". The data points that should be taken out were marked by an "e" on their
right sides. Page 5 of Figure 5 showed the results after the bad points were eliminated. It can be
seen that the standard deviation improved from 1.88 to 0.65 after removing the outliers.
The overall quality of the experimental data in this study was further evaluated by
cheeking carbon and total material balances between the inlet and outlet of the SPR. The
calculations were performed using the following equations:
for carbon balance,
G + G + G - G (Y + Y ~ Y ")
CI 14 "CO C02 OUT ^ CH4 1 CO 1 C02>
and for total mass balance
42
-------
2>, + Fmo ¦ ~ (1 «.04I-cm-28.01 yco +44.01 Ycol*2.0167„) ~ fFm0 (35)
where G, = Mass feed rate of CH,, CO, CO„ and H2, g/min
FH20 = Water feed rate, g/min
Fouf- Flow rate read from the dry gas meter, 1/min
Yj = Composition from DC analysis
WH20= Water condensate rate, g/min
The errors in the carbon and total material balances for each experimental run are
included in Table 8. It shows that the average error was 1.0 percent for carbon balances and was
1.2 percent for total material balances. The maximum error was 4.7 percent for carbon balances
and 6.4 percent for total material balances.
43
-------
SECTION 7.0
CONCLUSIONS AND RECOMMENDATIONS
This study has successfully demonstrated that the Ni-based catalysts commercially
available for conventional steam reforming can be used for the Hynol process to steam reform
the biomass gasification product gas with the addition of natural gas feedstock to make synthesis
gas for methanol production.
The catalyst has good mechanical strength and allows for the operation at temperatures
up to 1000°C.
The hydrogen in the Hynol feed gas assists in preventing carbon formation and maintains
the catalyst in the reducing state. An appropriate steam-to-carbon ratio of 2 to 2.5 is
recommended for Hynol steam reforming.
Within the range of this study, the carbon monoxide and carbon dioxide in the feed gas
have negligible effects on the catalyst activity.
The methane-steam reaction can be well described by a first-order kinetics. The reaction
rate is directly proportional to the methane partial pressure. By taking account of the increase in
the total flow rate as the reaction proceeds, the intrinsic reaction rate can be expressed by
Equation (11).
44
-------
The intrinsic reaction rate increases with temperature as described by the Arrhenius law.
The prc-exponcntial factor in the rate constant expression is 65552 mol/g-h-atm. The activation
energy is 27.97 kcal/mol, which is close to the published data that were measured in the absence
of diffusion effects.
The pore diffusion in commercial-size catalyst pellets strongly reduced the catalyst
activity. This effect can be quantitatively expressed by the effectiveness factor defined in
Equation (30). The effectiveness factor of the Ni-catalyst pellets is a function of reaction
temperature and has been correlated by Equation (31). The calculation of the methane
conversion over a known weight of catalyst pellets is provided by Equation (32).
A single reactor tube 0.7 m in diameter and 8 m long can be used for the steam pyrolysis
reactor in the Hynol demonstration plant.
The experimental equipment used in this study has been demonstrated to be generally
suitable for catalyst kinetic studies at high pressures and temperatures. However, the use of more
durable heavy-duty heating elements for the test reactor is recommended to avoid unnecessary
shutdowns.
Noble metal catalysts, such as rhodium, ruthenium, platinum, and palladium, are tolerable
to sulfur poison and are also much more reactive than Ni-based catalysts (Rostrup-Nielsen,
1975). The use of noble metal catalysts for steam reforming in the Hynol process can ease the
requirement to clean up the gasification product gas before steam reforming. Sulfur removal
from the process gas then can be done before it enters the methanol synthesis reactor where the
gas has been cooled. A two-stage technology particularly suitable for removing large amounts of
45
-------
hydrogen sulfide can be used (Carnell, 1989). The two-stage technology consists of a washing
stage followed by a reaction with zinc oxide. Using more active noble metal catalysts allows the
use of smaller reactors and, consequently, reduces the capital cost. Therefore, the evaluation of
the activity of noble metal catalysts under the I lynol conditions is suggested.
46
-------
SECTION 8.0
REFERENCES
Aeurcx Environmental Corporation (1995): Hvdroearb/Hynol Process Demonstration. Final
Report A021-1-95, Mountain View, CA,
Agnelli, ME., Ponzi, E.N., and Yeramian, A.A. (1987): Catalytic Deactivation on Methane
Steam Reforming Catalysts. 2. Kinetic Study. Ind. Eng. Chem. Res. 26, (8), 1707.
Akers, W.W. and Camp, D.P. (1955): Kinetics of the Methane-Steam Reaction. AlChE Journal,
1,(4), 471.
Allen, D.W., Gerhard, E.R., and Likins, M.R. (1975): Kinetics of the Methane-Steam Reaction.
Ind. Eng. Chem. Process Des. Dev., 14, (3), 256.
Bird, R.B., Stewart, W.E., and Lightfoot, E.N. (1960): Transport Phenomena. John Wiley &
Sons, New York.
Borgwardt, R.H. (1994): Laboratory Simulation Tests of Hvnol Flowsheet D. Memorandum of
December 12.
Borgwardt, R.I I. (1995): The Hvnol Process. Presented at Symposium on Greenhouse Gas
Emissions and Mitigation Research, Washington, D.C., June 27-29.
Carnell, P J.H. (1989): Chapter 4. Feedstock Purification, in Catalyst Handbook ed. by Martyn
V. Twigg, Second Edition, Wolfe Publishing Ltd, London, England.
Dong, Y. (1994): Kinetic Studies of Steam Reforming. Literature Survey. Aeurex Environmental
Corporation Project Report, March.
Dong, Y. and Borgwardt, R.H. (1996): Biomass Reactivity in Gasification bv the Hvnol Process.
Preprints of Papers Presented at the 212th ACS National Meeting, Vol. 41, No. 4, Division of
Fuel Chemistry, American Chemical Society, Orlando, FL, August 25-29.
47
-------
Froment, G,F. and Bischoff, K.B. (1979); Chemical Reactor Analysis and Design. John Wiiey &
Son, New York, p. 146,
Gerhard, E.R, and Moe, J.M. (1965); Chemical Reaction and Heat Transfer Rates in the Steam
Methane Reaction. AIChE Symposium, Fifty-sixth National Meeting, San Francisco, CA.
Holman, J.P. and Gajda, W.J. (1984): Experimental Methods for Engineers. 4th edition,
McGraw-Hill, Inc., New York, p.73.
Liu, J., Wang, F„ and Wu, G. (1986): Kinetic Model of CH,-CQ,-Steam Reaction at High
Temperature. Ind. Eng. Chem, Process Des. Dev., 25, (1), 273.
Motor Vehicle Emission Laboratory (1989): An Analysis of the Economic and Environmental
Effects of Methanol as an Automoti ve Fuel. EPA Report No. 0730 (NTIS PB90-225806), Ann
Arbor, MI, September.
Munster, P. And Grabke, H.J. (1981): Kinetics of the Steam Reforming of Methane with Iron-
Nickel. and Iron-Nickel Allovs as Catalysts. Journal of Catalysis, 12, 279.
Ridler, D.E. and Twigg, M.V. (1989): Chapter 5. Steam Reforming, in Catalyst Handbook ed.
by Martyn V. Twigg, Second Edition, Wolfe Publishing Ltd, London, England.
Rostrup-Nielsen, J.R. (1973): Activity of Nickel Catalysts for Steam Reforming of
Hydrocarbons. Journal of Catalysis, 31,173.
Rostrup-Nielsen, J.R. (1975): Steam Reforming Catalysis. Danish Technical Press, Inc.,
Copenhagen, Denmark, pi 01.
Steinberg, M. and Dong, Y. (1994); Process and Apparatus for the Production of Methanol from
Condensed Carbonaceous Materials. U.S. Patent No. 5,344,848.
Xu, J. and Froment, G.F. (1989): Methane Steam Reforming. Methanation and Water-Gas Shift:
I. Intrinsic Kinetics. AIChE Journal, 35, (1), 88.
48
-------
TABLE 1. ACTIVATION ENERGIES REPORTED IN PREVIOUS
WORK
Authors
Catalyst
size
P
T
1a
mm
a t in
°C
cal/mol
Akers & Camp (1955)
3.18
1
640
8,778
Ridlcr & Twigg (1989)
Foil
1
800-900
31,000
Rostrup-Nielsen (1973)
0.3-0.5
1-31
500
26,200
Munster & Grabke (1981)
N/A
1
700-850
38,000
Liu et al. (1986)
2
1.1
950-1100
12,000
Agnelli et al. (1987)
0.3-0.42
1
640-740
41,600
Xu & Froment (1989)
0.17-0.25
3-15
500-575
57,000
TABLE 2. PURITIES OF THE GASES USED
Gas
component
Cylinder
grade
Purity
%
Cylinder
pressure, psi
Cylinder
content, m3
ch4
Chemically pure
99.0
2,400
10.14
h2
Ultra-high purity
99.995
2,640
8.24
CO
Chemically pure
99.3
2,000
6.85
C02 in 1I2
Custom blending
30
2,000
5.63
n2
Industial
99.998
2,640
8.64
49
-------
TABLE 3. GG AND INTEGRATOR OPERATING CONDITIONS
Item
Setpoint
Initial oven temperature
35 "C
Final oven temperature
70°C
Temperature ramp
7.0°C/min
Injector temperature
150°C
TCD temperature
250°C
Carrier gas (He) flow
-30 mL/min
Reference gas flow
-45 mL/min
TCD sensitivity
High
Attenuation
2
Chart speed
1.0 cm/miii
Area reject
2000
Threshold
3
Peak width
0.04
TABLE 4. CATALYST PARTICLE DENSITY
CATALYST
Volume
Weight
Density
SAMPLE No.
cm3
g
g/em3
1
1,943
3.4182
1.76
2
1.970
3.4112
1.73
3
1.929
3.3804
1.75
AVERAGE
1.947
3.4033
1.75
50
-------
TABLE 5. CATALYST SOLID DENSITY
Ingredients
ps» g/cm3
wt.%
A1,03
4.00
60
MgO
3.65
25
Ni
8.90
15
Catalyst ps = (4 x 0.6) + (3.65 x 0.25) + (8.9 x 0.15) = 4.65 g/cm3
TABLE 6. CATALYST-SPECIFIC SURFACE AREAS AND PORE SIZES
Catalyst
Before Reaction
After Reaction
BET Surface Area (m2/g)
15.61
16.88
Average Pore Size (A)
242.16
363.71
TABLE 7. COMPARISON OF CATALYST PROPERTIES
Reference
Ps
Pp
S
4>s
e
g/cm3
g/cm3
m2/g
-
-
Agnelli et al. (1987)
2.80
1.02
36
-
0.49
Liu et al. (1986)
2.73
1.89
-
-
0.31
This work
4.65
1.75
15.6
0.39
0.62
51
-------
Table 8. Experimental data summary.
RUN
Catalyst
W
P
T
FEED GAS FLOW RATE (SLPM)
S
aim
C
H,
CO
CH<
CO,
HaO
R027-6
* Mif-7"
Powder
Powder
1.000
1.666"
25.1
" 25.6 ~
925
975"
3.181
3*187
0.798
~ 0.798
3.000
Tooo
0.360
0,360
8.36
~8,37
R027-B
Powder
1.060
25.0
901
IjiT
0.798
3.000
0.360**
~ 8 ~35
R028-4
Powder
1.000
25.0
950
3.181
0.798
3.000
0.360
8.39
R028-5
Powder
1 000
25.0
100Q
3.181
~6.798"
67666"
0.360
8.41
R028-6
Powder
1.000
25.0
901
3.181
0.798
3.000
0.360
8.42
R028-7
Powder
1.000
2S.0
950
3.183
0.798
3.000
0.358
8.42
R028-8
Powder
1.000
24.9
926'
3.183
0.798
3,000
6.358
8.44
R028-9
Powder
1.000
24.9
1000
3.183
0.798
3.000
0.353
8.41*
R032-2-1
Pellet
3.431
25.1
1001
3.181
0.798
3.000
0.360
10.40
R032-2-2
Pefiel
3.431
25.1
1001
3.181
0.798
3.000
0.360
8.45
R032-3
Pellet
3.431
25.0
1001
4.699
0.633
3,523
0.431
9.98
R032-4
Pellet
3.431
25.0
1001
5.476
0.736
4.113
0.503
11.62
R032-6
Pelfal
3.431
25.0
1002
3.181
0.798
3.000
0.360
8.38
R034-2
Pellet
3.339
25.0
951
3.931
0.529
2,935
0.350
10.41
R034-3
Pellet
3.339
25.2
901
3.931
0.529
2.935
0.350
10.48
R034-4
Pellet
3.339
25.2
1000
4.175
0.529
2.935
0.000
10,78
R034-5
Peliel
3.339
25.0
1001
4.175
0.529
2.935
0.000
8.57
R034-6
Pellet
3.339
25.0
1000
3.931
0.529
2.935
0.350
10.49
R035-1
PeHet
3.228
25.0
1001
3.810
0.529
2.935
0.350
10.50
R036-3
Pellet
3.234
25.0
902
4.175
0.529
2.935
O.OOO
8.26
R036-4
Pellet
3.234
25.1
901
3.931
0.529
2.935
6.350
10.37
R036-5
Pellet
3.234
25.1
901
3.678
0.529
2.935
0.697
10.41
R036-6
Peftet
3.234
25.0
902
3.931
0.000
2.935
0.350
10.43
R036-8
Pellet
3.234
25.1
899
3.943
0.529
2.935
0.338
10.30
R036-10
Pelfet
3.234
25.0
903
3.589
0.529
2.935
0.000~
10.34
R037-1
PeHet
3.195
25.1
953
3.181
0.798
3.000
0.359
i.67_
R037-2
Pellet
3,195
25.1
953
3.181
0.798
3.000
0.359
12.46 *
R038-2
Pelfet
3.162
25.0
1001
3.181
0.799
3.000
0.359
ToTif*
R038-4
Pellet
3.162
25.0
1000
3.181
1598
3.666"
*o73S9"
"1*0*745""
R038-5
Pellet
3.162
25.0
1000
3.181
67466"
T660"
67359
~ 10.51
R038-6
Pellet
3.162
25.0
1000
3.181
0.000
3.000
0.359
10*39
PRODUCT GAS
Hj CO
PLOW RATE
CH«' CO;
(SLPM)
HjO
W/F
gh/mol
HjO/C
mot/aim
meWg-h-alm
BALANCES{%)
Total
aC»m,c«i
6.88
7.35
6.39
7A2
8,3?"
6.40
7iV
6.64
8.48
5.94
5.95'
6.56
" 7.34 '
5.86
5.43
5.41
6.23
6.18
5.65
5.67
5.20
4.77
4.50
4.77
4.99
4.89*
4,84
4.66
1.0B
TsT
QM
~122
~1S*4
T.G3
lis"
1.14
153
0.89
0.95
0.85
0.96
0.89
0.61
0.57
0.56
0.65
0.68
0.66'
0.52
67s4
0.57
0.18
"o.iT
0.49
6771
2.23
1.88
2.39
2.10
~U2~
2.38
2.05
2.24
1~73
2.5?
2.50*
3.00
3.58
2.47*
2.55
2.61~
2.49
2.45
2.50~
2.50
2.60 '
~ZS8~
2.58
2.58"
2\59~
2.61
2.69
.0.83
0.85
~ojF
0,77
0J5
~075"
0,83
0.78
0^84
0.78'
0.73
0.72
0.80*
0.65
6,87
0.70
0.45*
0*42*
0.67
0,68 ~
0.27"
0,55
0.86"
0.42
"ase"
0.30
0.82*
0.80
7.25
6.93
7.41
7.20
6.95
7.69
7.20
7.44
* eiF
9.56
7.59
9.25
16.88
7.84
9.80
9*.98**
10.07
"7.72
"9.73"
9.97
7.86
9.92
9.99
10.08
9.74 *
"9.57"
"7.76
11.19*
0.124
6.124
ai24
6.124
0.124
6.124
0.124
0.124
0.?24
0.427*
0.427
0.364
0.311
6.427
0.425
0.425
0.425
D.425
0.425
0.411
0.411
6.411
£411
6.411*
0~4H
0.411
0.398
2.01
2.01
2.0i'
2.02
2.02
2.oT
2.03'
2.03
2.02
2.50
2.03*
2.18
"i.W
2.02
2.73
2.75
3.11
2.47
2.75
2.75
2.38*
2.72
2.S0
sTe"
J.71
2.98 "
25.67
37.33
20.33
30.00
42.67*
20.67
31.67
25.33
42.33
16.33
16.67
14.85
12.96
17.67
13,12
11.07
15.16
16.52
14.82
14.82
11.41
12.16'
12.10"'
12.16"
TmI"
11.75
0.847
0,390
-1.39
-0.43
5.64
5.98
0.638
SF
0Mir
0.684~
0.521-
1.615"
0.T01"
0.092
0.099
0.100~
0.099~
0.084
0.069"
o.ossT
0.097~
WWsT
6.099**
•1.64
-tllS
5.40
6^65"
0.05
0.34
"67io"
-1.35
0.53
oTsi"
-0.37
-0.22
-3.56
6.42
~1J3~
7.10
7.25"
6.45
6.52™
4.80
5.16
"6.38
Q.23
6*.26*
0.61~
TeF
1.62
1.62~
0.94
2.17
0.99
0.44
0.065
0,68
^2.U
2.28
-0.2S'
0.079
6.079~
6,077^
6.074 "
6.674
-3.78
-3.63~
*3.26*
-1.54
~109~
-3.47
-2.42
-0.91
2m
13.00
0.076
0,398
5.41
"5.35"
"^35"
~5.12
0.79
1.30
6759"
"g735~
2.55
2.50
J.10
0.75*
9.41
To9~
"9725*
9.52
0.393
3.00
!749~
15.33
0.103
0.393
1)7393*
"0*393"
2.11
~2.80
~2M'
15.00
16.67'
1fk67
0.105
„0"1"
~67l6T~
J.1S
1.22"
jO.46
0.66"
1.13
1.13
0.63
0.52
2,64
2,84
2 J?
i.eT
0.87
6.77
0.32
6754~
0.53~
*1.33*
f.00~
2.03
2.09
1.22
0.26
2.23
7.49
-0.43
^a27"
-1.92
~-6.1i~
-2.48^
-1.78
1.49
1.24
1.17
6.77
2.41 "
2.87
2.51
2! 15
1.20
1 .ia
105
0.097
0.089
0.33
6.36
1.03
0,83
t76T~
2.02
1.24
0.39
T5T
0.50
0.095
0,0*96 "
0.094 "
6.126
67t 70"
0.096
6.093
0.093
0.095~
0.159"
-1.11
-115
-67b5
-153
•1,84
0.07
~*sT
0.193
"67195
~6Jb7"
~o.m"
0,t77~~
o.m
0.108
16.11
17.80
1-188"
12J6*
17*87"
13.13
Tmo
15.22~
ti.ss"
15.24
14.92
12.46 "
"VumT
11.6V
1135
Tios"
1153*
14.22
11.77
-0.19
•1.01
0.098
6T02~
-1.42
-6.95
0.107
J4.94
jsTolT
15^8 "
(Continued)
-------
Table 8. Experimental data summary, (continued)
U>
RUN
Catalyst
W
P
T
FEED GAS FLOW RATE (SLPM)
PRODUCT GAS FLOW RATE (SLPM)
W/F
h2o/c
^cm
k0j,
BALANCES <%)
n.n
Xcm, cat
g
atm
C
h2
CO
O
X
co2
h2o
H:
CO
CH„
co2
HaO
gh/mol
mol/atm
%
mol/g-h-atm
C
H
0
Total
- ! %
R039-2
Pellet
3.204
25.1
950
4.000
0.799
3.000
0.359
10.42
5.76
0.78
2.63
0.76
9.42
0.399
2.51
12.33
0.083
0.29
0.10
-1.82
-1.18
0.126
12.39
R039-3
Pettet
3.204
25.1
950
1.999
0.799
3.000
0.359
10.42
3.84
0.67
2.60
0.87
9.28
0.399
2.51
13.33
0.081
-0.43
-0.53
-2.06;
•1.57
0.123
13.71
R039-4
Pellet
3.204
25.0
949
3.181
0.799
3.000
0.359
10.46
4.49
0.73
2.64
0.79
9.42
0.399
2.52
12.00
0.078
0.05
•2.30
-2.06
•1.73
0.119
12.79
R039-5
Pellet
3.204
25.1
951
1.096
0.799
3.000
0.359
10.40
2.81
0.63
2.71
0.90
9.30
0.399
2.50
9.67
0.054
1.97
0.20
-1.54
-0.69
R041-2
Pellet
3.275
25.2
903
3.181
0.799
3.000
0.359
9.69
4.79
0.85
2.64
0.63
8.90
0.408
2.33
12.00
0.072
-0.91
0.52
•1.76
-1.28
0.174
11.97
R041-3
Pellet
3.275
25.0
903
1.098
0.799
3.000
0.359
9.75
2.56
0.81
2.73
0.66
9.04
0.408
2.34
9.00
0.047
1.01
1.26
-0.86
-0.24
R041-4
Pellet
3.275
25.0
905
3.181
0.799
3.000
0.359
9.41
4.55
0.84
2.64
0.54
8.70
0.408
2.26
12.00
0.072
-3.32
. -0.33
-2.81
-2.56
0.169
12.14
R041-5
Pellet
3.275
25.0
903
1.999
0.799
3.000
0.359
9.17
3.34
0.81
2.69
0.61
8.67
0.408
2.21
10.33
0.056
-1.15
1.29
0.12
0.02
R041-6
Pellet
3.275
24.9
901
1.098
0.799
3.000
0.359
9.22
2.54
0.79
2.72
0.62
8.63
0.408
2.22
9.33
0.048
-0.67
1.79
-0.72
-0.39
R041-7
Pellet
3.275
25.1
901
4.000
0.799
3.000
0.359
9.18
5.40
0.84
2.63
0.53
8.62
0.408
2.21
12.33
0.076
~3.80
0.52
-1.65
•1.75
0.187
11.66
R041-8
Pellet
3.275
25.1
901
0.838
0.799
3.000
0.359
9.19
2.60
0.78
2.65
0.63
8.68
0.408
2.21
11.67
0.059
-2.36
3.44
0.12
0.05
R042-2
Pellet
3.115
25.1
900
3.184
0.799
3.000
0.355
9.19
5.00
0.84
2.64
0.58
8.15
0.388
2.21
12.00
0.074
•2.26
0.30
-5.13
-3.81
0.184
11.59
R042-3
Pellet
3.115
25.0
900
0.841
0.799
3.000
0.355
9.20
2.55
0.81
2.66
0.65
8.02
0.388
2.21
11.33
0.060
-0.82
-0.94
-5.41
-3.95
R042-4
Petlet
3.115
25.0
901
1.061
0.799
3.000
0.355
9.17
2.43
0.82
2.67
0.61
8.25
0.388
2.21
11.00
0.059
-1.30
-1.30
-3.64
-2.89
R042-5
Pettet
3.115
25.0
901
4.004
0.799
3.000
0.355
9.09
5.67
0,87
2.76
0.52
7.69
0.388
2.19
8.00
0.050
-0.10
-1.12
-9.43
-6.39
R044-1
Pellet
3.123
25.1
900
3.184
0.799
3.000
0.355
8.38
4.45
0.72
2.70
0.71
7.90
0.389
2.02
10.00
0.058
-0.58
1.06
1.53
1.02
0.143
12.16
R044-2
Pellet
3.123
25.1
950
3.184
0.799
3.000
0.355
8.34
4.72
0.76
2.65
0.70
7.76
0.389
2.01
11.67
0.068
•1.06
1.46
0.72
0.45
0.103
14.07
R044-3
Pellet
3.123
25.1
1000
3.184
0.799
3.000
0.355
8.35
4.76
0.76
2.56
0.71
7.74
0.389
2.01
14.67
0.087
-2.99
0.49
0.62
-0.15
0.085
16.41
R045-6
Pellet
3.150
25.0
950
3.184
0.799
1.000
0.355
9.29
3.98
0.57
0.84
0.71
8.42
1.176
4.31
16.00
0.088
-1.58
-2.72
-3.60
-3.27
R045-7
Pellet
3.150
25.0
951
3.184
0.799
3.000
0.355
9.29
4.63
0.72
2.57
0.74
8.29
0.392
2.24
14.33
0.090
-2,99
-2.24
-2.86
-2.81
0.135
13.46
R046-4
Pellet
3.203
24.9
951
3.179
0.799
1.000
0.360
9.23
4.64
0.58
0.84
0.76
8.52
1.196
4.28
16.00
0.086
0.97
2.99
-1.20
-0.43
R046-7
Pellet
3.203
25.0
950
3.179
0.799
2.000
0.360
9.24
5.02
0.65
1.72
0.77
8.36
0.598
2.92
14.00
0.080
-0.60
2.44
•1.94
-1.15
R046-9
Pellet
3.203
24.9
951
3.179
0.799
3.000
0.360
9.25
5.36
0.75
2.58
0.78
8.21
0.399
2.22
14.00
0.086
•1.18
1.63
-2.31
-1.54
0.130
13.64
Average
•0.84
0.19
•1.39
•1.09
-------
TABLE 9. BLANK TEST RESULTS (PR-09)
Time
ch4in
Out Rate
CH,
CH, OUT
A CH4
SLPM
L/min
%
L/min
%
1:00
3.14
11.80
28.68
3.38
7.64
1:30
3.14
11.06
27.98
3.09
-1.59
2:00
3.14
11.06
28.36
3.14
0.00
2:30
3.14
11.36
26.42
3.00
-4.46
3:00
3.14
10.93
28.00
3.06
-2.55
3:30
3.14
11.53
26.88
3.10
-1.27
4:00
3.14
10.95
27.74
3.04
-3.18
Average
3.14
11.24
27.72
3.12
-0.64
TABLE 10. TEMPERATURE DEPENDENCE OF CATALYST POWDERS,
(25 ATM, HYNOL FEED GAS, W/F = 0.124 G-H/MOL, H20/C = 2)
RUN No.
T
XcH4
kj
°C
%
mol/g-h-atm
R027-8
901
20.33
0.398
R028-6
901
20.67
0.407
R027-6
925
24.67
0.522
R028-8
926
25.33
0.521
R028-4
950
30.00
0.638
R028-7
950
31.00
0.684
R027-7
975
36.67
0.847
R028-5
1001
42.67
1.022
R028-9
1000
42.33
1.015
54
-------
TABLE 11. TRANSPORT PROPERTIES OF GAS COMPONENTS
Property
pTT
LxJnL^
h2
CO
CO,
h20
Mixture
Rate, SLPM
3.00
3.181
0.799
0.359
8.330
15.669
Pj, atm
4.79
5.08
1.27
0.57
13.29
25.0
Pc, atm
46
13.1
35
73.8
220.5
-
Tc, atm
190.6
33.2
132.9
304.2
647.3
-
Mj, g/mol
16.04
2.016
28.01
44.01
18.016
15.50
Pi xlO3, g/cm}
3.84
0.48
6.70
10.53
4.31
3.71
,u,x10\ g/crn-s
3.219
1.996
4.777
4.966
3.405
3.497
Dcll4i» cm%
-
0.417
0.122
0.097
0.289
0.263
TABLE 12. EFFECTIVENESS FACTORS OF CATALYST PELLETS
T, °C
900
950
1000
Average
0.177
0.127
0.096
Minimum
0.167
0.119
0.092
Maximum
0.186
0.134
0.099
sd
0.010
0.008
0.003
55
-------
TABLE 13. DATA QUALITY INDICATOR GOALS FOR ACCURACY
Parameter
Quality Objective
QC Methods
Catalyst weight
Accuracy: ± 0.01 percent
Mettler AT200 analytical
balance used
Feed rates of CH4, H2,
CO and C02
Accuracy: ± 2 percent
MFCs are calibrated at
high operating pressures
Water feed rate
Accuracy: ± 2 percent
HPLC pump used and
calibrated during runs
System pressure
Accuracy: ± 0.2 kg/cm2
Mity-Mite S-91 W back
pressure regulator used
Reaction temperature
Accuracy: ± 5°C
LabTeeh Control
software used for
temperature controlling
Condensates rate
Accuracy: ± 2 percent
(estimate)
Calculated from the
condensate weight
collected and the time
recorded by stopwatch for
water accumulation
Offgas flow rate
Accuracy: ± 1.5 percent
Dry gas meter is
calibrated
Offgas composition
Accuracy: ± 2 percent
A QA check is run on the
GC daily
56
-------
TABLE 14. DATA QUALITY OF GC ANALYSIS (GAS CYLINDER No. 5)
Calibration Date: 5/20/96
Gas #5
H2
N2
CO
CH4
C02
%
76.30
2.01
4.90
12.80
3.99
Cor. Factor
0.992
0.987
0.992
0.998
0.988
GC Run
1
73.17
2.01
4.96
12.87
4.15
2
72.46
2.05
4.99
12.91
4,05
3
78.37
2.14
5.01
12.99
4.03
4
78.55
2.07
4.99
12.99
4,04
5
76.22
2.01
4.94
12 85
4.04
6
78.21
2.06
4.95
12.87
4.01
7
77.47
2.00
4,92
12.77
4.02
8
75.71
2.03
4.93
12.84
4.01
9
77.44
2.01
4.93
12.81
4.06
10
76.69
2.04
4.91
12.76
4.03
11
78.52
2.03
4.91
12.77
4.02
12
75.43
2,03
4.93
12.74
4.01
13
74.42
2.05
4.92
12.78
4.02
14
80.17
2.05
4.98
12.87
4.07
15
78.77
2.01
4.90
12.75
4.00
18
75.75
2.02
4.92
12.75
4.03
17
75.08 j
2.03
4.93
12,84
4.03
18
79.85 i
2.02
4.93
12.79
4.04
19
78.71
2.01
4.90
12.76
4.02
20
77.20
2.08
4.94
12.78
4,09
Average
76.91
2.04
4.94
12.82
4.04
i
sd
2.20
0.04 |
0.05
0.08
0.06
Uncertainty
1.41
0.03
0.03
0.05
0.04
U, %
1.85
1.38
0.65
0.39
0.96
57
-------
TABLE 15. DATA QUALITY OF GC ANALYSIS (GAS CYLINDER No. 6)
Calibration Date: 7/23/96
Gas #6
H2
N2
CO
CH4
C02
%
52.61
10.50
24.70
5.10
7.09
Cor. Factor
1.006
0.994
0.994
1.017
1.013
GC Run
1
52.36
10.52
24.73
4.97
7.04
2
52.75
10.47
24.70
4.99
6.98
3
52.85
10.53
24.76
5.00
7.13
4
52.10
10.46
24.60
4.98
6.89
5
52.67
10.52
24.74
4.99
7.01
6
52.61
10.47
24.70
5.00
7.02
7
53.76
10.49
24.75
5.02
6.93
8
50.88
10,59
24.89
5.01
8.96
9
51.26
10.67
25.08
5.08
7.09
10
51.55
10.63
25.06
5.06
7.11
11
52.39
10.59
24.99
5.05
6.98
12
52.28
10.71
25.06
5.08
7.01
13
52.00
10.78
24.92
5.03
6.95
14
52.68
10.54
24.89
5.01
7.03
15
52.87
10.56
24.82
5.00
7.00
16
53.28
10.58
24.89
5.02
7.02
17
52.56
10.58
24.86
5.00
7.03
18
51.24
10.55
24.76
4.99
6.82
19
52.00
10.51
24.78
4.99
6.93
20
52.20
10.54
24.76
4.99
6.99
Average
52.31
10.56
24.84
5.01
7.00
sd
0.77
0.10
0.19
0.10
0.12
Uncertainty
0.49 !
0.07
0.12
0.06
0.08
U, %
0.93 I
0.64
0.50
1.19
1.09
58
-------
HP Steam
HP Steam
Flue Gas
v©
NG Fuel
J_
Filter
Removal
HP Steam
Biomass
Heat
Exchanger
HP Steam
HP Steam
X
Water
Compressor
Heat
Exchanger
Compressor
Methanol
t
Water
HPR Hydrogasifier
SPR Steam reformer
MSR Methanol converter
CON Condenser
SEP Separator
REF Distillation column
Figure 1. Hynol process flowsheet.
-------
Nitrogen Out
Reactor Inlet
Top Flanae Plate
Top Flange
Pressure-Retaining
Vessel
ThermocouD e
Reactor tube
Top Heater
Ceramic
Catalyst
Midd e Heater
Pe et
Ceramic
Thermocouple
Thermocouple
Catalyst
Supporter
Bottom Flange
Bottom Heater
Thermocouple
Nitrogen In
J Reactor Exit
Figure 2. Schematic diagram of the steam reforming reactor.
60
-------
Filter
TC
,tc
—TC
TC
TC
CH4
Condenser
TC
TC
C02
Ice
Bath
Back-Pressure
Regulator
CO
TC
TC —¦
> Vent
TC
TC
Gas Mixer
Dry Gas
Meter
Dryer
TC
Trap
TC
TC
Steam
Reformer
Integrator
H2Q +
TC
Steam
MFC
MFC
MFC
MFC
MFC
GC
Steam
Generator
Figure 3. Flowsheet of the experimental equipment.
-------
-START ;
2 .962
4 .137
4 , 64S
8.091
TIM^flBLE STOP
13 .895
Closing signal
Storing report
RUN#
97
file R:Q225494E.BNC
to R«Q225494E.RPT
FEB 14, 1996 15=27:09
ESTQ-RREB
RT TYPE
2.962 N PB
1,13? PB
1.645 BB
8.091 PB
13.895 BB
AREA
WIDTH
CflLtt
MOLARS
NRME
41898
.067
1R
66 .632
H2
473356
.103
2R
8 .962
N2
942094
.120
3R
17 .329
CO
104680
.179
4R
2 .132
CH4
490682
.278
5R
6 .244
C02
TOTAL fiREfi=20S2?10
MUL f RCT0R = 1 . 0000E+ 00
Figure 4. Typical printout of the the GC integrator.
62
-------
SPR DATA SHEET
RUN No:
R028-4
7/12
Page 1
DATE:
GC CHECK:
h2
n2
CO
ch4
CN
O
o
Calibration Date
5/20/96
Gas #5
76.300
2.010
4.900
12.800
3.990
GC
79.849
2.015
4.925
12.792
4.042
Error%
4.65
0.25
0.51
-0.06
1.30
Correction Factor:
0.994
0.988
0.992
0.998
0.989
GC
Run
1
73.167
2.012
4.956
12.871
4.146
2
72.462
2.047
4.988
12.908
4.048
3
78.365
2.135
5.009
12.988
4.030
4
78.554
2.074
4.987
12.986
4.043
5
76.216
2.007
4.941
12.853
4.042
6
78.212
2.056
4.947
12.868
4.008
7
77.473
1.996
4.918
12.769
4.018
8
75.711
2.030
4.933
12.843
4.006
9
77.444
2.006
4.933
12.810
4.063
10
76.688
2.036
4.909
12.759
4.033
11
78.522
2.034
4.912
12.768
4.024
12
75.430
2.025
4.927
12.742
4.00Sf
13
74.424
2.054
4.919
12.780
4.021
14
80.167
2.045
4.978
12.868
4.067
15
78.766
2.005
4.895
12.747
3.997
16
75.753
2.016
4.915
12.752
4.029
17
75.083
2.033
4.932
12.837
4.025
18
79.849
2.015
4.925
12.792
4.042
19
20
Average
76.794
2.035
4.940
12.830
4.036
sd | 2.182
0.033
0.032
0.076
0.033
Figure 5. Example of SPR data sheets (page 1).
-------
SPR DATA SHEET
c*
4^.
RUN No:
R028-4
Catalyst Type:
HT
1.00
Diluent Type:
CR
90
Page 2
DATE:
7/12
W (g):
Diluent (g):
Dp
(mm):
EXPERIMENTAL CONDITIONS
HEATE
Set
Gas Rate (SLPM)
Balance N2 (SLPM)
SG
400
400
450
650
650
Set Poin
Actual
set
OUTLET
ch4
2.785
2.726
0.700
1.089
3.000
0.10
IN #1
h2
2.342
IN #2
CO
0.798
IN #3
o
o
ro
1.199
C02%
30 0
DATA:
Time
P
SG
INLET
REACTOR
OUTLET
ICE BATH
HEATER ACTUAL
H:M
atm
T201
T202
T203
T204
T205
T207
T208
T112
T113
SG
OUTLET
H212
H213
H214
9:00
25.02
294
380
950
950
952
478
369
19
9
377
398
923
900
793
9:30
25.02
333
381
950
950
950
496
365
19
9
434
398
924
903
804
10:00
25.02
357
381
951
950
950
516
371
19
10
452
401
923
904
800
10:30
25.02
327
381
952
951
950
500
370
20
11
402
400
923
906
801
11:00
25.02
342
382
949
950
951
476
369
20
11
445
400
923
909
802
11:30
25.02~l
346
381
948
952
954
533
370
21
11
411
400
925
910
804
12:00
25.02
328
381
950
952
950
475
368
21
11
412
398
926
910
803
12:30
25.02
355
381
948
950
950
489
370
21
11
459
400
926
908
805
13:00
24.99
319
381
950
951
950
482
367
21
11
391
398
927
909
805
13:30
24.99
344
381
951
951
951
479
366
21
11
439
398
927
909
804
14:00
24.99
350
382
949
950
950
534
371
21
11
430
405
926
906
810
14:30
24.94
332
381
948
950
950
485
367
22
12
413
398
929
909
807
END
Figure 5. Example of SPR data sheets (page 2).
-------
SPR DATA SHEET
RUN No.
R028-4
Page 3
Date
7/12
Time
FI-101
AP
T
Dry Gas
Trap Water
Water In
Product Gas (mol%)
H:M
sec/.3cf
in H20
°C
l/min
9
min.
sec.
g/min
set
s/10ml
ml/min
H2
N2
CO
CH4
C02
9:00
40.5
1.9
23
12.79
190.4
31
42
6.01
6.10
89.2
6.73
60.359
10.006
10.419
15.908
6.371
9:30
41.5
1.8
23
12.45
152.3
27
19
5.58
6.10
89.1
6.73
60.601
8.423
10.494
16.779
6.425
10:00
42.5
1.8
23
12.16
193.3
31
47
6.08
6.10
89.0
6.74
59.437
9.313
9.021
18.101
6.915
10:30
43.0
1.8
23
12.02
148.5
26
39
5.57
6.10
88.9
6.75
59.508
7.074
10.659
17.323
6.506
11:00
43.2
1.8
23
11.96
190.3
32
13
5.91
6.10
89.0
6.74
61.540
6.741
10.504
17.406
6.678
11:30
43.7
1.8
24
11.81
191.7
31
13
6.14
6.10
88.6
6.77
58.835
9.095
9.570
18.146
6.831
12:00
43.7
1.8
24
11.79
156.0
28
56
5.39
6.10
88.5
6.78
61.061
6.543
10.559
17.836
6.575
12:30
44.1
1.8
24
11.68
196.0
32
20
6.06
6.10
88.9
6.75
61.740
6.169
10.402
18.157
6.829
13:00
44.6
1.7
25
11.52
147.7
27
24
5.39
6.10
89.3
6.72
59.108
6.152
10.544
18.117
6.612
13:30
44.3
1.7
25
11.60
190.2
30
58
6.14
6.10
89.1
6.73
60.561
6.192
10.558
18.084
6.578
14:00
44.0
1.7
25
11.68
183.4
29
34
6.20
6.10
88.8
6.76
59.222
8.282
9.181
19.190
6.896
14:30
44.2
1.7
25
11.61
150.7
27
18
5.52
6.10
89.1
6.73
62.046
6.075
10.573
18.080
6.611
END
Figure 5. Example of SPR data sheets (page 3).
-------
SPR DATA SHEET
CJ\
ON
RUN No.
R028-4
P atm
25.0
Page 4
Date
7/12
INPUT
H2
CO
CH4
C02
H20
Total
C
H20/C
W/F
SLPM
SLPM
SLPM
SLPM
SLPM
SLPM
atom
mol/mol
gh/mol
Rate
3.181
0.798
3.000
0.360
8.386
15.725
0.186
2.02
0.12
p, atm
5.06
1.27
4.77
0.57
13.34
25.01
In
Output
t
T"av
H20
CO
CH4
C02
H2
H20
XcH4
d
dC
dH
do
Total
hr
°C
l/min
l/min
l/min
l/min
l/min
l/min
%
%
%
%
%
0.5
951
8.36
1.32
2.03
0.81
7.67
7.47
32.36
2.69
e
-0.05
9.40
5.25
4.77
1.0
950
8.37
1.30
2.08
0.79
7.50
6.93
30.51
0.85
0.33
5.96
-0.81
0.41
1.5
950
8.38
1.09
2.20
0.83
7.18
7.56
26.81
2.86
e
-1.03
8.94
4.17
3.80
2.0
951
8.39
1.27
2.08
0.77
7.11
6.93
30.77
1.10
-0.90
3.50
-1.66
-0.75
2.5
950
8.38
1.25
2.08
0.79
7.31
7.34
30.76
1.09
-1.08
7.11
2.73
2.59
3.0
952
8.42
1.12
2.14
0.80
6.90
7.64
28.76
0.90
-2.48
6.87
4.15
3.19
3.5
951
8.43
1.23
2.10
0.77
7.15
6.70
30.10
0.43
-1.46
2.48
-4.80
-3.06
4.0
950
8.39
1.20
2.12
0.79
7.16
7.54
29.49
0.18
-1.19
7.73
4.13
3.57
4.5
951
8.35
1.21
2.08
0.75
6.77
6.70
30.57
0.90
-2.80
0.58
-4.63
-3.50
5.0
951
8.37
1.22
2.09
0.75
6.98
7.64
30.22
0.56
-2.29
7.13
4.76
3.66
5.5
950
8.40
1.06
2.24
0.80
6.87
7.71
25.45
4.22
e
-1.47
8.40
4.54
3.87
6.0
950
8.37
1.22
2.09
0.76
7.16
6.86
30.20
0.53
-2.11
3.73
-2.95
-1.81
Ave
950
8.39
1.21
2.11
0.78
7.15
7.25
29.67
2.60
sd
1.88
Chau
1.39
Carbon loss (g/h)
0.00
END
Figure 5. Example of SPR data sheets (page 4).
-------
SPR DATA SHEET
ON
-J
RUN No.
R028-4
P atm
25.0
Page 5
Date
7/12
INPUT
H2
CO
CH4
C02
H20
Total
C
H20/C
W/F
SLPM
SLPM
SLPM
SLPM
SLPM
SLPM
atom
mol/mol
gh/mol
Rate
3.181
0.798
3.000
0.360
8.386
15.725
0.186
2.02
0.12
p, atm
5.06
1.27
4.77
0.57
13.34
25.01
In
Output
t
TaV
H20
CO
CH4
C02
H2
H20
XcH4
d
dC
dH
dO
Total
hr
°c
l/min
l/min
l/min
l/min
l/min
l/min
%
%
%
%
%
0.5
1.0
950
8.37
1.30
2.08
0.79
7.50
6.93
30.51
0.36
0.33
5.96
-0.81
0.41
1.5
2.0
951
8.39
1.27
2.08
0.77
7.11
6.93
30.77
0.62
-0.90
3.50
-1.66
-0.75
2.5
950
8.38
1.25
2.08
0.79
7.31
7.34
30.76
0.61
-1.08
7.11
2.73
2.59
3.0
952
8.42
1.12
2.14
0.80
6.90
7.64
28.76
1.39
-2.48
6.87
4.15
3.19
3.5
951
8.43
1.23
2.10
0.77
7.15
6.70
30.10
0.05
-1.46
2.48
-4.80
-3.06
4.0
950
8.39
1.20
2.12
0.79
7.16
7.54
29.49
0.67
-1.19
7.73
4.13
3.57
4.5
951
8.35
1.21
2.08
0.75
6.77
6.70
30.57
0.41
-2.80
0.58
-4.63
-3.50
5.0
951
8.37
1.22
2.09
0.75
6.98
7.64
30.22
0.07
-2.29
7.13
4.76
3.66
5.5
6.0
950
8.37
1.22
2.09
0.76
7.16
6.86
30.20
0.04
-2.11
3.73
-2.95
-1.81
Ave
950
8.39
1.22
2.10
0.77
7.12
7.14
30.15
0.91
sd
0.65
Chau
1.39
Carbon loss (g/h)
0.00
END
Figure 5. Example of SPR data sheets (page 5).
-------
100 kg dry biomass
11.1 kg moisture
Carbon 0.085 mol
Gasmer
go
C02
CH4
H20
H2
N2
2.51
1.13
5.68
4.17
10.00
0.53
24.02 mol
22.2 mol
Steam 2.4 mol.
Purge 0.2 mot
CO
C02
CH4
H20
H2
N2
MeOH
18.3
5.4
22.9
0.13
133.5
5.0
2.17
CO
1.94
C02
0.57
CH4
2.42
H20
2.41
H2
14.10
N2
0.53
MeOH
0.23
167 mol
187 mol
MeOH 7.87 mol
H2O0.51 mo
Condenser
CO
18.3
C02
5.4
CH4
22.9
H20
0.7
H2
133.5
N2
5.0
MeOH
10.0
196 mol
CH4 3.75 mol Steam 6.65 mol
1
Reformer
48.4 mol
H203.85 mol
44.6 mol
CO
9.83
C02
0.80
CH4
2.44
H20
0.31
H2
30.67
N2
0.53
212 mol
Methanol
Converter
Figure 6. Hynol process simulation results.
68
-------
40
35
30
25
20
15
10
5
0
I i i I I 1 I I I I I i ' I I I I I I I I I 1 I 1 I I I L.
O
o
%
o
o
R025
25 atm and 1000°C
W/FCH4 = 0.1 g-h/mol
H20/C = 2.07
qOXD ° Q o 03 CK °° 00000°
O O u o o O wo
I 1 1 1 1 1 1 1 1 | I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 T"
0 5 10 15 20 25 3(
Time, hours
Figure 7. Stabilization of catalyst activity.
-------
0.15
* ' ' ' i '
i » i i i i -i-
J i—L.
-L 1 i
Ni-catalyst pellet
25 atm
CO
i
L—
JZ
6)
0.10
CD
c
•§->
U)
c
o
O
0
15
0.05
c
o
ts
as
0
Qu
0.00
1000°C
950°C
Zx
o
2.0
2.2
-V
i I i •
2.4
2.6
o
~
A
V
R032
R037
R034
"i 1 i i s i i i r*
2.8
3.0
3.2
H20/C
Figure 8. Effects of steam ratio on catalyst activity.
70
-------
50
Ni-catalyst powders
O Series R027
~ Series R028
i t 1 1 1 1 1 1 1 1 1 1 j 1 1 i 1 1 r—| ? : ? ?
900 920 940 960 980 1000
T, °C
Figure 9. Effects of temperature on methane conversion
by catalyst powders.
71
-------
0.5
Ni - catalyst powder
25 atm
Ea = 27.97 Kcal/mol
0.0 -
-Q
£ -0.5
_i
k1 = 65552 exp(-27970/RT)
-1.5
0.00078
0.00080
0.00082
0.00084
0.00086
1/T (1/ K)
Figure 10. Arrhenius plot for catalyst powders.
72
-------
0.10
~ R-034
O R-044
0.09 -
E
-i—•
o
E
JD
O
0.07 -I
0.06
0.05
880 900 920 940 960 980 1000 1020
Temperature, °C
Figure 11. Activities of commercially sized catalyst pellets.
73
-------
Ni-catalyst pellets
25 atm
1 it | i r i r"""7 i i f i r~r 1 1 1 j 3 1 i f i i i r
880 900 920 940 960 980 1000 1020
Temperature, °C
Figure 12. Effectiveness factors of catalyst pellets.
74
-------
25
Ni-catalyst pellets
25 atm
20 -
8
xr
x
p
X
10
15
20
5
25
Y
CH4, obs
Figure 13. Comparison of calculation results with
experimental methane conversion data.
75
-------
R038 1000°C
0.10
0.05
0.00 —-i—i—i—i—|—i—i—i—i—|—i—i—i—i—|—i—i—i—i—|—i—i—i—i—
0.0 0.5 1.0 1.5 2.0 2.5
CO partial pressure, atm
Figure 14. Effects of the CO partial pressure
in the feed gas on catalyst activity.
o
o
o
~
R036 900°C
Ni-catalyst pellet
25 atm
76
-------
0,12
0.08
6'
J§ 0.04
o
J*:
o.oo
0.00
1000°C
900°C
-O
-
-
Ni-catalyst pellet
_
25 atm
-
~
R-034
i f 'T r j
o
i \ \ " \ | i r
R-036
i i | i i i i
0.25 0.50 0.75
C02 partial pressure, atm
Figure 15. Effects of the C02 partial pressure
in the feed gas on catalyst activity.
1.00
77
-------
J 1 I L_
_l I I L_
PH2 (atm) = 5.38 3.01
4.47
1.75
RUN SERIES: R-039
T = 900°C
H20/C = 2.5
W/F = 0.399 g-h/mol
t 1 1 r
0
15
5 10
Time, hrs
Figure 16. Effects of the H2 partial pressure in the feed gas. (R039)
20
-------
' ' I
111)11
I I 1 i > ' ' '
' ' '
PH2 (atm) = 4.70 1.83 4.75 3.26 1.89 5.78
RUN SERIES: R-041
T = 900°C
H20/C = 2.2 - 2.3
W/F =.0.408 g-h/mol
I ' ' r
25
0
I i i
~ l i i i r~
10
15
Time, hrs
20
30
Figure 17. Effects of the H2 partial pressure in the feed gas. (R041)
-------
20
15
00
o
c
o
'(/>
CD
| 10
O
0
sz
CO
JZ
»
-------
0.10 -
~
Ni-catalyst pellet
950°C
O R045
~ R046
0
1 2 3
Methane feed rate, SLPM
Figure 19. Constant k„bs observed at various methane feed rates.
81
-------
~i y~" ) 1 i i r™ i i i i i r
0
n—p—i—i—i—i—j—i—i—i—i—|—i—i—r™ r
10 15 20 25 30
N
Figure 20. Chauvenet's criterion for rejecting a reading.
82
------- |