- Email: [email protected]
Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation
JID: PROCI
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
Available online at www.sciencedirect.com
Proceedings of the Combustion Institute 000 (2016) 1–7 www.elsevier.com/locate/proci
Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation Dan Janecek∗, David Rothamer, Jaal Ghandhi University of Wisconsin-Madison, Engine Research Center, 1500 Engineering Dr., Madison, WI 53706, United States Received 2 December 2015; accepted 3 August 2016 Available online xxx
Abstract A novel fuel-substitution strategy was used to investigate the relationship between octane number and cetane number under representative low temperature combustion (LTC) thermodynamic conditions. The cetane number of the test fuels was known by using mixtures of the cetane number secondary reference fuels (SRFs) T-26 and U-19, which are full-blend certification fuels used in the place of 2,2,4,4,6,8,8heptamethylnonane and n-hexadecane for engine cetane number testing. The octane number was derived from the fuel-substitution methodology in terms of the octane primary reference fuel (PRF, consisting of n-heptane and isooctane) mixture needed to maintain combustion phasing during homogeneous-charge compression-ignition (HCCI) engine operation as increasing amounts of the SRF mixture was substituted for the PRF mixture. A linear blending relationship was found to exist between the cetane and octane numbers. Results agree with other similar relationships from the literature acquired under substantially different operating conditions. The current study significantly expands the range of the cetane–octane correlation to a range of cetane numbers from 20 to 75 and validates the correlation at low equivalence ratios of interest to LTC operation. © 2016 by The Combustion Institute. Published by Elsevier Inc. Keywords: Cetane; Octane; SRF; LTC
1. Introduction Octane and cetane ratings are two important parameters that describe a fuel’s ignition qualities. The octane number is related to the fuel’s resistance to autoignition and is an important indicator of engine knock, which ultimately limits a spark-ignition ∗
Corresponding author. E-mail address: [email protected] (D. Janecek).
engine’s compression ratio and, thus, efficiency. The cetane number describes a fuel’s propensity to ignite, and plays a vital role for compression ignition (CI) engines. The octane number (ON) of a fuel is experimentally found using both the research octane number (RON) and motored octane number (MON) methods as specified by ASTM D2699 and D2700, respectively [1,2]. The octane number determined by either test refers to the mixture of octane number primary reference fuels (PRFs)
http://dx.doi.org/10.1016/j.proci.2016.08.015 1540-7489 © 2016 by The Combustion Institute. Published by Elsevier Inc.
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
JID: PROCI
2
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
isooctane (2,2,4-trimethylpentane) and n-heptane that exhibits a standardized knocking intensity at the same compression ratio (CR) as the test fuel for prescribed operating conditions. The resulting octane number corresponds to the volumetric percentage of isooctane in the PRF mixture. The RON test conditions are: intake air temperature of 52 °C and ambient pressure, spark ignition timing 13° before top dead center (bTDC), and an engine speed of 600 rpm. The MON test conditions are: intake air temperature of 149 °C and ambient pressure, spark ignition timing of 26° to 14° bTDC depending on engine CR, and an engine speed of 900 rpm. Both tests use a variable CR engine (ranging from 4 to 18), and fuels typically tested using these methods have octane ratings between 40 and 100. Due to the low speeds and high intake temperatures used in these tests, numerous studies have found that fuel reactivity cannot be adequately described by the RON and MON tests; many fuels exhibit a strong sensitivity to engine operating conditions [9,10]. Therefore, it is useful to have the ability to test fuels of interest at a variety of different engine-relevant operating conditions; the current testing method will be described in more detail later in the paper. Cetane number (CN) is found using a variable CR pre-chamber diesel engine according to ASTM D613 [3]. The cetane number of a fuel, refers to the mixture of cetane number primary reference fuels isocetane (2,2,4,4,6,8,8-heptamethylnonane) and n-hexadecane that exhibits the same ignition delay of 13 crank angle degrees (CAD) at the same CR for a set engine operating condition. The cetane test conditions specify an intake air temperature of 66 °C and ambient pressure with a start of injection (SOI) occurring in the pre-chamber at 13° before top dead center. Compression ratio, which can vary between 8 and 36, is adjusted until ignition occurs at TDC. Typical diesel fuels tested have cetane numbers between 30 and 65. In practice, due to the high cost of the cetane primary reference fuels, secondary reference fuels (SRFs) T and U, which are tightly controlled full-blend fuels made by Chevron Phillips, are used for testing. The specific SRFs used in this study are T-26 and U-19, having tested cetane numbers of 75.2 and 19.4, respectively. Formerly, a direct comparison between gasoline-like fuels described by octane number and diesel-like fuels specified described by cetane number was not necessary since they had separate applications; high-octane (low reactivity) fuels were predominantly used in SI engines while high-cetane (high reactivity) fuels were used in CI engines. However, due to the increased interest in LTC, combustion strategies are being investigated involving the compression-ignition of low-reactivity fuels [4], and the use of mixtures of low- and high-reactivity fuels in dual-fuel combustion strategies [5]. Many of these strategies utilize fully or partially premixed quantities of fuel at
Fig. 1. Cetane number as a function of RON. Data compiled from [6] and [7].
much lower than stoichiometric equivalence ratios. An advantage of using a HCCI testing method is that it allows for the ignition quality of the fuel to be characterized fully by its kinetic behavior while avoiding fuel spray injection complications. Computational fluid dynamics (CFD) is an important tool for the investigation and development of low temperature combustion strategies. However, the implementation of detailed chemistry of large molecules, like those found in diesel fuel, in CFD is still beyond the capability of current computing power for engineering applications. Thus, simplified chemistry needs to be used. The most common surrogate for diesel fuel in CFD is nheptane because skeletal chemical kinetics models are available. Skeletal models also exist for the PRF blends. A well-established link between cetane number and octane number would potentially allow more accurate CFD modeling of fuels, i.e., it would provide an easy-to-implement means for varying the cetane number. Direct comparisons of the cetane and octane number have been undertaken using the results of RON/MON and cetane testing. Kalghatgi [6] measured the cetane number of PRF blends and mixtures of n-heptane and toluene, and compared them to their measured (or known) research octane numbers (RON). The results showed a linear correlation between the two metrics. The National Renewable Energy Laboratory (NREL) [7] compiled results for an array of single-component hydrocarbons, including oxygenates, and also recovered a linear relation between RON and cetane number. These results are reproduced in Fig. 1, including confidence intervals that arise from the linear fitting. The results in tabular form can be found in
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
JID: PROCI
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7 Table 1 Engine geometry. Compression Ratio Displacement (L) Stroke (mm) Bore (mm) Con Rod Length (mm) Intake Valve Closing Exhaust Valve Opening Swirl Ratio Piston Bowl Type
9.5 0.477 90.4 82 145.54 −132 °aTDC 112 °aTDC 1.5 Stock (Re-entrant)
Appendix A. Both data sets show a strong correlation between CN and RON over the limited range of CNs tested. The NREL correlation covers a larger range of CNs but has a larger scatter in the data resulting in a weaker correlation between CN and RON (as seen by the larger confidence interval bands and an R2 value of 0.87 compared to an R2 value of 0.98 for Kalghatgi’s data). In the current study, mixtures of the cetane secondary reference fuels T-26 and U-19 were tested using a novel fuel-substitution methodology that provides an effective PRF (octane) number for each tested fuel mixture. The main features of the fuel substitution method [8] are: engine tests are performed using homogenous-charge compressionignition (HCCI) combustion to avoid fuel spray and flame propagation complications while allowing engine-relevant conditions to be investigated; combustion phasing is controlled independently of intake and boundary conditions allowing the fuel chemistry to be isolated; multiple engine operating conditions are tested to cover a range of temperature and pressure regimes; and a wide CN range of fuels can be tested due to an absence of the requirement of the effective octane number being between 0 and 120. The octane number test limits of 0–120 correspond to the octane number limits of the reference fuels n-heptane (0) and isooctane mixed with 6 mL/gallon of tetraethyllead (120.3) [1]. The main objective of this study is to investigate the relationship between the CN of the SRF fuel blends and the ON of the PRF mixtures that exhibit matched ignition characteristics (reactivity) independent of fuel spray complications at representative LTC operating conditions.
3
and the intake air temperature was controlled using heaters both upstream of and surrounding the outer walls of the intake surge tank. Fuel pressure and injection timing were controlled using the National Instruments (NI)Drivven hardware/software suite. NI hardware was also used for all high- and low-speed controls such as intake air pressures and temperatures, coolant temperatures, and the exhaust gas recirculation (EGR) valve. In-cylinder pressure was measured using a Kistler 6125A pressure transducer in series with a Kistler 5010 charge amplifier. A 0.25° resolution encoder fixed to the crankshaft was used as the timing clock for the high-speed data acquisition, giving a resolution of about 22 μs, at the tested speed. Pressure data were acquired for 300 cycles for each test condition. Horiba gas analyzers were used to determine relevant emissions measurements. The engine was operated in an HCCI combustion mode. Three fuel streams were delivered to the engine; n-heptane and isooctane were independently port injected, and the SRF fuel mixtures were delivered from an upstream prevaporizing system. To ensure the complete vaporization of the low-volatility fuels, an air-assisted fuel injector was used to inject a highly atomized fuel spray into the main intake air stream directly downstream of an in-line heater. Tape heaters were utilized to maintain the downstream wall temperatures at the gas temperature up to the engine intake port. At the engine operating speed of 1900 rpm, which was used for all tests, the fuel has a residence time of approximately 3.5 s in the intake system. The engine speed was chosen in order to achieve reasonable combustion phasing for the intake temperature and pressure conditions. Due to the relatively high intake temperatures, tests were performed to verify that no fuel decomposition reactions occurred in the intake system. Heptane, which has similar ignition characteristics to the diesel fuels tested, was injected both through the upstream pre-vaporizer system and at the intake port. Pressure and heat release data were analyzed, and no measureable effect was found, verifying that intake temperatures up to 200 ºC likely had no effect on the combustion characteristics for the residence times used in the experiments. Details about the prevaporizing system are described in more detail in [8].
2. Experimental setup 3. HCCI constant phasing fuel blend studies Engine experiments were carried out in a singlecylinder version of the General Motors/Fiat JTD 1.9 L four-cylinder diesel engine. Engine specifications are given in Table 1. In order to maintain reasonable fuel ignition timing, the compression ratio was reduced from a stock value of 16.7–9.5.The stock four-cylinder engine head was mounted to a Ricardo Hydra single-cylinder block. The air flow rate was controlled using choked flow orifices,
In this study, the fuel substitution strategy described in [8] was utilized. It was shown in [8] that, at a minimum, combustion phasing, load and intake temperature must be kept constant in order to keep the boundary conditions, such as the wall and piston temperatures, constant. In previous work it was shown that there is a strong connection between piston temperature and combustion phasing,
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
JID: PROCI
4
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
Fig. 2. Pressure and AHRR for SRF20 Substitution Sweep.
and piston temperature also has a strong influence on ignition. The reader is referred to [8] for details. The fuel substitution strategy works as follows. An operating point is first run using only the portinjected PRFs. The test fuel is then introduced through the upstream pre-vaporizer system at the desired mass fraction; the port fuel mass is reduced to maintain the engine load, and the isooctane-ton-heptane ratio is adjusted to maintain the same combustion phasing. Figure 2 shows the pressure and apparent heat release rate (AHRR) for SRF20, i.e., a mixture of SRFs that gives a CN of 20, substitution. For all of the sweeps run, the typical variation of key combustion parameters were as follows: IMEP varied by less than 1%, CA50 by less than 0.5° from the mean, and the IVC temperature determined using the trapped mass and ideal gas law varied by less than 3 °C about the mean. It is seen that nearly identical in-cylinder thermal conditions and combustion performance were achieved for the entirety of the sweep, where yhf is the mass fraction of the heavy fuel, i.e., the SRF blend. Figure 3 shows the required PRF number of the port-injected PRF fuel mixture (PRFmeas ) required to keep phasing and load constant as a function of heavy-fuel mass fraction (yHF ) of SRF20. Since the IVC temperature, load, and combustion phasing are kept constant, the total fuel mixture reactivity, which is expressed as the overall PRF number, should be constant throughout the sweep and be approximately equal to the baseline PRF mixture when no heavy fuel is injected upstream, i.e., 50 for the case shown in Fig 3. It is evident that the SRF20 (CN=20) mixture has a lower reactivity than a PRF50 mixture. Therefore, the port-injected PRF mass ratio had to be adjusted such that more n-heptane was injected, i.e., the PRFmeas was decreased, as the mass fraction of SRF20 increased.
Fig. 3. Measured PRF as a function of the mass fraction of SRF20.
If the resulting mixture reactivity obeys a linear mass-based blending rule, the required PRF vs. heavy fuel fraction data can be used to calculate an effective PRF number for the heavy fuel, PRFHF . The effective PRF of the three-component fuel (nheptane / isooctane / heavy fuel) mixture, PRFmix , which is constant as discussed above, can be written as PRFmix = (1 − yHF )PRFmeas + yHF PRFHF
(1)
where PRFmeas is the port-injected mixture PRF number. This equation has two unknowns, PRFmix and PRFHF , that can be solved for using a leastsquares approach. This method has been used in previous studies to calculate an effective PRF number for diesel-like fuels, and the linear blending rule was seen to accurately describe the overall mixture reactivity [8]. As seen in fig. 3, an effective PRFHF of 86.5 is calculated for the SRF20 mixture, and the predicted PRFmeas as a function of SRF20 mass fraction matches very well with the experimental data points, indicating that the linear blending rule accurately describes the overall mixture reactivity for this case. Figure 4 shows the raw data and the curve fits using Eq. (1) for all of the SRF mixtures tested at this operating condition. It can be seen that the fits to the data are excellent and correctly capture the curvature of the data points suggesting that, despite its simplicity, the linear blending rule applied in Eq. (1) adequately represents the results. In order to investigate whether the effective PRF of the heavy fuel is dependent on the engine operating conditions, four different operating conditions were tested. An overall mixture PRF number of 20 and 50, which gave rise to
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
JID: PROCI
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
5
Table 2 Engine operating conditions for constant phasing sweeps. Operating condition
Intake P [bar]
Intake T [°C]
Gross IMEP [bar]
1 2 3 4
1.2 1.2 2 2
165 165 165 165
4.2 4.6 5.4 5.6
CA50 [°aTDC] −8 3 −15 −11
[-]
T (CA10) [K]
0.36 0.36 0.28 0.28
850 925 775 800
ρ(CA10) [kg/m3] 7.8 10.2 9.9 10.5
Baseline PRF [-] 20 50 20 50
Fig. 5. Calculated SRF PRF for all sweep conditions. Fig. 4. Measured PRF and the regression using the linear blending model for a range of SRF mixtures.
substantially different combustion phasing, were each run at intake pressures of 1.2 bar and 2.0 bar. A summary of operating conditions can be seen in Table 2. Temperature and density are estimated at the crank angle that 10% of the cumulative heat release is reached (CA10). Density at CA10 is modest due to the high intake temperatures required to ensure full fuel vaporization and the low compression ratio. The density for operating condition 2 compared to condition 1 is substantially higher due to the later combustion phasing, causing CA10 to occur closer to TDC. Figures 5 and 6 summarize the results from all of the data collected. Figure 5 shows the effective PRF values obtained for each SRF mixture tested, with confidence intervals, calculated at each of the four operating conditions given in Table 2. The horizontal lines represent the average for each fuel mixture. Varying thermodynamic operating conditions such as intake temperature and pressure can have a large effect on fuel chemistry and ignition behavior; however it can be seen that over the range of conditions tested the operating condition did not have a significant influence on the effective PRF number when the uncertainties are accounted for. Figure 6 shows the direct correlation between the effective PRF number and the cetane number
Fig. 6. Summary of experimental data with linear fit and CI.
along with a best fit line. The linear fit was calculated using a weighting function based on the individual condition confidence intervals. The linear fit is shown with 95% confidence intervals. The best-fit line relating PRF and cetane number (CN) is given
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
JID: PROCI
6
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
Fig. 7. Comparison of experimental correlation of current study with the results from with previous studies [6,7]. Table 3 Summary of cetane vs. RON/PRF curve fits [6,7].
Current testing Kalghatgi NREL
Slope
Y-intercept
−1.96+/−0.07 −2.34+/−0.28 −2.18+/−0.31
125+/−4 128+/−10 132+/−10
as PRF = 125 − 1.96 CN 2
(2)
with an R value of 0.99. In general, there is an excellent correlation between the effective PRF number measured using the fuel-substitution method and the cetane number. Figure 7 shows the linear correlation from Eq. (2) along with the linear fits to the data of Kalghatgi [6] and NREL [7], previously shown in Fig. 1. The 95% confidence intervals are included for all three linear fits. There appears to be a noticeable difference in the slope of the fit to the current data relative to the other two correlations from the literature. However, within the limits of uncertainty, there is no significant difference between all three sets of data in terms of slope or intercept of the curve fits. A summary of the calculated slopes and y-intercepts along with the corresponding uncertainties can be seen in Table 3. The Kalghatgi data were obtained using the ASTM D613 Cetane test on fuels of known RON, whereas the NREL data were obtained using a variety of Cetane tests (ASTM D613, ASTM D6890, and ASTM D7170) and the ASTM D2699 RON test. The conditions for the current measurements differ substantially from those for the NREL and Kalghatgi correlations. For the present data, igni-
Fig. 8. Comparison between fuel substitution CA10 conditions and estimated RON, MON, and cetane test conditions [11,12].
tion occurs at conditions that are quite lean and with a homogeneous air–fuel mixture, whereas for the RON, MON, and cetane number tests ignition takes place at richer conditions, and the cetane number tests occur with significant inhomogeneity due to the indirect fuel injection. Further, HCCI combustion is purely based on chemical kinetics, while RON and MON tests are also influenced by physical properties that affect flame speed, and the cetane number test is affected by physical properties involved with the fuel injection, vaporization, and mixing processes. Another issue that could influence the comparison between the different measurement procedures is the temperature-density region that the different experiments encompass. Figure 8 shows the estimated thermodynamic conditions for the four operating conditions (as seen in Table 2) at CA10, which is used as a proxy for the ignition timing. The RON, MON, and CN curves in Fig. 8 were taken from the simulations of Yates et al. [11,12]. This simulation work involved modeling the RON, MON, and CN engine experiments in order to estimate in-cylinder thermodynamic conditions. It can be seen that a much larger range of temperatures and densities encompasses the RON or MON tests than is seen for the cetane test. This is partly due to a smaller range of compression ratios required for the typical range of fuels tested with the cetane test. Another cause for the RON and MON tests reaching a higher temperature-density operating zone than the cetane test is that as spark ignition occurs, the end-gas is further compressed by increasing cylinder pressures caused by combustion occurring behind the flame front. Overall, the temperaturedensity data for the current testing can be seen to
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
JID: PROCI
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
agree quite well with the operating conditions seen in the other tests. The range is smaller than the spread seen in the RON and MON tests due to operating constraints involved with HCCI combustion. The current results allow a significantly wider range of fuel ignition quality to be tested, although it is acknowledged that negative PRF numbers may have limited direct applicability at present.
7
CN and RON data for a number of single component hydrocarbons, including oxygenates reproduced from [7] CN RON CN RON
8 104 28 73
12 109 29 83
12 118 30 62
14 98 30 74
15 106 31 59
17 83 32 55
17 98 34 73
19 78 36 46
21 91 41 28
22 71 41 29
22 93 43 56
23 84 44 43
24 75 47 25
24 92 54 0
27 77 57 22
References 4. Conclusions A strong linear correlation was found between cetane number, which was defined by the mixture ratio of secondary reference fuels tested, and octane number, which was defined as the effective PRF number obtained using a novel fuel- substitution method, was found over cetane numbers from 20 to 75. This relationship was found to be similar to trends seen under substantially different operating conditions previously reported in the literature; it deviated most significantly at the upper limit of cetane numbers previously tested. The current results cover a wider range of cetane number than previous literature measurements and span a different range of conditions. The advantage of the method used in the current study relative to enginebased MON, RON, and cetane number testing is that it can be used to test high-reactivity fuels and fuels with low volatility that would not be practical to test in a spark-ignition engine. A unique aspect of the fuel-substitution method is that it can provide negative values of octane number, i.e., fuels with higher reactivity to n-heptane can be tested. While this testing was done at low equivalence ratios and representative LTC conditions, the method could easily be expanded to different operating conditions such as varying engine speed, higher intake pressures and equivalence ratios, exhaust gas residuals (EGR) dilution, etc.
Acknowledgments Support for this work was provided by the Office of Naval Research (ONR) Contract N00014-12-10650, Sharon Beermann-Curtin, grant monitor.
Appendix A CN and RON data for a number of PRF blends and mixtures of n-heptane and toluene reproduced from [6]
[1] American Society for Testing Materials: Designation: D2699, “Standard Test Method for Research Octane Number of Spark-Ignition Engine Fuel”, ASTM Inernational, West Conshohocken, PA [2] American Society for Testing Materials: Designation: D2700, “Standard Test Method for Motor Octane Number of Spark-Ignition Engine Fuel”, ASTM International, West Conshohocken, PA. [3] American Society for Testing Materials: Designation: D613, “Test Method for Cetane Number of Diesel Fuel Oil”, ASTM International, West Conshohocken, PA. [4] Y. Ra, P. Loeper, R. Reitz, et al., Study of High Speed Gasoline Direct Injection Compression Ignition (GDICI) Engine Operation in the LTC Regime, 2011 SAE Technical Paper 2011-01-1182, doi:10. 4271/2011- 01- 1182. [5] S.L. Kokjohn, R.M. Hanson, D.A. Splitter, R.D. Reitz, Int. J. Engine Res. 12 (2011) 209–226, doi:10. 1177/1468087411401548. [6] G.T. Kalghatgi, Auto-Ignition Quality of Practical Fuels and Implications for Fuel Requirements of Future SI and HCCI Engines, 2005 SAE Technical Paper 2005-01-0239, doi:10.4271/2005- 01- 0239. [7] J. Yanowitz, M.A. Ratcliff, R.L. McCormick, J.D. Taylor, M.J. Murphy, Compendium of Experimental Cetane Numbers, National Renewable Energy Laboratory, 2014 Technical Report 5400-61693. [8] Janecek, D., Rothamer, D., and Ghandhi, J., “Fuelsubstitution method for investigating kinetics of low volatility fuels under engine-like operating conditions”, 9th U.S. National Combustion Meeting 114IC–0377. [9] G.T. Kalghatgi, Fuel Anti-Knock Quality – Part I. Engine Studies, 2001 SAE Technical Paper 2001-013584, doi:10.4271/2001- 01- 3584. [10] G.T. Kalghatgi, Fuel Anti-Knock Quality – Part II. Vehicle Studies – How Relevant is Motor Octane Number (MON) in Modern Engines?, 2001 SAE Technical Paper 2001-01-3585, doi:10.4271/ 2001- 01- 3585. [11] D.B. Yates, A. Swarts, C.L. Viljoen, Correlating Auto-Ignition Delays and Knock-Limited SparkAdvance Data for Different Types of Fuel, 2005 SAE Technical Paper 2005-01-2083, doi:10.4271/ 2005- 01- 2083. [12] D.B. Yates, A. Swarts, C.L. Viljoen, Understanding the Relation Between Cetane Number and Combustion Bomb Ignition Delay Measurements, 2004 SAE Technical Paper 2004-01-2017, doi:10.4271/ 2004- 01- 2017.
CN 26 26 26 32 32 33 36 36 39 42 47 RON 70 69 69 50 50 50 42 41 40 30 20
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
Available online at www.sciencedirect.com
Proceedings of the Combustion Institute 000 (2016) 1–7 www.elsevier.com/locate/proci
Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation Dan Janecek∗, David Rothamer, Jaal Ghandhi University of Wisconsin-Madison, Engine Research Center, 1500 Engineering Dr., Madison, WI 53706, United States Received 2 December 2015; accepted 3 August 2016 Available online xxx
Abstract A novel fuel-substitution strategy was used to investigate the relationship between octane number and cetane number under representative low temperature combustion (LTC) thermodynamic conditions. The cetane number of the test fuels was known by using mixtures of the cetane number secondary reference fuels (SRFs) T-26 and U-19, which are full-blend certification fuels used in the place of 2,2,4,4,6,8,8heptamethylnonane and n-hexadecane for engine cetane number testing. The octane number was derived from the fuel-substitution methodology in terms of the octane primary reference fuel (PRF, consisting of n-heptane and isooctane) mixture needed to maintain combustion phasing during homogeneous-charge compression-ignition (HCCI) engine operation as increasing amounts of the SRF mixture was substituted for the PRF mixture. A linear blending relationship was found to exist between the cetane and octane numbers. Results agree with other similar relationships from the literature acquired under substantially different operating conditions. The current study significantly expands the range of the cetane–octane correlation to a range of cetane numbers from 20 to 75 and validates the correlation at low equivalence ratios of interest to LTC operation. © 2016 by The Combustion Institute. Published by Elsevier Inc. Keywords: Cetane; Octane; SRF; LTC
1. Introduction Octane and cetane ratings are two important parameters that describe a fuel’s ignition qualities. The octane number is related to the fuel’s resistance to autoignition and is an important indicator of engine knock, which ultimately limits a spark-ignition ∗
Corresponding author. E-mail address: [email protected] (D. Janecek).
engine’s compression ratio and, thus, efficiency. The cetane number describes a fuel’s propensity to ignite, and plays a vital role for compression ignition (CI) engines. The octane number (ON) of a fuel is experimentally found using both the research octane number (RON) and motored octane number (MON) methods as specified by ASTM D2699 and D2700, respectively [1,2]. The octane number determined by either test refers to the mixture of octane number primary reference fuels (PRFs)
http://dx.doi.org/10.1016/j.proci.2016.08.015 1540-7489 © 2016 by The Combustion Institute. Published by Elsevier Inc.
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
JID: PROCI
2
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
isooctane (2,2,4-trimethylpentane) and n-heptane that exhibits a standardized knocking intensity at the same compression ratio (CR) as the test fuel for prescribed operating conditions. The resulting octane number corresponds to the volumetric percentage of isooctane in the PRF mixture. The RON test conditions are: intake air temperature of 52 °C and ambient pressure, spark ignition timing 13° before top dead center (bTDC), and an engine speed of 600 rpm. The MON test conditions are: intake air temperature of 149 °C and ambient pressure, spark ignition timing of 26° to 14° bTDC depending on engine CR, and an engine speed of 900 rpm. Both tests use a variable CR engine (ranging from 4 to 18), and fuels typically tested using these methods have octane ratings between 40 and 100. Due to the low speeds and high intake temperatures used in these tests, numerous studies have found that fuel reactivity cannot be adequately described by the RON and MON tests; many fuels exhibit a strong sensitivity to engine operating conditions [9,10]. Therefore, it is useful to have the ability to test fuels of interest at a variety of different engine-relevant operating conditions; the current testing method will be described in more detail later in the paper. Cetane number (CN) is found using a variable CR pre-chamber diesel engine according to ASTM D613 [3]. The cetane number of a fuel, refers to the mixture of cetane number primary reference fuels isocetane (2,2,4,4,6,8,8-heptamethylnonane) and n-hexadecane that exhibits the same ignition delay of 13 crank angle degrees (CAD) at the same CR for a set engine operating condition. The cetane test conditions specify an intake air temperature of 66 °C and ambient pressure with a start of injection (SOI) occurring in the pre-chamber at 13° before top dead center. Compression ratio, which can vary between 8 and 36, is adjusted until ignition occurs at TDC. Typical diesel fuels tested have cetane numbers between 30 and 65. In practice, due to the high cost of the cetane primary reference fuels, secondary reference fuels (SRFs) T and U, which are tightly controlled full-blend fuels made by Chevron Phillips, are used for testing. The specific SRFs used in this study are T-26 and U-19, having tested cetane numbers of 75.2 and 19.4, respectively. Formerly, a direct comparison between gasoline-like fuels described by octane number and diesel-like fuels specified described by cetane number was not necessary since they had separate applications; high-octane (low reactivity) fuels were predominantly used in SI engines while high-cetane (high reactivity) fuels were used in CI engines. However, due to the increased interest in LTC, combustion strategies are being investigated involving the compression-ignition of low-reactivity fuels [4], and the use of mixtures of low- and high-reactivity fuels in dual-fuel combustion strategies [5]. Many of these strategies utilize fully or partially premixed quantities of fuel at
Fig. 1. Cetane number as a function of RON. Data compiled from [6] and [7].
much lower than stoichiometric equivalence ratios. An advantage of using a HCCI testing method is that it allows for the ignition quality of the fuel to be characterized fully by its kinetic behavior while avoiding fuel spray injection complications. Computational fluid dynamics (CFD) is an important tool for the investigation and development of low temperature combustion strategies. However, the implementation of detailed chemistry of large molecules, like those found in diesel fuel, in CFD is still beyond the capability of current computing power for engineering applications. Thus, simplified chemistry needs to be used. The most common surrogate for diesel fuel in CFD is nheptane because skeletal chemical kinetics models are available. Skeletal models also exist for the PRF blends. A well-established link between cetane number and octane number would potentially allow more accurate CFD modeling of fuels, i.e., it would provide an easy-to-implement means for varying the cetane number. Direct comparisons of the cetane and octane number have been undertaken using the results of RON/MON and cetane testing. Kalghatgi [6] measured the cetane number of PRF blends and mixtures of n-heptane and toluene, and compared them to their measured (or known) research octane numbers (RON). The results showed a linear correlation between the two metrics. The National Renewable Energy Laboratory (NREL) [7] compiled results for an array of single-component hydrocarbons, including oxygenates, and also recovered a linear relation between RON and cetane number. These results are reproduced in Fig. 1, including confidence intervals that arise from the linear fitting. The results in tabular form can be found in
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
JID: PROCI
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7 Table 1 Engine geometry. Compression Ratio Displacement (L) Stroke (mm) Bore (mm) Con Rod Length (mm) Intake Valve Closing Exhaust Valve Opening Swirl Ratio Piston Bowl Type
9.5 0.477 90.4 82 145.54 −132 °aTDC 112 °aTDC 1.5 Stock (Re-entrant)
Appendix A. Both data sets show a strong correlation between CN and RON over the limited range of CNs tested. The NREL correlation covers a larger range of CNs but has a larger scatter in the data resulting in a weaker correlation between CN and RON (as seen by the larger confidence interval bands and an R2 value of 0.87 compared to an R2 value of 0.98 for Kalghatgi’s data). In the current study, mixtures of the cetane secondary reference fuels T-26 and U-19 were tested using a novel fuel-substitution methodology that provides an effective PRF (octane) number for each tested fuel mixture. The main features of the fuel substitution method [8] are: engine tests are performed using homogenous-charge compressionignition (HCCI) combustion to avoid fuel spray and flame propagation complications while allowing engine-relevant conditions to be investigated; combustion phasing is controlled independently of intake and boundary conditions allowing the fuel chemistry to be isolated; multiple engine operating conditions are tested to cover a range of temperature and pressure regimes; and a wide CN range of fuels can be tested due to an absence of the requirement of the effective octane number being between 0 and 120. The octane number test limits of 0–120 correspond to the octane number limits of the reference fuels n-heptane (0) and isooctane mixed with 6 mL/gallon of tetraethyllead (120.3) [1]. The main objective of this study is to investigate the relationship between the CN of the SRF fuel blends and the ON of the PRF mixtures that exhibit matched ignition characteristics (reactivity) independent of fuel spray complications at representative LTC operating conditions.
3
and the intake air temperature was controlled using heaters both upstream of and surrounding the outer walls of the intake surge tank. Fuel pressure and injection timing were controlled using the National Instruments (NI)Drivven hardware/software suite. NI hardware was also used for all high- and low-speed controls such as intake air pressures and temperatures, coolant temperatures, and the exhaust gas recirculation (EGR) valve. In-cylinder pressure was measured using a Kistler 6125A pressure transducer in series with a Kistler 5010 charge amplifier. A 0.25° resolution encoder fixed to the crankshaft was used as the timing clock for the high-speed data acquisition, giving a resolution of about 22 μs, at the tested speed. Pressure data were acquired for 300 cycles for each test condition. Horiba gas analyzers were used to determine relevant emissions measurements. The engine was operated in an HCCI combustion mode. Three fuel streams were delivered to the engine; n-heptane and isooctane were independently port injected, and the SRF fuel mixtures were delivered from an upstream prevaporizing system. To ensure the complete vaporization of the low-volatility fuels, an air-assisted fuel injector was used to inject a highly atomized fuel spray into the main intake air stream directly downstream of an in-line heater. Tape heaters were utilized to maintain the downstream wall temperatures at the gas temperature up to the engine intake port. At the engine operating speed of 1900 rpm, which was used for all tests, the fuel has a residence time of approximately 3.5 s in the intake system. The engine speed was chosen in order to achieve reasonable combustion phasing for the intake temperature and pressure conditions. Due to the relatively high intake temperatures, tests were performed to verify that no fuel decomposition reactions occurred in the intake system. Heptane, which has similar ignition characteristics to the diesel fuels tested, was injected both through the upstream pre-vaporizer system and at the intake port. Pressure and heat release data were analyzed, and no measureable effect was found, verifying that intake temperatures up to 200 ºC likely had no effect on the combustion characteristics for the residence times used in the experiments. Details about the prevaporizing system are described in more detail in [8].
2. Experimental setup 3. HCCI constant phasing fuel blend studies Engine experiments were carried out in a singlecylinder version of the General Motors/Fiat JTD 1.9 L four-cylinder diesel engine. Engine specifications are given in Table 1. In order to maintain reasonable fuel ignition timing, the compression ratio was reduced from a stock value of 16.7–9.5.The stock four-cylinder engine head was mounted to a Ricardo Hydra single-cylinder block. The air flow rate was controlled using choked flow orifices,
In this study, the fuel substitution strategy described in [8] was utilized. It was shown in [8] that, at a minimum, combustion phasing, load and intake temperature must be kept constant in order to keep the boundary conditions, such as the wall and piston temperatures, constant. In previous work it was shown that there is a strong connection between piston temperature and combustion phasing,
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
JID: PROCI
4
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
Fig. 2. Pressure and AHRR for SRF20 Substitution Sweep.
and piston temperature also has a strong influence on ignition. The reader is referred to [8] for details. The fuel substitution strategy works as follows. An operating point is first run using only the portinjected PRFs. The test fuel is then introduced through the upstream pre-vaporizer system at the desired mass fraction; the port fuel mass is reduced to maintain the engine load, and the isooctane-ton-heptane ratio is adjusted to maintain the same combustion phasing. Figure 2 shows the pressure and apparent heat release rate (AHRR) for SRF20, i.e., a mixture of SRFs that gives a CN of 20, substitution. For all of the sweeps run, the typical variation of key combustion parameters were as follows: IMEP varied by less than 1%, CA50 by less than 0.5° from the mean, and the IVC temperature determined using the trapped mass and ideal gas law varied by less than 3 °C about the mean. It is seen that nearly identical in-cylinder thermal conditions and combustion performance were achieved for the entirety of the sweep, where yhf is the mass fraction of the heavy fuel, i.e., the SRF blend. Figure 3 shows the required PRF number of the port-injected PRF fuel mixture (PRFmeas ) required to keep phasing and load constant as a function of heavy-fuel mass fraction (yHF ) of SRF20. Since the IVC temperature, load, and combustion phasing are kept constant, the total fuel mixture reactivity, which is expressed as the overall PRF number, should be constant throughout the sweep and be approximately equal to the baseline PRF mixture when no heavy fuel is injected upstream, i.e., 50 for the case shown in Fig 3. It is evident that the SRF20 (CN=20) mixture has a lower reactivity than a PRF50 mixture. Therefore, the port-injected PRF mass ratio had to be adjusted such that more n-heptane was injected, i.e., the PRFmeas was decreased, as the mass fraction of SRF20 increased.
Fig. 3. Measured PRF as a function of the mass fraction of SRF20.
If the resulting mixture reactivity obeys a linear mass-based blending rule, the required PRF vs. heavy fuel fraction data can be used to calculate an effective PRF number for the heavy fuel, PRFHF . The effective PRF of the three-component fuel (nheptane / isooctane / heavy fuel) mixture, PRFmix , which is constant as discussed above, can be written as PRFmix = (1 − yHF )PRFmeas + yHF PRFHF
(1)
where PRFmeas is the port-injected mixture PRF number. This equation has two unknowns, PRFmix and PRFHF , that can be solved for using a leastsquares approach. This method has been used in previous studies to calculate an effective PRF number for diesel-like fuels, and the linear blending rule was seen to accurately describe the overall mixture reactivity [8]. As seen in fig. 3, an effective PRFHF of 86.5 is calculated for the SRF20 mixture, and the predicted PRFmeas as a function of SRF20 mass fraction matches very well with the experimental data points, indicating that the linear blending rule accurately describes the overall mixture reactivity for this case. Figure 4 shows the raw data and the curve fits using Eq. (1) for all of the SRF mixtures tested at this operating condition. It can be seen that the fits to the data are excellent and correctly capture the curvature of the data points suggesting that, despite its simplicity, the linear blending rule applied in Eq. (1) adequately represents the results. In order to investigate whether the effective PRF of the heavy fuel is dependent on the engine operating conditions, four different operating conditions were tested. An overall mixture PRF number of 20 and 50, which gave rise to
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
JID: PROCI
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
5
Table 2 Engine operating conditions for constant phasing sweeps. Operating condition
Intake P [bar]
Intake T [°C]
Gross IMEP [bar]
1 2 3 4
1.2 1.2 2 2
165 165 165 165
4.2 4.6 5.4 5.6
CA50 [°aTDC] −8 3 −15 −11
[-]
T (CA10) [K]
0.36 0.36 0.28 0.28
850 925 775 800
ρ(CA10) [kg/m3] 7.8 10.2 9.9 10.5
Baseline PRF [-] 20 50 20 50
Fig. 5. Calculated SRF PRF for all sweep conditions. Fig. 4. Measured PRF and the regression using the linear blending model for a range of SRF mixtures.
substantially different combustion phasing, were each run at intake pressures of 1.2 bar and 2.0 bar. A summary of operating conditions can be seen in Table 2. Temperature and density are estimated at the crank angle that 10% of the cumulative heat release is reached (CA10). Density at CA10 is modest due to the high intake temperatures required to ensure full fuel vaporization and the low compression ratio. The density for operating condition 2 compared to condition 1 is substantially higher due to the later combustion phasing, causing CA10 to occur closer to TDC. Figures 5 and 6 summarize the results from all of the data collected. Figure 5 shows the effective PRF values obtained for each SRF mixture tested, with confidence intervals, calculated at each of the four operating conditions given in Table 2. The horizontal lines represent the average for each fuel mixture. Varying thermodynamic operating conditions such as intake temperature and pressure can have a large effect on fuel chemistry and ignition behavior; however it can be seen that over the range of conditions tested the operating condition did not have a significant influence on the effective PRF number when the uncertainties are accounted for. Figure 6 shows the direct correlation between the effective PRF number and the cetane number
Fig. 6. Summary of experimental data with linear fit and CI.
along with a best fit line. The linear fit was calculated using a weighting function based on the individual condition confidence intervals. The linear fit is shown with 95% confidence intervals. The best-fit line relating PRF and cetane number (CN) is given
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
ARTICLE IN PRESS
JID: PROCI
6
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
Fig. 7. Comparison of experimental correlation of current study with the results from with previous studies [6,7]. Table 3 Summary of cetane vs. RON/PRF curve fits [6,7].
Current testing Kalghatgi NREL
Slope
Y-intercept
−1.96+/−0.07 −2.34+/−0.28 −2.18+/−0.31
125+/−4 128+/−10 132+/−10
as PRF = 125 − 1.96 CN 2
(2)
with an R value of 0.99. In general, there is an excellent correlation between the effective PRF number measured using the fuel-substitution method and the cetane number. Figure 7 shows the linear correlation from Eq. (2) along with the linear fits to the data of Kalghatgi [6] and NREL [7], previously shown in Fig. 1. The 95% confidence intervals are included for all three linear fits. There appears to be a noticeable difference in the slope of the fit to the current data relative to the other two correlations from the literature. However, within the limits of uncertainty, there is no significant difference between all three sets of data in terms of slope or intercept of the curve fits. A summary of the calculated slopes and y-intercepts along with the corresponding uncertainties can be seen in Table 3. The Kalghatgi data were obtained using the ASTM D613 Cetane test on fuels of known RON, whereas the NREL data were obtained using a variety of Cetane tests (ASTM D613, ASTM D6890, and ASTM D7170) and the ASTM D2699 RON test. The conditions for the current measurements differ substantially from those for the NREL and Kalghatgi correlations. For the present data, igni-
Fig. 8. Comparison between fuel substitution CA10 conditions and estimated RON, MON, and cetane test conditions [11,12].
tion occurs at conditions that are quite lean and with a homogeneous air–fuel mixture, whereas for the RON, MON, and cetane number tests ignition takes place at richer conditions, and the cetane number tests occur with significant inhomogeneity due to the indirect fuel injection. Further, HCCI combustion is purely based on chemical kinetics, while RON and MON tests are also influenced by physical properties that affect flame speed, and the cetane number test is affected by physical properties involved with the fuel injection, vaporization, and mixing processes. Another issue that could influence the comparison between the different measurement procedures is the temperature-density region that the different experiments encompass. Figure 8 shows the estimated thermodynamic conditions for the four operating conditions (as seen in Table 2) at CA10, which is used as a proxy for the ignition timing. The RON, MON, and CN curves in Fig. 8 were taken from the simulations of Yates et al. [11,12]. This simulation work involved modeling the RON, MON, and CN engine experiments in order to estimate in-cylinder thermodynamic conditions. It can be seen that a much larger range of temperatures and densities encompasses the RON or MON tests than is seen for the cetane test. This is partly due to a smaller range of compression ratios required for the typical range of fuels tested with the cetane test. Another cause for the RON and MON tests reaching a higher temperature-density operating zone than the cetane test is that as spark ignition occurs, the end-gas is further compressed by increasing cylinder pressures caused by combustion occurring behind the flame front. Overall, the temperaturedensity data for the current testing can be seen to
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015
JID: PROCI
ARTICLE IN PRESS
[m;October 3, 2016;11:42]
D. Janecek et al. / Proceedings of the Combustion Institute 000 (2016) 1–7
agree quite well with the operating conditions seen in the other tests. The range is smaller than the spread seen in the RON and MON tests due to operating constraints involved with HCCI combustion. The current results allow a significantly wider range of fuel ignition quality to be tested, although it is acknowledged that negative PRF numbers may have limited direct applicability at present.
7
CN and RON data for a number of single component hydrocarbons, including oxygenates reproduced from [7] CN RON CN RON
8 104 28 73
12 109 29 83
12 118 30 62
14 98 30 74
15 106 31 59
17 83 32 55
17 98 34 73
19 78 36 46
21 91 41 28
22 71 41 29
22 93 43 56
23 84 44 43
24 75 47 25
24 92 54 0
27 77 57 22
References 4. Conclusions A strong linear correlation was found between cetane number, which was defined by the mixture ratio of secondary reference fuels tested, and octane number, which was defined as the effective PRF number obtained using a novel fuel- substitution method, was found over cetane numbers from 20 to 75. This relationship was found to be similar to trends seen under substantially different operating conditions previously reported in the literature; it deviated most significantly at the upper limit of cetane numbers previously tested. The current results cover a wider range of cetane number than previous literature measurements and span a different range of conditions. The advantage of the method used in the current study relative to enginebased MON, RON, and cetane number testing is that it can be used to test high-reactivity fuels and fuels with low volatility that would not be practical to test in a spark-ignition engine. A unique aspect of the fuel-substitution method is that it can provide negative values of octane number, i.e., fuels with higher reactivity to n-heptane can be tested. While this testing was done at low equivalence ratios and representative LTC conditions, the method could easily be expanded to different operating conditions such as varying engine speed, higher intake pressures and equivalence ratios, exhaust gas residuals (EGR) dilution, etc.
Acknowledgments Support for this work was provided by the Office of Naval Research (ONR) Contract N00014-12-10650, Sharon Beermann-Curtin, grant monitor.
Appendix A CN and RON data for a number of PRF blends and mixtures of n-heptane and toluene reproduced from [6]
[1] American Society for Testing Materials: Designation: D2699, “Standard Test Method for Research Octane Number of Spark-Ignition Engine Fuel”, ASTM Inernational, West Conshohocken, PA [2] American Society for Testing Materials: Designation: D2700, “Standard Test Method for Motor Octane Number of Spark-Ignition Engine Fuel”, ASTM International, West Conshohocken, PA. [3] American Society for Testing Materials: Designation: D613, “Test Method for Cetane Number of Diesel Fuel Oil”, ASTM International, West Conshohocken, PA. [4] Y. Ra, P. Loeper, R. Reitz, et al., Study of High Speed Gasoline Direct Injection Compression Ignition (GDICI) Engine Operation in the LTC Regime, 2011 SAE Technical Paper 2011-01-1182, doi:10. 4271/2011- 01- 1182. [5] S.L. Kokjohn, R.M. Hanson, D.A. Splitter, R.D. Reitz, Int. J. Engine Res. 12 (2011) 209–226, doi:10. 1177/1468087411401548. [6] G.T. Kalghatgi, Auto-Ignition Quality of Practical Fuels and Implications for Fuel Requirements of Future SI and HCCI Engines, 2005 SAE Technical Paper 2005-01-0239, doi:10.4271/2005- 01- 0239. [7] J. Yanowitz, M.A. Ratcliff, R.L. McCormick, J.D. Taylor, M.J. Murphy, Compendium of Experimental Cetane Numbers, National Renewable Energy Laboratory, 2014 Technical Report 5400-61693. [8] Janecek, D., Rothamer, D., and Ghandhi, J., “Fuelsubstitution method for investigating kinetics of low volatility fuels under engine-like operating conditions”, 9th U.S. National Combustion Meeting 114IC–0377. [9] G.T. Kalghatgi, Fuel Anti-Knock Quality – Part I. Engine Studies, 2001 SAE Technical Paper 2001-013584, doi:10.4271/2001- 01- 3584. [10] G.T. Kalghatgi, Fuel Anti-Knock Quality – Part II. Vehicle Studies – How Relevant is Motor Octane Number (MON) in Modern Engines?, 2001 SAE Technical Paper 2001-01-3585, doi:10.4271/ 2001- 01- 3585. [11] D.B. Yates, A. Swarts, C.L. Viljoen, Correlating Auto-Ignition Delays and Knock-Limited SparkAdvance Data for Different Types of Fuel, 2005 SAE Technical Paper 2005-01-2083, doi:10.4271/ 2005- 01- 2083. [12] D.B. Yates, A. Swarts, C.L. Viljoen, Understanding the Relation Between Cetane Number and Combustion Bomb Ignition Delay Measurements, 2004 SAE Technical Paper 2004-01-2017, doi:10.4271/ 2004- 01- 2017.
CN 26 26 26 32 32 33 36 36 39 42 47 RON 70 69 69 50 50 50 42 41 40 30 20
Please cite this article as: D. Janecek et al., Investigation of cetane number and octane number correlation under homogenous-charge compression-ignition engine operation, Proceedings of the Combustion Institute (2016), http://dx.doi.org/10.1016/j.proci.2016.08.015