Currently biofuel has attracted a lot of interests because of the energy crisis. DMF, known as 2,5-Dimethylfruan, came into researchers' eyes when its improved production methods were published in Nature and Science in 2007. It can become a possible alternative to gasoline for its high energy density. The aims of this project are to do some study on DMF and make a comparison between DMF, ethanol and gasoline of their various properties in different tests.
DMF (2,5-Dimethylfuran, liquid state under normal temperature) is a heterocyclic compound with the formula (CH3)2C4H2O. Its molecular structure is shown below.[1]
Figure 1 The molecular structure of DMF
Some of DMF's physical and chemical properties make it competitive to ethanol, another biofuel as well. For example, DMF's energy density is 31.5 MJ/L, higher than ethanol's value 23MJ/L, and closer to gasoline of which the value is 32.2MJ/L. Besides, DMF, unlike ethanol, is insoluble in water, which makes it which makes it stable in storage and unlikely to be contaminated by water in transportation pipelines or contaminate underground supplies. And with its higher boiling point(92℃) than ethanol(78℃), it is proved to be less volatile more practical for transportation as a liquid biofuel.[2]
The table below shows the basic properties of DMF, ethanol and gasoline.[3]
properties
DMF
ethanol
gasoline
Molecular formula
C6H8O
C2H6O
C2-C14
Molecular mass(kg/kmol)
96.13
46.07
100-150
Density at 20℃(kg/m3)
889.7
790.9
744.6
Water solubility at 25℃(mg/mL)
Insoluble,≤1.47
Highly soluble,≥100
Insoluble
Gravimetric oxygen content (%)
16.67
34.78
0
H/C ratio
1.33
3.00
1.865
O/C ratio
0.17
0.5
0
Stoichiometric air/fuel ratio
10.72
8.95
14.56
Gravimetric calorific value (LCV, liquid fuel)(MJ/kg)
33.7
26.9
43.2
Volumetric calorific value (LCV, liquid fuel)(MJ/L)
30
21.3
32.2
Research octane number (RON)
119
110
95.8
Auto-ignition temperature(℃)
285.85
423
257
latent heat of vaporization at 20℃(kJ/mol)
31.91
43.25
38.51
Heat of vaporization (kJ/kg)
332
840
373
Initial boiling point (℃)
92
78.4
32.8
Price (₤/L)
30-100
0.61-0.66
1.3
Table 1 Basic properties of DMF, ethanol and gasoline
Comparisons between DMF, ethanol and gasoline in different tests
Nomenclature
ATDC = after top dead center BTDC=before top dead center
CAD = crank angle degree CO=carbon monoxide
COV = coefficient of variation CPC= condensation particle counter
DI= direct injection DMA =differential mobility analyzer
DMF = 2,5-dimethylfuran EGR =exhaust gas recirculation
EVC =exhaust valve closing HC=hydrocarbon
IMEP= indicated mean effective pressure IVO= intake valve opening
LCV =lower calorific value MFB = mass fraction burned
NOx = oxides of nitrogen PM= particulate matter
RON = research octane number rpm = revolutions per minute
SMPS= scanning mobility particle sizer SOI=start of injection
TDC = top dead center λ= excess air ratio
2.1 Literature research section
The results of the experiments in this section are from past papers.
2.1.1Combustion and Emissions of DMF in a Direct-Injection Spark-Ignition Engine [4]
The engine for the test is a single-cylinder, four stroke, spark ignition, direct injection (DI) engine.
The information of the engine is shown in the figure and table below.
Engine Geometry and Valve Event Details
geometry details
valve timing details
engine type
four stroke, four valve
intake valve lift (mm)
swept volume (cm3)
565.6
intake valve duration
stroke (mm)
88.9
exhaust valve lift (mm)
bore (mm)
90
exhaust valve duration
connecting rod length (mm)
160
IVO
piston offset (mm)
0.6
EVC
compression ratio
11.5
spark timing
fueling type
spray-guided DI
fuel injection pressure and timing
Table 2
Figure 2 the Medusa Engine
Experimental setup
The engine is fitted with a Direct-Injection (DI) fuel system.
Fuel delivering pressure: 150bar (using bottled nitrogen).
Fuel injection pressure is measured by a fuel-line pressure meter.
Nitrogen is used to minimize contamination when changing fuels.
Inlet system: an air filter, a rotary gas volumetric flow meter and a manually adjusted throttle.
A Kistler-type 6125A pressure transducer is used to measure the cylinder pressure.
Samples taken condition: 0.5 degrees CAD (crank angle degree) intervals, 100 consecutive cycles.
A digital shaft encoder is used to measure the crankshaft position.
Oil temperature: 95±3℃, coolant temperature: 85℃
The engine-operating parameters are set as the figure shown below (valve, injection and spark timing strategy).
Figure 3
Engine speed: 1500rpm
Equipment for gaseous emissions measurement: a HORIBA MEXA 7100DEGR gas analyzer, a heated line and prefilter.
Gas samples taken position: 0.3m downstream of exhaust valve.
Equipment for air/fuel ratio measurement: an ETAS LA4λmeter and a Bosch LSUλ sensor.
Particulate Sampling System: a model 3936 scanning mobility particle sizer, measuring size range: 7.23-294.3nm; dilution ratio: 100:1
Test fuels: RON=95 gasoline, bioethanol and DMF (99.8%)
Inlet air temperature: 25℃
Air/fuel ratio: 0.1
Results and discussion
All testing results graphs are in appendix (figure A1-A10). According to the graphs:
From figure A1 (a) it's clear to see that the fuel rate of ethanol is about 33% more that of gasoline, while DMF is very close to gasoline.
From figure A1 (b) it's clear to see that the initial combustion duration of DMF is lower than that of ethanol and gasoline for the range of IMEPs tested.
In figure A2 (a), the initial combustion duration of DMF is between ethanol and gasoline (ethanol < DMF < gasoline) when spark timing is from 24° and 44° at 3 bar IMEP.
In figure A2 (b), the spark timing remains at 34° BTDC. It's clear to see that the combustion duration of DMF and gasoline decreases more rapidly than ethanol respect to load. And compare to gasoline, the combustion duration of DMF deceases relatively slower than that of gasoline.
From figure A3 (a), it's clear to see that the indicated efficiency of ethanol is the highest of all, while DMF and gasoline are similar to each other.
Figure A3 (b) shows the volumetric efficiency respect to load. Volumetric efficiency is the volume flow rate of air into the cylinder for a given volume displacement rate of piston. it's clear to see that the efficiency of DMF is higher than ethanol and similar to gasoline.
Figure A4 shows that the pumping loss of ethanol is bigger than those of DMF and gasoline, which are very similar to each other.
In figure A5, the results of DMF and gasoline are very close to each other, while ethanol has the lowest values for both maximum in-cylinder temperature and maximum in-cylinder pressure, which increase its burning duration.
The knock tendency of DMF is lower than that of gasoline because of its higher RON. And as load increases, the severity of knock increases as well. After 6.5 bar, as shown in figure A6, the engine becomes unstable. On the other hand, ethanol does not cause knock at any load, which means the combustion of ethanol is more stable even its RON is 110 lower than DMF of 119, as shown figure A7.
From figure A8 (a), it's clear to see that the levels of CO emission for DMF and gasoline have a similar trend, while ethanol has a lower value. And compared with gasoline, the level of DMF is relatively lower. In figure A8 (b), the level of HC emission for DMF is lower than that of gasoline. And ethanol is much lower than the other two. However, as load increases, the difference becomes smaller. As for the NOx emission level, ethanol still has the lowest value while DMF and gasoline are similar to each other. However, the level of gasoline increases more rapidly when load is increased.
In figure A9 (a), it's clear to see that at 3.5 bar IMEP, the three fuels have a similar distribution of particulates. And most of them are in nucleation mode and the peak is between 20 and 30nm. When increasing the IMEP to 5.5 bar, as shown in figure A9 (b), the number of nucleation mode particles decreases while the number of accumulation mode particles increases. This is more distinct when using DMF and gasoline as fuels. The reason for this is because the diameter of nucleation mode particles peak becomes smaller, as shown in figure A9 (c). On the other hand, more accumulation mode particles formed another peak. And in figure A9 (d), it is clear that the combustion of gasoline produces the highest number of accumulation mode particles among the three fuels, especially when IMEP = 5.5 bar.
Ethanol produces the lowest PM emission compared to the other two according to figure 10A (a) while DMF the highest. And according to figure 10A (b), the combustion of gasoline produces the largest mean diameter of particles while DMF the smallest.
Conclusion
The initial combustion duration of DMF is shorter than that of gasoline. And when compare to ethanol, the duration is longer when load is lower but is shorter when load is higher.
Under same condition (constant ignition timing of 34° BTDC), ethanol shows no sign of knock (can be extended to 8.5 bar IMEP safely), while DMF cause a knock at 7.1 bar IMEP and gasoline at 6.5 bar IMEP. The result of DMF was unexpected and need further research.
When using DMF as fuel, the emissions of CO, HC, and NOx are similar to those of gasoline, while ethanol produces the lowest emissions of the three.
The PM emission of DMF is also very similar to that of gasoline.
In conclusion, it's clear to see that the combustion and emissions characteristics of DMF are similar to those of gasoline.
2.1.2 Laminar Burning Velocities of DMF Compared with Ethanol and Gasoline [5]
The laminar burning velocity of a fuel can affect the burning rate and thus the performance of the engine. The velocity is highly affected by temperature but less by pressure. And the compositions of the residual unburned gases and the equivalence ratio also have effect on the velocity. In this experiment, the laminar burning velocities of the three fuels are measured at three different initial temperatures (50, 75 and 100℃) under 0.1MPa initial pressure. The system for testing is shown below.
Figure 4 experimental system
Experimental setup
TTL control signal width is fixed to 10ms for all experiments to adjust the discharge energy.
The safety valve is set to 0.7MPa.
Light source: a 500W xenon lamp.
Camera for capturing the schlieren images: a Phantom V7.1 high-speed camera, 800 x 600pixels, 6600 picture per minute. Sample rate: 3 kHz.
Pressure in vessel for each test: 10 kPa absolute pressure. Fuel injection pressure (direct injection): 10 MPa.
The vapor pressures (kPa) of the three fuels under different initial temperatures are listed in the table below.
Fuel/ temperature(℃)
50
75
100
DMF
20.3
54.7
126.3
ethanol
29.5
88.8
225.7
gasoline
147.5
272.6
464.0
Table 3
The figure below shows the partial pressures respect to increasing equivalence ratio (mixture pressure = 0.1MPa).
Figure 5
The mixture of fuel and air is ignited after 5 minutes' quiescence for the complete mixing.
An ETAS LA4 lambda meter is used to measure the equivalence ratio.
An in-house MATLAB code is used for data processing.
The measurement of spherical flame front is to detect the change of the gas density between the burned and unburned gases in the vertical direction (see figure A11 in the appendix).
Results and discussion
All testing results graphs are in appendix (figure A11-A16).
Definitions
Sn = the stretched laminar flame speed (the rate of the change of the schlieren flame radius)
Ss = the unstretched flame speed
u1 = the unstretched laminar burning velocity
Lb = the Markstein length (used for measuring the effect of curvature on a flame)
Ф = equivalence ratio α = stretched ratio
In figure A12, it's clear to see that the flame speed of ethanol is the highest among the three fuels. And there is hardly a difference between the case of DMF and gasoline.
The unstretched flame speeds of the three fuels can be determined by analyzing the stretched flame speed (see figure A13). In figure A14, it's clear to see that the fastest flame propagation happens when the equivalence ratio is between 1.0 and 1.2. When the initial temperature is 50℃, the unstretched flame speeds of the three fuels follow this order: Ethanol > DMF > gasoline. In fact, the results between DMF and gasoline are quite close to each other. And this situation remains the same when the initial temperature is 75℃ or 100℃.
Figure A15 shows the Markstein length of the three fuels at different initial temperature. Lb > 0 means that the flame speed decreases when the stretched rate increases while Lb < 0 means that the flame speed increases when the stretched rate increases. And Lb decreases as the equivalence ratio increases according to the graph. The Markstein length has a higher value when the equivalence ratio is low (0.8-0.9) as temperature increases. This means that when temperature is higher, the lean flame is more stable. From the graph, it is clear to see that the curve of DMF is always under the curve of ethanol, and it is very close to gasoline. This means that compared to ethanol, the flame of DMF and gasoline is more unstable.
The laminar burning velocity is one of the key indicators of flame behavior. The test results are shown in figure A16 in appendix. it's clear to see that ethanol has the highest laminar burning velocity among the three fuels for all temperature conditions, especially when the initial temperature is 50℃. However, as the initial temperature increases, the superiority of ethanol becomes smaller. As to DMF, the graph indicates that it has the lowest laminar burning velocity among the three fuels, though the cure of gasoline is very close to the curve of DMF. And as the equivalence increases, the distance between the curves of DMF and gasoline becomes bigger except when the initial temperature is 50℃.
Conclusions
The highest stretched flame speeds of the three fuels occur when equivalence ratio is between 1.1 and 1.3.
The differences in flame stability of the three fuels are marginal.
Ethanol has the highest laminar burning velocity among the three fuels. The cases of DMF and gasoline are close to each other, but the velocity of DMF is slightly lower than that of gasoline.
Experiment section
2.2.1 Testing the performances of DMF, ethanol and gasoline in a boosted engine
Introduction
Boosting, in simple words, is using some methods or certain equipment, such as turbocharger, to increase the air flow rate entering the engine. The pressure of air entering the engine is forced to increase so that the engine create more power and have a higher efficiency compared to normal engines.[6][7]
The purpose of the test is to analysis the performance of DMF in a boosted engine and to make a comparison with ethanol and gasoline. The engine for testing is the Medusa engine.
The designs of the system
The pressure of air flow in the engine room is too low to boost the engine and equipment for measuring and controlling the air flow or pressure is needed in order to get more accurate results.
Design A
The method of increasing the air pressure is connecting the outlet pipe to a big tank and then connects the tank to the engine. Then the air flow meter is installed on the pipe between the tank and the engine and the controller is between the high pressure air pipe and the tank. The design is shown in the figure below.
Figure 7
The tank, pipes and cooling system can be manufactured in the lab.
The air flow meter/controller was planned to be purchased from Flowtech as shown below.
Figure 8 The flow meter
Figure 9 The flow meter
MODEL NUMBER
8716MPNH
SPECIFICATIONS
MNPT 2"
Length 14"
Linear signal output 0-5 VDC & 4-20 mA
Signal Interface RS232 & RS485
Accuracy, including linearity (Ref.: 21°C) ±[1% of Reading + (.5% + .02%/°C of Full Scale)]
Repeatability ±0.2% of Full Scale
Sensor response time 1 second
Turn down ratio 100:1 minimum
Electronics temperature range -40°-85°C (-40°-185°F)
Gas temperature range -40°-200°C (-40°-392°F), extended range available
Gas pressure effect Negligible over ± 20% of absolute calibration pressure
Pressure rating maximum 500 PSI Std., > 500 PSI special
Input power requirement 24VDC @ 250mA
115 VAC 50/60 Hz optional
230 VAC 50/60 Hz optional
Flow Transmitter power requirements 5 watts maximum
Flow Transmitter enclosure NEMA 4X, ABS plastic with clear polycarbonate cover, 5" x 5" x 4"
Wetted materials 316 Stainless Steel (Hastelloy optional)
Standard temperature & pressure (STP) 70°F & 29.92" Hg (Air .075 lb./cubic foot)
NIST traceable calibration Standard
The flow meter is used to measure the air flow rate into the engine and report the result to computer through its own software. And with the respond of the flow meter we can use the valve to control the air flow rate
Although the engine can be boosted by using this design, there are some problems.
The cost of the flow meter is high (£2808.4) and it would take about one month to deliver. This is the biggest problem of the design.
The pressure may not be enough even with the tank installed. If we increase the volume of the tank, the engine room may not have enough space.
The installation of the flow meter requires critical environment as the figure shown below.
Figure 9
In order to have testing results with high accuracy, the minimum length requirement of the pipe for installing the flow meter is 15x14"=210"=5.334m. This is also difficult to be achieved in the engine room.
As a result of fact, design A cannot be realized because of the problems listed above.
Design B
Because the biggest problem of design A is the cost, design B uses a cheaper way.
The design is shown in Figure 10.
Figure 10
The flow meter is from the lab (cheaper but lower accuracy compared to the one in design A). To prevent the insufficient pressure condition, an air compressor is added between the air inlet and the tank (or using the compressor only). The tank is smaller than the one in design A. The valve is used to release the air inside the tank in case the pressure is too high and also used to control the air flow rate (inaccurate). The air compressor can be purchased. The water pipe is used to cool down the air temperature from the air compressor.
Compared to design A, the accuracy of design B is much lower (both the measuring and controlling) so that the testing results might not be reliable. And also because the manufacture of the components would take a long time, design B was not realized either.
According to the situation above, performance tests by boosting the engine cannot be made. So another method, using Miller Cycle to test the performance of DMF, is used.
2.2.2 Using Miller Cycle engine to test the performance of DMF, ethanol and gasoline
Introduction
Miller Cycle is an upgrade of Otto Cycle. As we all know that the power of an engine is from the explosion procedure. If the procedure can be remained and the time for the piston through the bottom dead center is long enough, then we will be able to make full use of the pressure of combustion gas expansion to produce work. A value called the expansion ratio is used to define the volume ratio of the combustion gases and the combustion chamber. Miller Cycle engine is one kind of engine with a large expansion ratio, which means combustion gases have more expansion and more work is down. However, one problem is that the bigger the combustion ratio, the bigger chance that a knock will occur. In a Miller Cycle engine, this problem is solved by releasing some of the gases, which means in the combustion stroke the gases in the combustion chamber is less so that knocking is avoided. The reason for this is that in Miller Cycle engine the intake valve closes very slowly, which is at about 70 degrees after the piston passes the bottom dead center. During this procedure, the piston will push some of the gases back into the inlet pipe, where the gases will be stored and then be absorbed again into the chamber during the next cycle.[8][9]
Experiment
The engine for testing is the also the Medusa Engine. The whole system using for testing is the same as the 2.1.1 experiment except that the intake valve is controlled and measured by computer. The intake valve closing time was delayed by 30 degrees. The test for ethanol has failed for unknown reason, so this experiment is for DMF and gasoline only. The data of experiment setup and testing results please see to the tables in appendix (Table A1 & Table A2).
Results and discussions
Graphs in the appendix show the comparisons between DMF and gasoline in a Miller Cycle engine (Figure A17-A25).
Figure A17 shows the covariance (COV) of IMEP for DMF and gasoline, a bigger value of COV means that the combustion is more unstable, before IMEP = 6bar, the cure of DMF is slightly higher than that of gasoline, which means before this point, the combustion of gasoline is more stable than DMF. However, after this point, the instability of the combustion of gasoline increases rapidly and the curve of gasoline becomes higher than that of DMF.
Figure A18 shows the maximum pressure inside the chamber of IMEP for DMF and gasoline. Normally, a higher pressure means a higher work output. According to the graph, before the point IMEP = 6bar, the difference of pressure between DMF and gasoline is quite small, but as the load increases, Pmax of DMF increases while gasoline remains almost the same.
Figure A19 shows the crank angles of a piston moved between 10% and 90% Mass Fraction of Burned Fuel of IMEP for DMF and gasoline. This value is used to compare the work output of the combustion. If a piston has rotated a smaller angle, then this means that the pressure on the piston is higher thus there would be more combustion work output. According to the graph, the curve of DMF is always below that of gasoline, which indicates that the combustion work output of DMF is more than that of gasoline.
Figure A20 shows the temperature of tail gas of IMEP for DMF and gasoline. The fuel in the combustion chamber could never be completely burned. Some of the fuel will get out of the chamber with other gases. If the temperature of tail gas is high enough, then the "escaping" fuel could burn and this will increase the efficiency of the engine. According to the graph, the temperature difference between DMF and gasoline is very small, and the comparison results vary with load.
Figure A21-A24 shows the indicated specific emissions (CO, CO2, HC and NOx) of IMEP for DMF and gasoline. Firstly, the results of CO vary with the increasing of load. Before the IMEP is about 5.5bar, the emission of CO for gasoline is higher than that of DMF. Then between 5.5bar and 7bar, the emissions of CO for DMF and gasoline are similar. After 7bar, the emission of CO for DMF becomes higher than that of gasoline. Secondly, it is clear to see that the emission of CO2 for DMF is always higher than that of gasoline. Thirdly, the result of HC emission for DMF is higher than that of gasoline before IMEP = 4.2bar and then lower after the point. However, after 8bar, the emission for DMF suddenly rises. This may be caused by experiment error. And the last one is the emission of NOx. It is clear to see that the emission of NOx for gasoline is always higher than that of DMF.
Figure A25 shows the indicated efficiency of IMEP for DMF and gasoline. Ordinarily, the higher the efficiency, the better because lower efficiency means more thermal energy lost. According to the graph, before 5bar, the efficiencies of DMF and gasoline are similar to each other. After that the efficiency of DMF becomes lower than that of gasoline. However, as is shown in the graph, the curve of DMF suddenly rises after IMEP = 6.8bar. According to scientific theory, this is not supposed to happen. This may be caused by experimental error so that the results between 6.8bar and 8.8bar are probably unreliable.
Conclusion
From the results of this experiment, it can be inferred that the combustion properties of DMF in a Miller Cycle engine are quite similar to that of gasoline. Those two fuels both have advantages and disadvantages compared to each other. And possible errors could have some negative effects on reliability of the results. In conclusion, according to the similarity of the results for DMF and gasoline, it can be confirmed that DMF is a good choice as an alternative fuel to gasoline.
3 Conclusions
From all the testing results listed above, it is clear that the thermal properties of DMF are quite similar to those of gasoline. It can be inferred that DMF could be an alternative fuel to gasoline. Compared to ethanol, DMF's high energy density and low water insolubility are its biggest advantages. And compared to gasoline, the main advantage of DMF is its high initial boiling point. However, most thermal properties of DMF, such as laminar burning velocity, are no better than those of ethanol. And the biggest problem for DMF is that its price is rather high compare to the other two. This is because there are few industries on earth produce DMF and the producing procedure is more complicated than that of ethanol and gasoline. Because of all those disadvantages, DMF, at least for now, have to stay in the lab for research only rather than being used as an alternative fuel to gasoline.
