Transcript ElizabethEhrkePosterx - Physics

Catalytic Ignition Conditions of Hydrogen in a Tubular Reactor
Elizabeth Ehrke
Department of Physics, Case Western Reserve University, Cleveland, OH 44106
Research Advisor: Chih-Jen Sung
Department of Mechanical and Aerospace Engineering, Case Western Reserve University
Abstract
Widespread use of hydrogen fuel cell vehicles can potentially decrease national energy use and reduce harmful emissions. One of the main obstacles to public release is the development of validated safety codes and standards. A
tubular reactor system was developed to test the ignition propensity of hydrogen/air mixtures for fire safety applications. The simulated hydrogen/air mixture enters a heated metal tube, which acts as a catalyst to ignite the
hydrogen at a specific temperature. The mixture composition and flow rate can be varied to test the affect of equivalence ratio and residence time on ignition propensity. Initial testing shows that ignition occurred in a .4 mm
diameter platinum tube with varied flow parameters. The temperature of ignition is unique to the specific flow rate and mixture composition. Future testing using this system will record complete ignition profiles for multiple
materials, which can be used to identify high-risk leak situations and improve safety standards for hydrogen-powered devices.
Introduction
Sonic Nozzle Calibration
Results and Discussion
Hydrogen has been successfully used in industry for decades, but current
safety codes and standards must be updated for the situations
encountered during public use of hydrogen fuel cell vehicles.
Sonic nozzles are used to set the specified flow rate and equivalence
ratio (ϕ). They allow low flow rates, which are needed due to the small
diameter tubes used in the system. Small tubes are used to prevent gas
phase reactions, which release too much heat and could melt the tube,
connections, or thermocouples.
Catalytic ignition was observed for multiple test parameters. Figure 3 shows the
temperature and voltage plots of two tests with ϕ of .4 and .5. The arrows show
the manual voltage increases. The red circles show the ignition temperature. The
large temperature spikes do not correspond to a voltage increase, which
indicates catalytic ignition through surface reactions. Ignition occurs at about 900
K for ϕ=.4 and about 1000 K for ϕ= .5. The temperature spike is the flame front
moving past thermocouple 1 to the entrance of the tube. After the voltage is
turned off the temperature stays elevated, which indicates that the flame is self
sustaining. The ignition plots and their differences demonstrate the system is
functioning correctly, and it is possible to characterize the critical ignition
conditions for a range of initial parameters.
Hydrogen/air mixtures have been found to ignite anywhere between 4%75% hydrogen by volume1. Under normal leak conditions, this is an
acceptable risk because the low molecular weight of hydrogen allows
quick dissipation below flammability limits. When the leak is confined to a
small heated area, the ignition propensity drastically increases since the
hydrogen cannot dissipate quickly.
A hydrogen leak within a fuel cell vehicle allows for the possibility of
trapped gas coming into contact with heated metal surfaces. Certain
metals can act as a catalyst to ignite hydrogen/air mixtures, which can be
sustained through surface reactions.
A system has been designed to determine the critical ignition conditions
for hydrogen/air mixtures within heated metal tubes to eventually develop
an ignition envelope to aid in the improvement of fire safety standards for
hydrogen fuel cell vehicles.
Test System Design
The test system is designed to flow a specific mixture of hydrogen, oxygen,
and nitrogen gas into a heated metal tube to determine the ignition
temperature. As seen in figure 1 below, the metal tube is heated by the
power supply, and the temperature of the tube is measured at three points
using thermocouples and recorded using a National Instruments data
acquisition system. Increasing the voltage across the tube increases the
temperature by the resistance of the metal. The pressure regulators are
used to control the backpressure of the sonic nozzles to set the gas flow
rates. The solenoid valves are controlled using a LabView program to
switch the flow to allow only nitrogen into the tube to aid in cooling and
extinguish self sustaining flames.
The flow through the nozzle is choked, so only changes in backpressure
will affect the flow rate through the nozzle. Each nozzle was calibrated
using a bubble meter which measures the flow volume over time. As
expected, the flow rates for hydrogen, oxygen, and nitrogen were linearly
proportional to the nozzle backpressure.
Volumetric flow rate Q of a choked flow through a nozzle is determined
by
(1)
where Po is the backpressure, A* is the throat area, ρ is the gas density, To
is the gas temperature, R is the specific gas constant, and γ is the ratio of
specific heats. The theoretical and experimental pressure dependence of
the hydrogen sonic nozzle is shown in figure 2 below. The linear fit is well
within the error on the calibration data. The calibration is consistent with
theory when adjusting for the assumptions made. The linear fit is used to
determine the backpressure for the desired flow rate.
Figure 3: Plots of the applied voltage and temperature of the platinum tube
at three locations with respect to time. Two different trials are shown with
ϕ=.4 and .5. The arrows indicate manual voltage increases to increase the
tube temperature. The red circles indicate the spontaneous ignition
temperature.
Conclusions and Future Work
Pressure Gauge
Pressure Regulator and Sonic Nozzle
Figure 2: Pressure dependence plot of hydrogen flow rate
through choked nozzle including experimental calibration, linear
fit, and theoretical data plot
On/Off Valve
Thermocouple
Solenoid Valve
H2
Testing
Computer
O2
N2
Chamber
Power Supply
Figure 1: Schematic of test system including flow structure and
tubular reactor
A flow rate and equivalence ratio (ϕ) were set using the linear fits of
each gas to specify the backpressure. The flow was allowed to stabilize
then the tube temperature was incrementally increased by increasing the
voltage until ignition was observed. Ignition is observed by a large sudden
increase in temperature due to exothermic surface reactions. Three
thermocouples recorded temperature with number 1 being closest to
the tube entrance.
Initial tests of the tubular reactor show that ignition has occurred in a 14
cm long, .4 mm diameter platinum tube. Variation of the flow rate and
equivalence
ratio resulted in different ignition locations and
temperatures.
The system can accurately determine ignition conditions for various flow rates
and equivalence ratios. This system will be used to test the catalytic ignition
propensity of several metal materials that are likely to come into contact with
leaking hydrogen gas. The affect of residence time and equivalence ratio on
ignition temperature will be investigated to determine critical ignition
conditions. This data can then be used to help develop improved codes and
standards to increase the fire safety of hydrogen fuel cell vehicles.
Acknowledgements
I would like to thank the NIST Fire Safety Laboratory for funding this project. I would
also like to express my gratitude to Kyle Brady for his help on the design and
construction of the system.
References
1.J.L. Alcock, L.C. Shirvill, R.F. Cracknell, “ Compilation of Existing Safety Data on
Hydrogen and Comparative Fuels,” Deliverable Report EIHP2, (2001)