Journal of Scientifi c & Industri al Research Vol. 62 , January-Febru ary 2003. pp 64-70

Natural Gas/Hydrogen Mixtures for Low Noxious Emissions

1 R Roger

University of Gent , Laboratory of 'T'ransporttechnology, Sint-Pietersnieuwstraat B-9000 Gent , Belgium

Adding hydrogen to natural gas ex tends the lean limit of combustion and ex tremely low emission levels can he obtained. Even the equivalent zero emission vehic le (EZEV) requirement s can be reached . The emi ss ions red ucti on is especiall y important at light engine loads. In thi s study, tes ts have been carried out with natural gas, pure hydrogen, and different bl ends of these two fu els. The fuel suppl y system provides natural gas/hydrogen mixtures in va ri able proporti on, regul ated independentl y of the engine operating condition . The inlluence of the fu el compositi on on the engine characteri sti cs and on the exhaust emi ssions has been examined, mainl y but not exc lusively for 10 and 20 per cent hydrogen addi ti on. It is shown that to obtain max imum engine efliciency for the whole load ran ge whi le taking low ex haust emissions into account, the mixture compositi on shou ld be varied with respect to engine load.

1 Introduction

Hydrogen-enriched natural gas has been the subj ect of severa l research projects. These mixtures of natural gas and hydrogen are commonl y named ' Hythane' (a reg iste red trademark of Hydrogen Consu ltants Inc.).

Main point of inte rest is that with the addition of hydrogen, the lea n limit of natural gas operation can be extended , without going into the lean misfire region. This results in low and even ex treme ly low NO, leve ls with only a s li ght increase in hydrocarbons. The low exhau st e mi ss ion levels are obtained without emi ss ion contro l equ ipme nt (w ithout a catalytic converte r). The CO and CO2

va lues are lower for any gaseous fue l compared to gaso line, and for hydrogen no CO or CO2 is formed at a ll (from the fue l itse lf).

Although hydrogen is an a lternative fue l with very c lean burning characte risti cs, a high flame propagation speed and w ide flammability limits, it also has disadvantages. The compl ex ity and we ight of hydrogen storage, the loss of power associated with the use of pure hydrogen and the backfire pheno menon are the most important ones. The addition of natural gas to hydrogen (al so hythane, bu t with a high percentage of hydrogen) can solve the backfire proble m. In many cases, backfire restricts the operating regIOn of the a ir-fue l mixture on the

' rich ' side. With natural gas additi on sto ichiometri c mixtures can be run without any othe r precauti ons.

The proposition of an Equiva le nt Zero Emiss ion Vehic le (EZEV) at 10 per cent of the 1997 U LEV (u ltra low emiss ion vehicle) requ ireme nts by the California Air Resources Board (CARB) has encouraged the research in lean burn hyd rogen o r hythane spark ignited eng ines .

In the lite rature l.'), simi lar trends are found

using hythane blend s. On comparing the results in detail , one has to keep in mind that the compos iti on of natural gas can be quite diffe rent (for diffe rent countries). The methane content can change from 90 to 98 per cent , with sometimes hi gh nitrogen concentrati ons ( I to 8 pe r cent ). Thi s w ill of course influence the ex perimental results.

2 Description of the Test RIG

2. 1 Engine

A Crusader T7400 spark ignit d eng ine (based on the GM 454 engine, best known as the Chevrol e t Big Block) was adapted for gaseous fu e ls .

The engine specifications are:

- 8 cylinders in V -bore: 107 .95 mm - stroke : 101.60 mm - swept volume: 7.4 I (454 in 3

)

ROGER: NATURAL GAS/HYDROGEN MIXTURES FOR LOW NOXIOUS EM ISS IONS 65

- compress ion ratio : 8.5: 1 - engine speed : 1000 - 4500 rpm - ignition sequence: 18436572.

The engine IS connected to a water (Froude) brake.

The gas (natural gas, hydrogen or hythane) is mixed with the air in a gas carburettor (see 2.2). The venturi of the gas carburettor is s lightly under­dimensioned , which causes a lower volumetric efficiency and some power loss. For the compari son of the different fuels, thi s has no importance.

The composition of the natural gas consists of 9 1 per cent methane, 6 per cent ethane, 1.5 per cent propane, 0.8 per cent nitrogen and smaller fractions of higher hydrocarbons. The ignition is done by a s ingle-firin g system (one spark for each cycle of 720°c a) which is necessary to avoid backfire when using hydrogen. The ignition timing is regulated by a spec ial disc on the di st ributor.

2.2 The Fuel Supply Sysrem

The layout of the sys tem is shown in Figure 1. The fuel is delivered to the engine through a gas venturi , which is supplied with the fuel mixture at a slight overpressure. The richness is controlled using a control valve in the supply line.

Both, pure hydrogen and natural gas, as well as mixtures of these fuels can be used with this system. The hydrogen is stored in nine steel bottles at a pressure of 200 bar; the natural gas is obtained from the city gas net at 30 mbar overpressure. Measurement of the fuel flow rate is made using mass flow meters in both of the supply lines.

If hythane is used the hydrogen flow is controlled by a mass flow controller (the same device is used as mass flow meter for pure hydrogen). From the measured natural gas flow the necessary hydrogen fl ow is computed and supplied as input signal to the mass flow controller. Thi s results in a constant hydrogen content, independent of engine speed and load . The hydrogen concentration is given in volume per cent (the term mass flow meter/controller only means that the measurement is automatically compensated for temperature and pressure changes the reading is in Nm3/h). Alternative ly, hythane (or any other fuel) from a high pressure tank (200 bar) can be used for short runs: it contains a very limited amount of hythane and

Venturi

Mass flow meter Natural

!I---O'<l--- gas

Non-return valve supply

r---------

Oplional

hythane

lank

Control unit

c)l<] Valve (manual contro~

£ Eledromegnelic valve

o ZertJ pressunI l'8Qulator

Mass Flow Controller

Hydrogen

consul'TlPl ion

measurement

bailie

(oplional)

Hydrogen bottles (9)

Figure I - The fuel preparation system

freezing problems can occur. By putting thi s tank on a scale fuel consumption can be measured .

The system described above is onl y able to provide mixtures with up to 67 per cent hydrogen, due to limitations in the control unit. For higher hydrogen content the mass fl ow controller has to be regulated independently of the natural gas flow or opened completely. Doing the latter results in up . to 79 per cent hydrogen . For even higher hydrogen fractions the natural gas flow must be throttled.

2.3 Appararus

The engine is fully equipped with the usua l sensors. The measurement/control s ignal s are read and controlled by a PLC system (Programmable Logic Controller). Thi s system monitors engine speed, oil and coolant temperature, exhaust gas temperatures, etc . and shuts off the engine when necessary. With a Microsoft Excel worksheet all values are stored and can be made visible on a screen.

The exhaust temperatures and the exhaust gas composition can be measured at the exhaust of each cylinder and at the end of each bank (V engine). Two

66 J SCIIND RES VOL 62 JANUARY-FEBRUARY 2003

O2 (A)-sensors are installed at the common ex haust pipe of each bank , wh ich allows an immediate value of the air-fue l ratio of each bank . The A-sensors and the ex haust temperatures give the possibility to check if all cy linders behave similarly.

The ex haust gas components are measured with the follow ing methods of measurement : CO - CO2 -

NO - NO] (multor 610, non-di spersive infra red); O2

(servomex model OA 11 00, paramagneti c), HC (Signal model 3000, flame ioni zat ion).

A hi gh pressure transducer (type A VL QC 32) is located in one cy linder head (mounted flush wi th the combusti on chamber wa ll of cy linder I) and is used for the ca lculati on of the heat rel ease.

3 Experimental Results and Discussion

All measurements were made at 3800 rpm, with the throttle wide open (WOT). These conditi ons are chosen because at tha t engine speed, maximum power is reached. No other throttle positions were used because for hythane and hydrogen the best efficiency is achieved when regu lating power through adjustment of the air excess rati o instead of the inlet pressure (except for the very low loads). Thi s is not the case for natural gas because its combustion becomes so slow at lean mixtures that effi ciency suffers and even misfire occurs below a certain equival ence rati o (thi s depends, to great ex tent , on the engine considered, however). Therefore the lean limit as sllch has not been in vestigated here.

The measurements are anal ysed with respect to mean effec ti ve pressure, in stead of power, to make the results less dependent on engine size and speed. For a similar reason, ex haust emiss ions are represented in g/kWh instead of ppm or vo lume per cent : thi s way the dependence on effective engine power is avo ided. More spec ifica ll y, thi s also avoids the effect of the dilut ion with air of the nox ious emissions in lean mi xtures.

3. 1 Natllral Gas alld Hythan c Ivith 10 alld 20 Volume

Pcr Cellt Hydrogen

3.1.1 Brake Mean Effecti ve Pressure, Fuel Consumption, Thermal Efficiency, Volumetric Efficiency And Ignit ion Advance

In a first set of measurements, a comparison is made between natural gas and two blends of hythane (hythane wi th 10 and 20 volume per cent hydrogen, respective ly). Each time, spark timing was optimised for maximum power (MET).

Figure 2 shows how the air excess factor <I> or air-fuel ratio A (A = 1/<1» influences bmep. As is evident the different fuel mixtures gi ve very . imi jar

results: the on ly major difference is the abi lity of hythane to run leaner the more so the hi gher the hydrogen content. The A range is also determined by the fuel supply system: on the rich side the suppl y pressure of the natura l gas is a limiting factor. For pure natural gas the lean limit gives a large reducti on in engine power (no significant benefit is to be ex pected from such lean running, however: low effi ciency and large hydrocarbon emi ss ions).

The specific fuel consumption be (ex pressed in g/kWh) for the same condi ti ons is shown in Figure 3. A A-value between 1.1 and 1.2 gives best efficiency for each of the fuels. Because of the very different lower heat va lues of combusti on (for hydrogen 120 mJ/kg and for methane 50 mJ/kg), effi ciency compari sons shou ld be made using Figure 4. This shows that a hydrogen addition of 10 per cent increases efficiency moderately, whereas 20 per cent hydrogen gives no significant ex tra benefit (for the

7r-------------------------------~

D O~. H2 0 IO~. H;

"' 6 1< 20~.H; ,

0 Q

Q.

" E 5 Q

~~---L----~--~----~----~--~--~ I 1.1 1.2 U U 1.5 1.6 1.7

). (-)

Figure 2 - The infl uence of· the air excess racto r on hmep

250.------------------------------, c 0 "/. HZ 010 "/. HZ

~.20%H2

220

210 L-__ -1...-__ --' ____ ....L.-__ ---1. ____ ..l..-__ ---1. __ ----1

7 1.1 1.2 1.3 1. 4 7.5 7.6 1.7 ).,( - )

Figure 3 - The influence or the air excess fac tor on the speciri c ruel consumpti on

, ,

ROGER: NATURAL GAS/HYDROGEN MIXTURES FOR LOW NOX IOUS EM ISS IONS 67

.. '"

0.31

0.37

0.30

0)9

0.28 I 1.1 1.2 1.3 /.I. 7.5 1.6

)..{-)

Figure 4 - The inlluence of the air excess factor on the elTiciency

1.7

same A-va lue). This result is in agreement w ith Bell

, ,------------------------------------,

I> O"/.Hz o 10"1. HZ

/( ZO"/. H2

OL_ __ -L ____ L-__ ~ ____ ~ __ ~L_ __ _L __ ~

I 1.1 I.Z 1.3 1.1. 1.5 1.5 1.7

).. (-)

Figure 5 - The hydrocarbon emissions as a function of the air excess factor

and Gupta who also found a signi fican tl y hi gher 10.

e fficiency whe n adding 10 per cent hyd rogen, and no fu rther improvement when going to 15 per cent hydrogen.

3.1.2 Emissions

The hydrocarbon (UHC) and the NOx emi ss ions )- as fu ncti ons of the air excess fac tor are shown in

Figure 5 and 6. Minimum hydrocarbon and max imum NOx emi ss ions are found for an air excess factor of about 1. 1. Thi s result is in agreement with the literature, and ho lds for any fue l (see for example Heywood'-l for gaso line and Be ll and Gupta9 for

hythane).

For leaner mi xtures the combusti on temperature tS lower because of the lower heat of combu stion ava il ab le in the mi xture, which reduces NOx, and for richer mi xtures less oxygen is ava ilable for the formation of NO,. The minimum in unburned hydrocarbon emiss ions in this range is caused by the imposs ibility of complete combustion on the rich side (local/g lobal lack of oxygen) and on the lean side the fl ame ex tingui shes further from the wa ll s resu lting in more unburned emiss ions from wa ll s, and crev ices. For even leaner mi xtures the fl ame may be unable to trave l through the whole combust ion chamber and in some cases the spark may be unable to ignite the mixture at a ll.

For comparison at constant bmep, Figure 7 and 8 are g iven. These figures show that 'at the same bmep' hydrogen addition inc reases NOx and dec reases unburned hydrocarbons, irrespective of bmep level. Thi s result seems to di sagree with results from the literature, which c laim lower NOx emi ss ions

"' D 0"1. HZ ~ o TO 0/. HZ ~ IZ ~ )( ZO% HZ ,.

B ~ lOo.%HZ 0 <:

~

0 I.Z I.~ 1.6 1.8 2.0 2.Z Z/.

l. (-)

Figure 6 - The NO" emission

~

~ 3 I> o "I. HZ

~ 10 %HZ ~ ~ 20."1. HZ 0. 1 '-

l.J :t. :::>

o.L-____ L-____ ~ ____ ~ ____ ~ ____ ~ ____ ~

I. 5 55 6 6.5 7

bmrp (bar)

Figure 7 - The hydrocarbon emissions and bmep

when adding hydrogen. On in spection , however, it becomes clear that they compare natura l gas with hythane with the latter running leaner (see fo r example Be ll and Gupta\ or with a reduced spark advance (see for example Raman el al.\ Both these measures have the di sadvantage of IIlcreas lllg

68 J SCIIND RES VOL 62 JANUARY-FEBRUARY 2003

20~---------------

5

o 0'/. HZ () 10"1. H2

1< 20"1. H2

+ 100 "l.H2

J 3,5 5 5.5 6 bmtp (bar)

Figure 8 - The NO, emission and bmep

65

unburned hydrocarbon emissions as well as decreasing efficiency. Also the NO, emissions can be reduced by retarding the spark from MBT fo r any fuel and therefore also for pure natu ral gas .

3.2 Hythane with 50, 67, 84 Volume Per Cent Hydrogen and Pure Hydrogen

In a second set of tests, a preliminary in vesti gati on was made on ~ he use of hi gher hydrogen fracti ons. Due to time constraints (hi gh hydrogen consumption) the igniti on advance was not optimised for each conditi on in thi s set of measurements , but was kept constant at 20°C before TOe, except fo r pure hydrogen, where the optimal spark advance was used (which varies from 12° at A = 1.5 to 22° at A = 3.5).

For pure hydrogen the engine could functi on reli abl y (without backfire) fo r A ~ 1.5. Hythane with less than 79 per cent hydrogen never causes backfire nor knock in thi s engine, whatever the ri chness . During the measurements with 84 per cent hydrogen knock (without backfire) occurred fo r A = 1.06. Thi s suggests that a hydrogen content of 80 per cent or less guarantees safe operati on of the engine, whatever the air excess fac tor.

In Figure 9 the bmep achi eved by pure hydrogen and hythane with 50, 67, and 84 per cent hydrogen is shown. Remarkably all hyth ane mi xtures give a very comparable output fo r A between I and 1.3. This is in agreement with the results presented by Raman ef al. s, who compared pure natural gas with hythane (30 per cent hydrogen) and found the same bmep at A = 1.3 and only 2 per cent bmep loss at stoi­chiometry with hythane. The reason for thi s

2-E .Q

7,--------------------------------

3

1,5

o 50"1. HZ () 67"1. HZ x 8~ "I. H2

+ 100'1, H2

3 ). ( - )

Figure 9 - Thc bmep for hi gh hydrogen conccntrati ons

behaviour is that increase in the hydrogen content increases fl ame speed (increasing bmep) but decreases the volumetric energy content (for A < 1.43) (decreas ing bmep)R) . In the A region considered, both effects compensate each other large ly. For A > 1.43 the vo lumetric energy content as well as the fl ame speed increase with hi gher hydrogen contents, which causes hi gher bmep fo r those mi xtures. At these air excess rat ios the bmep drops significantl y though.

For strongly lean mixtures, pure hyd rogen gives the highest output , as it is the onl y fue l capable of fas t combusti on at these conditions.

4 The Low Emission Potential of Hythane

Most, if not all , of the literature about the use of hythane concern s the possibility to ach ieve extremely low emiss ions. We will now in vestigate how thi s can be done.

Assuming that an ultra low emi. sion vehicle (ULEV) with a fuel consumption equi va lent to 10 L gaso line /100 km (fu el density of 0.75 kg/L) during the tes t cyc le has to be des igned the emi ss ions performance of the engine can be estimated. If we take the average engine effi ciency equal to 20 per cent , the ULEV limits ( 1.7 g CO/mile, 0.04 g NMOG/mile and 0.2 g NOJ mile) convert to an average engine emiss ion level of no more th an 15 g CO/kWh, 0.35 g NMOG/kWh and 1.75 g NOJkWh. These average engine emi ss ion leve ls are roughl y calcul ated as follows: 10 11100 km x 0.75 kg/L x 0.20 Ole) X 43.5 106 l/kg (Hu).(3600 101rl kWh/J gives 0.18 1 kWh/km. For CO one finds 1.7 g/mile x 1.609 mile/km/0.18 1 k Wh/km == 15 g/k Who

"

ROGER: NATURAL GAS/HYDROGEN MIXTURES FOR LOW NOXIOUS EMISSIONS 69

Looking at the emi ss ion graphs, it seems that the hydrocarbon emiss ions are always too hi gh. However, because the unburned hydrocarbons (UHC's) originate mainl y from the fuel and more specificall y from the natural gas frac ti on of the fuel, they are mostl y methane. Because the ULEV limits concern non-methane organic gases (NMOG; methane is not relevant with respect to the formati on of tropospheric ozone) , thi s improves the chances to comply with the limits. Unfortunately the exhaust gas analys is equipment avail able for the experiments described here, did not all ow the measurement of NMOG directl y.

The CO emiss ions from the tested engll1e are very low, as is the case for any gaseous fuel, and therefore does not cause mayor problems when trying to compl y with the limits.

As can be seen from Figure 6, the NO, emiss ions are only low enough for very lean hydrogen and lean hythane mi xtures . In the case of hythane the hi gher the hydrogen fracti on the leaner the engine has to run and the lower the achievable bmep. At these conditi ons the engine produces a lot of UHC (Figure 5), whi ch makes it extremely unlikely th at the NMOG limit can be attained without exhaust gas treatment. The use of gives no problem with NMOG, negli gible amount of UHC IS

hydrocarbons in the fuel).

pure hydrogen since onl y a

emitted (no

For lean mi xtures, it is however poss ible to reduce CO and UHC (and therefore NMOG) emi ssions by using an ox idation catalyst. If we take in to account that the NMOG fracti on is reduced more . ~ t rong l y th an the methane frac ti on, it is very probable that in thi s way the NMOG limits can be fulfill ed.

Therefore the fo ll owing strategy to run the engine can be proposed.

• At low loads, pure hydrogen is used (A. > 2), limiting NO, and eliminating NMOG and CO emi ss ions.

• At intermedi ate loads, hythane is used in such a way that NO, remains low. The ox idati on catalys t reduces CO and NMOG emi ss ions to acceptable levels.

• At full load (neady) pure natural gas IS used, providing hi gh max imum bmep.

Although thi s means hi gh emiss ions at full load conditi ons, this is not a major concern with respect to the legal limits since these conditi ons do not occur in the tes t cycles used to measure emiss ions. Also, during everyday vehicle use the engine is mainl y operated at parti al load, reducing the need to optimise emissions at full load. If low emi ss ions are to be reached at all operating conditi ons, a three way catalyst may be used with A. = I operati on at hi gh loads. This necess itates the use of a throttl e to regulate power in the high load region and is unpractical without a dri ve by wire system. A less complicated solution would be to make the hi o-h load b

region simply unava il able for the dri ver (not an acceptable so lution to the end user).

-.. I ......

Such a system has the foll owing advantages:

• High effi ciency - WOT is used at all but the lowest loads, avo iding throttling losses . Figure 4 and 10 show that near optimal effi ciency IS

reached. Spark advance is set at MBT.

• Low emiss ions.

• Low knock sensiti vity - At hi gh loads pure natural gas with a hi gh NO, is used; at lower loads lean running avo ids knock. Thi s opens the poss ibility for hi gh compress ion rati os, wh ich increase effi ciency.

• Backfire is avo ided - Since hydrogen is not used at A. < 1.5 , backfire can be avo ided withou t taking any spec ial precauti ons.

• The di sadvantages are:

• Weight and vo lume - Two fuels must be stored .

O.32r--------------::-:-:--- -. a sao;. H2 Q 67·/. H2 >< 8""/. H] ~ IOO"loH2

.. r::--O.25

aI8·~---L----~---~ _ _ ~ 1 1.5 2 2.5

..,.. (-)

Figure 10 - The effi c iency fo r hi gh hydrogen con ce ilirati on ~

J

70 J SCIIND RES VOL 62 JANUA RY-FEB RU AR Y 2003

• Complexity and cost - A mixing system is necessary to control the hydrogen/natural gas ratio .

5 Conclusions

A fu el supply system was des igned and implemented that provides the engme with a hydrogen/natural gas mixture in a variable proportion. The composition of thi s mixture can be set independently of engine operation. For future practical applications, minimal modifications should allow this variable mixture composition with respect to engine load for maximum efficiency and minimal poll uti on .

For hythane with low hydrogen content (up to 20 per cent) a limited improvement in emissions can be obtained. Because of the conflicting requirements for low hydrocarbons and low NO, extremely low emiss ions are not poss ible without exhaust after treatment: To reduce hydrocarbon emiss ions Ie must be less than 1.3, while for low NO, Ie must be at least 1.5 .

For lower bmep, high efficiency can be achieved by increasing the hydrogen content and thus avoiding throttling losses. At the same time unburned hydrocarbon emissions are minimized, while (for lean mixtures) NO, emi , sions stay limited. This means that for optimal resu lts the composition of the fuel should depend on load .

References

Hoekstra R L, Collier K, Mulligan N & Chew L, Experiment al stud y of a clean burning vehi cle fuel , 1111 J Hydrogell Eller, 20 (NR 9) ( 1995) 737-745.

2 Hoekstra R L. Colli er K & Mulligan N, Demonstration of hydrogen mi xed gas vehicl es , Tellth World Hwlrog Eller COli , Cocoa Beach, Proc, 3 ( 1994) 178 1-1796.

3 Hoekstra R L, Van Blarigan P & Mulligan N. NO, emissions and e.fjiciell cy of hwlrog ell , n{l{{(ral gas (lnt! hvdrogelli ll atllral gas hlelldedIllels, SAE Paper 96 1 103.

4 Swain M R, Yusuf M, Dulger Z & Swain :vI N, The e.fji'cts oI hwlrogen additioll 0 11 lIatllral gas ellgine operatioll . SAE Paper 932775.

5 Yusuf M, Swain M R, Swain M N & D(i1ger Z, An approac h to lean burn natural gas fu ell ed engine through hydrogen addition, Thirteellth ISATA Coni; Paper 97EL08 1. Florence, June 1997.

6 Larsen J F & Wall ace J S, Comparison of emissions and efllcien cy of a turbocharged lean-burn natural gas and hythane-fuelled engine, Trails ASME, 119 ( 1997) 21 8-226.

7 Cat tel an A I & Wallace J S, Hythane and CNG fu ell ed engine ex haust emission and engine elTiciency comparison, Tellth World Hydrogell Ell er COllI Beach Proc, 3 ( 1994) 1761 - 1770.

8 Raman V, Hansel J, Fulton J, Lynch F & Bruderly D. Hythane - an ultra clean transportation fu el, Tenth World Hydrogell EneI' Cor~r Cocoa Beach Proc, 3 (1994) 1797-1806.

9 Bell S R & Gupta M, Ex tension of the lean operating limit for natural gas fuelli ng of a spark ignited engine using hydrogen blending, Comhllst Sci Technol, 123, ( 1997) 23-48.

10 Dhooge D & Pussemier B, Omhorll vell vall een GM-II f{)to r tot een \IIatersto./inotor (Converting a GM-engine into a Hydrogen Engine) , End of year Thesi", Laboratory for Machines, Un iversity of Gent , 1997 .

I I Sierens R, Comparati ve tests on a SI engine fuelled with natural gas or hydrogen, ASME Winter Meering HOllstOIl 93, ASMEIICE Engille Syrllp Eiler-SOliI' Techno/ COlli; Paper No. 93 ICE- I 5.

12 Milton B E & Keck J C, Laminar /)lIrnirrg velocities ill stoichiometric h.l'drogell (llId hydrogell-!rw lrocarholl g{ls mixtllres.

13 Karim G A, Hydrogen as an additi ve to methane for engine appli cati ons. Elevellth World Hrdrogell EneI' COllI Stllllgart, Proc, 2 ( 1996), 1921- 1926.

14 Heywood J, illlem al cOlllbllstion engille ./illrdallrenwls (McGraw- Hili , Bork Co., Inc. New York ) 1988.

Prof Sierells Roger has doctorate ill Applied Science, Uni,'ersitv oI Gellt. He is present;r working as ({ professor irr FaCIlity (~f Applied Sciences at the Un iversity of Gellt. He is also Director of the La/Joratorr of Trarr sport TechrlOlogv (RUG), Chainrw lI oI the Mechanical Eng in eerillg Education COIIIIII ill el' and par-tillle proIessor a t the Ull iversity oI AlltlVerp (UFSIA). He has mallY scienti./ic a IVa rd.,· to his credit and has pllblished 1110 1'1' than 100 papers in intemational jOllmals 0 11 IC engines. His areas (4 interest are sillllliation of the gasdvnarnic alld thenrwdyn({/nic cycle oI IC engilles (diesel, recillrocatillg) and rorar." SI engines. He has lVorked on altem ative /irels, lIatllral gas, liqllid LPC, and hydrogen.