Abstract
A study was
carried out to evaluate the potential of hydrogen enrichment to increase the
tolerance of a stoichiometrically fuelled natural gas (NG) engine to high
levels of dilution by exhaust gas recirculation (EGR). This provides
significant gains in terms of exhaust emissions without the rapid reduction in
combustion stability typically seen when applying EGR to a methane-fuelled
engine. This paper gives the envelope of
benefits from hydrogen enrichment. In parallel the performance of a catalytic
exhaust-gas reforming reactor was studied in order that it could be used as an
onboard source of hydrogen-rich EGR. It was shown that sufficient hydrogen was
generated with currently available prototype catalysts to allow the engine, at
the operating points considered, to tolerate up to 25% EGR, while maintaining a
coefficient of variability of indicated mean effective pressure (IMEP) below
5%. This level of EGR gives a reduction in NO emissions greater than 80% .
Notation
CNG –
Compressed Natural Gas
COV –
Coefficient of Variation
EGR –
Exhaust Gas Recirculation
IMEP –
Indicated Mean Effective Pressure (indicated work per unit displacement
volume of engine, has units
of pressure)
NG –
Natural Gas
RON –
Research Octane Number
rpm –
revolutions per minute
TDC – Top
Dead Centre
Introduction
CNG as an alternative
vehicle fuel :
Protection of the environment and energy issues have
become increasingly important world-wide concerns with regard to internal
combustion engines. Natural gas, being a clean burning and plentiful resource,
is in many ways a good alternative fuel to meet these current and future
requirements. Typically, emissions of carbon monoxide, reactive (non-methane)
hydrocarbons and particulate matter are low, but emissions of oxides of
nitrogen have been seen to be relatively high.
The nature of a compressed natural gas (CNG) fuel system, being sealed
and utilising a gaseous fuel, tends to avoid problems associated with
evaporative emissions and cold start enrichment seen in gasoline engines. Additionally,
the total emissions of reactive organic gases from fuel storage and refuelling
associated with the use of CNG vehicles have been shown to be low . In
comparison to gasoline CNG has a low energy density, intake of air is reduced
(due to the feed of gaseous fuel into the intake manifold), and the flame speed
is also lower.
However, natural gas has high activation energy in
comparison to other hydrocarbon fuels. The high ignition temperature and
resistance to self-ignition result in excellent ‘antiknock’ properties; pure
methane has an equivalent research octane number (RON) of 130, and so natural
gas can safely be used in engines with higher compression ratios than are
possible with gasoline. Work at Toyota on a converted ‘Camry’ SI-engined model
has shown that many of the problems of low energy density can be overcome, with
the range being two-thirds of the petrol fuelled model for an increase in kerb
weight of 55kg.
Exhaust Gas
Recirculation (EGR) for NG engines :
It has been noted that
natural gas fuelled engines can produce undesirably high NOx emissions. However, NOx production is a
by-product of good, efficient combustion.
Thus, the concept explored
in this study involves the use of relatively high levels of EGR dilution as a
means of emissions reduction. The effect of EGR is to reduce both flame
temperature and flame speed. Oxides of
nitrogen form at high temperatures from nitrogen and free oxygen. The rate of
NO formation depends exponentially on temperature and is a function of oxygen availability.
In general, a given volume of exhaust gas has a greater effect on flame speed
and NOx emissions than the same quantity of excess air. For a given heat
release, the resultant temperature increase of a charge that contains EGR will
be smaller due to the fact that the average specific heat of the exhaust gas is
greater than that of air.
However, as with
excess air, there is a limit to the degree of EGR that can be used before
unacceptably poor combustion quality is encountered. This is a particularly significant problem
with natural gas engines, because the characteristics of the fuel make it
particularly difficult to burn lean or in high dilution by exhaust gas
recirculation. Thus the potential gains are severely limited by the rapid
decrease in combustion stability.
Hydrogen enrichment of Natural Gas :
The concept explored in this paper utilises hydrogen
enrichment to extend the tolerance of the natural gas engine to EGR, thus
allowing operation at high levels of EGR dilution whilst maintaining combustion
stability. It is suggested that the high flame speed and low flammability limit
of hydrogen have the effect, when added to the inlet charge, of increasing
flame speed, improving combustion stability and reducing burn duration in a
stoichiometric NG engine with high levels of EGR. The use of hydrogen
enrichment can open up engine operating conditions which would not otherwise be
viable, thus allowing improvements in emissions and efficiency through the use
of extended lean burn or high level EGR. The addition of hydrogen allowed a
significant extension in the lean operating limit of the engine, also reduced
emissions of NOx and an improvement in engine thermal efficiency, compared to
operation on gasoline.
Fuel Reforming :
The use of hydrogen enrichment requires a suitable
source of hydrogen on-board the vehicle. The concept tested here proposes a
system of on-board fuel reforming to generate a hydrogen-rich reformed EGR
stream.
Work at the Jet Propulsion Laboratory using a partial oxidation, process, or a
combination of partial oxidation and steam reforming processes, demonstrated
the use of forms of catalytic hydrogen generator to provide a hydrogen-rich gas
for the enrichment of gasoline fuelled engines ( Here rich combustion of a part
of the hydrocarbon fuel provides heat for the endothermic reaction. ).
Early works looked at the
processes of exhaust gas reforming of liquid automotive fuels (gasoline and
heptane) by direct contact with simulated engine exhaust gases. The aim there
was to study the feasibility of developing a system that would utilise the
waste heat in the engine exhaust to drive the endothermic reforming reactions.
The results obtained at high exhaust temperatures demonstrated the potential to
generate over 30% hydrogen in the reformed fuel and to increase its calorific
value by up to 28%.
The work presented here concerned the application of
fuel reforming in natural gas engines in view of the aforementioned problems
with the use of EGR in such engines. The
latter part of this paper presents the specific improvements obtained in
combustion quality by hydrogen addition and the potential of exhaust gas fuel
reforming to generate the necessary hydrogen.
Experimental Results
The experiments on single-cylinder
‘Medusa’ engine, largely based on an original design by Stone , and fitted with
one quarter of a Rover K16 engine cylinder head with the following
specifications -
Displacement: 446.86 cm 3
Bore: 80 mm
Stroke: 88.9 mm
Compression ratio: 10.5 : 1
Valves: 2 intake, 2 exhaust
Ignition controller: Intelligent Controls
IC5460.
Fuel 1 :
unleaded gasoline ( To take the baseline results ).
Fuel 2: Natural gas ( 85% CH4,
6% C2H6, 1.7% C3H8, 1.8% CO2,
6% N2 by volume).
Carburetor : Standard Gas Carburetor.
revealed the following
results – Figure 1 shows that the engine could tolerate levels of EGR up to about 8%, maintaining a COV of IMEP at 5% or less. Beyond this point the percentage COV is seen to increase more rapidly, and combustion quality is defined as unacceptable. It was found that extremely advanced ignition timing was necessary to achieve results in this region as burn duration increased with a similar pattern to that for the increase in percentage COV. Beyond 13% EGR it becomes impossible to run the engine without significant misfiring.
Figure 2 shows the corresponding reduction in NO emissions for this test. Similar results were obtained by the enrichment of the natural gas fuel with hydrogen.
A typical result set is presented in
Figure 3. The relationship appears to be approximately linear with an increased
proportion of hydrogen in the fuel producing a corresponding reduction in the
percentage COV.
In order to assess the feasibility of the
fuel reforming / reformed EGR concept it is necessary to know what proportion
of the EGR, on leaving the reforming reactor, must be hydrogen in order for it
to be sufficient to stabilise the engine running with a given EGR proportion.
The following figure is obtained based on experiments that illustrates the
amount of hydrogen in EGR at different
speeds of the engine.
It can be seen from Figure 4 that in order
to achieve a 5% COV of IMEP with an EGR proportion of 20%, the EGR stream would
have to contain a maximum of approx. 25% hydrogen, in the case of the 2000 rpm,
2 bar IMEP operating point, and a minimum of 5% hydrogen for the higher load
2000 rpm 4 bar IMEP operating point. These quantities are based on adoption in
both cases of fixed ignition timing
strategy. It is clear from the catalyst test results presented in the following
part of the paper, that hydrogen content of 20% in the reformed EGR is
currently achievable as a conservative estimate. This would allow the engine to
operate, at any of the operating points considered, with 25% EGR, which has
been seen to give significant benefits in terms of engine-out emissions.
Figure 5 shows engine-out total
hydrocarbon emissions versus percentage EGR .
Here COV of IMEP is less than 5% following
hydrogen enrichment. There is a general trend for engine-out hydrocarbons to
increase as EGR is increased, the increase being most significant in the 4bar
IMEP higher load case Here the major sources of hydrocarbon emissions are
combustion chamber crevices. Results obtained with a high -speed HC analyser suggested
that it is the rate of oxidation of these crevice-bound hydrocarbons late in
the cycle that determines the actual level of exhaust hydrocarbons. This
indicates that total hydrocarbon emissions decrease as exhaust temperature
raises, the rate of oxidation being partially dependant upon the in-cylinder
temperature. It should be noted that there are apparently anomalous points in
the data presented here, showing some reduction in hydrocarbon emissions, this
perhaps illustrates the difficulty of maintaining a stoichiometric mixture with
hydrogen enrichment without on-line monitoring of hydrogen emissions.
Figure 6 shows engine-out exhaust gas
temperature (approx. 200 mm downstream of
exhaust valve) versus percentage EGR .
Here a slight trend for a reduction in exhaust gas temperature as EGR and
hydrogen addition levels are increased, most significantly in the high load
case, which is consistent with the argument presented above. That is the lower
temperature gives a lower oxidation rate and thus a higher level of engine-out
hydrocarbons.
Conclusions
§ The
use of EGR with addition of reformed fuel could potentially offer significant
emissions improvements.
§ Tolerance
of the NG engine to EGR, as measured in terms of combustion stability was shown
to be greatly extended by addition of main components ( hydrogen and carbon
monoxide) of a reformed fuel.
§ The
proportion of hydrogen or hydrogen – carbon monoxide mixture required to
maintain combustion stability has been quantified for a range of operating
points.
§ Certain
catalysts can be used to produce a reformed fuel of the required composition
(over 20% hydrogen) from exhaust gases with natural gas added.
References
1. Kato
K, Igarashi K, Masuda M et al:
Development of an Engine for Natural Gas Vehicle. SAE Paper 1999-01-0574, Society of Automotive Engineers, 1999.
2. Heywood, JB, Internal
Combustion Engine Fundamentals, (McGraw-Hill, NewYork), ISBN 0-07-100499-8,
1988.
EmoticonEmoticon