Ignition Behavior of Biodiesel and Diesel under Reduced Oxygen

Sep 17, 2015 - Although biodiesel produced slightly longer physical ignition delays as a result of an extended atomization and evaporation process, it...
0 downloads 9 Views 1MB Size
Subscriber access provided by UNIV OF CAMBRIDGE

Article

Ignition behavior of biodiesel and diesel under reduced oxygen atmospheres Michael P. Mayo, and André Louis Boehman Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.5b01439 • Publication Date (Web): 17 Sep 2015 Downloaded from http://pubs.acs.org on September 17, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Ignition behavior of biodiesel and diesel under reduced oxygen atmospheres Michael P. Mayo†‡ and André L. Boehman‡* †

Department of Chemical Engineering, Penn State University, University Park, PA 16802 ‡

Walter E. Lay Automotive Laboratory, University of Michigan, Ann Arbor, MI 48109

KEYWORDS.

Biodiesel,

soybean,

diesel,

ULSD,

autoignition,

ignition

delay,

chemiluminescence, high speed imaging, spray, constant volume combustion chamber, CID, low temperature combustion, LTC, EGR

ABSTRACT

The effects of atmospheric oxygen concentration on the physical and chemical ignition behavior of biodiesel and diesel fuel blends were investigated using an optically-accessible constant volume combustion chamber (600°C and 20 bar) equipped with a light-duty diesel injector. High speed imaging was used to measure the transient cone angle and penetration length of the liquid spray jet as it developed within the chamber. Structural fluctuations of the developing spray jets were suggested to be a function of injector pin movement, with their frequency and amplitude a function of fuel viscosity and density. Physical and chemical ignition delay was analyzed through detection of excited formaldehyde (CH2O) and hydroxide (OH) chemiluminescence using a system of optical filters and photomultiplier tubes (PMTs). Although biodiesel produced

ACS Paragon Plus Environment

1

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

slightly longer physical ignition delay values as a result of an extended atomization and evaporation process, its chemical ignition delay was much shorter than diesel, attributed to the long chain hydrocarbon fuel structure and fuel oxygen availability. From an engine control and performance standpoint, the short ignition delay and robust heat release profile of biodiesel produced under simulated low temperature combustion (LTC) operating conditions demonstrates the compatibility of biodiesel for use in next generation compression ignition (CI) engines.

INTRODUCTION As automotive manufacturers continue to optimize the compression ignition (CI) engine by exploring technologies that increase fuel economy and mitigate the production of toxic emissions, many have found benefits in using combustion strategies categorized under “low temperature combustion” (LTC). In order to guide the chemistry of combustion towards a cleaner conversion of energy at a high efficiency each proposed LTC strategy utilizes some form of modification to the intake oxygen concentration, fuel injection timing, and fuel type. With the introduction of government regulated fuel standards, the choice of fuel type has gained greater focus. One such fuel type that has been attractive for next generation engines is biodiesel due to its high renewable content and low carbon impact to the environment. A major benefit of using a LTC strategy is the reduction of nitrogen oxides (NOx) and particulate matter (PM), achieved primarily by reducing the oxygen content in the intake air via exhaust gas recirculation (EGR). Reduced atmospheric oxygen concentration results in lowered peak combustion temperatures known to promote the formation of these pollutants.1 These strategies have been studied extensively and proven to be quite effective with diesel,2 however

ACS Paragon Plus Environment

2

Page 3 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

the added benefits of utilizing a renewable fuel such as biodiesel have driven recent studies to focus on its behavior in a LTC environment..3 Understanding the physical and chemical ignition behavior differences between diesel and biodiesel spray combustion is essential towards optimizing their use in CI engines. Similar work by Nerva4 and Genzale5 focusing on physical behavior found that biodiesel spray jets maintained greater liquid-phase momentum than diesel resulting in greater penetration depths; a situation that may promote wall-impingement and consequently lube oil dilution in an engine. For this paper, the influence of fuel properties and atmospheric oxygen concentration on biodiesel/diesel ignition are examined with high speed optical systems to capture liquid spray development and intermediate chemical reaction chemiluminescence. In addition to comparing results to previously published literature, this work intends to provide a deeper understanding from a physical and chemical stand-point into the behavior of biodiesel/diesel ignition under reduced oxygen atmospheres, similar to conditions ran in next generation CI engines that utilize a LTC strategy. Moreover, this work uses an ASTM standard instrument for measurement of the derived cetane number (DCN) as the basis for the experiments. Addition of optical access and combustion air dilution to this ASTM standard instrument provides a unique means of exploring the variations in spray and combustion behavior of fuels, which complements the many prior studies of biodiesel combustion in IC engines, optically accessible IC engines, spray chambers, flow reactors and rapid compression machines. This work also complements other ongoing efforts to understand and characterize the ignition behavior of conventional and alternative fuels, such as the “Compendium of Experimental Cetane Numbers” which has been compiled from various instruments including constant volume combustion chambers for cetane rating such as used in the present work.6

ACS Paragon Plus Environment

3

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 36

High Speed Spray Imaging. Understanding the physics of liquid spray injection is essential for researchers to develop more accurate computational models to optimize their combustion strategy. This starts with providing a detailed characterization of the spray injection process to observe the behavior of pressurized liquid fuel (on the order of 1000 bar) as it enters a closed combustion chamber at a high velocity to mix with heated combustion air. Capturing a developing spray has most commonly been achieved with a high speed camera and some form of illumination either by light-scatter or light-extinction methods. Light-scattering illumination involves directing the light source at the side7 or the head of the spray jet.8 In light-extinction, the light source is directed towards the backside of the spray, initially passing through either a diffuser9 or collimating lens (Schlieren imaging).10 Pickett investigated the uncertainty of liquid length measurements and found that light-scatter methods are sensitive to illumination source orientation, where varied maximum intensity locations are created.11 Generally, sophisticated post-processing algorithms that utilize statistical analysis have been shown to reduce the effects of these intensity variations in light-scattering setups.12 By coupling high speed imagery with statistical algorithms, liquid spray development can be quickly characterized and used to verify computational spray models. In order to quantify spray jet features such as cone angle and penetration length, each image is pre-processed, segmented according to pixel value, and geometrically analyzed.7, 13-15 Generally, pre-processing techniques mitigate noise that can be falsely identified as the spray by employing the use of filters or manual removal of background objects (artifacts).13 Then, in order to segment images accurately, each pre-processed image is statistically analyzed to automatically calculate an optimal threshold value; a method researchers commonly use in order to significantly reduce processing time. Each digital pixel is identified as either spray or

ACS Paragon Plus Environment

4

Page 5 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

background by converting grayscale pixels (values ranging from 0 to 255) to black and white (values of 0 and 1, respectively) according to the selected “threshold” value. Macian evaluated the two most common statistical methods employed in spray characterization, Otsu’s thresholding method (OTM) and Likelihood Ratio Test (LRT), and found that LRT was more reliable than OTM due to the inaccuracies of OTM created by a significant sensitivity to mean luminosity variations, discussed above.12 LRT was first adapted to spray characterization by Pastor who calculated optimal thresholds by fitting independent Gaussian distributions to the spray and background histograms15; later the method was expanded to cover non-Gaussian probability density functions (PDFs) that were characteristic of background distributions observed in greater variety of light-scattering experimental setups.13 Once segmented, the spray’s physical features such as cone angle and liquid penetration are calculated through simple edge detection and integrative fitting methods. The macroscopic spray measurements captured through image analysis are then used to validate computational models developed for rapid prototyping of combustion chambers. Chemiluminescence of Combustion. A powerful method to characterize the chemical behavior of ignition is through optical observation of photons emitted via chemiluminescence; the spontaneous emission of a photon from an electronically-excited molecule of as the result of a chemical reaction.16 The phenomenon known as autoignition occurs when evaporated fuel molecules mix with oxygen at an optimal ratio, temperature, and pressure to promote a transition from low temperature reactions to high temperature oxidation. The process involves a complex reaction mechanism, of which a few key intermediate reactions can be optically identified according to the wavelengths of photons emitted by excited intermediate radical species. Researchers have used the chemiluminescence of excited CH2O*, OH*, CH*, and C2* as

ACS Paragon Plus Environment

5

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 36

markers for the transitions between the major hydrocarbon oxidation temperature regimes.17, 18 An illustration of measured emission spectra for these key species is displayed in Figure 1, adapted from work by Sheinson19 and Docquier.17 Excited formaldehyde (CH2O*) chemiluminescence, characterized by a broadband emission spectrum20, has been observed during low-temperature hydrocarbon oxidation reactions.19 Coincidentally, the CH2O* emission spectrum overlaps the spectra of CH* and C2*, however these species are not intrinsic to low temperature flame chemistry.19, 21 Therefore, photon emission at 430nm in typically marked as CH2O* emission below 1000K, signifying the onset of low-temperature combustion chemistry.1 OH* chemiluminescence is most commonly marked as the onset of high temperature combustion chemistry with a strong emission at 307nm.16, 22 The key reaction responsible for the formation of excited OH* has been identified by Dandy to be CH + O2 → OH* + CO.23 By capturing the time points at which these intermediate reactions occur with high speed optical sensors, the chemical behavior of the combustion system can be accurately characterized to support computational reaction mechanism development.

ACS Paragon Plus Environment

6

Page 7 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 1. Emission spectra for key hydrocarbon oxidation intermediate species, adapted from Sheinson19 and Docquier17

In-situ measurement of photon emissions provides a clear picture of the chemical reactions that take place during the ignition delay time period. Researchers have employed laser-induced chemiluminescence methods which are very effective but costly and require ample optical access to the combustion chamber.2 Another effective but less costly technique is through use of passive optical equipment to capture the natural chemiluminescence. The in-situ detection of excited radical chemiluminescence has been achieved through the use of band-pass filters and photomultiplier tubes (PMTs) coupled to a high speed data acquisition system.24 In this paper, we investigate the use of PMTs to accurately detect light emissions in a constant volume combustion chamber (CVCC) outfitted with a six-jet fuel injector common in light-duty diesel engines. We anticipate this work to provide a successful demonstration toward characterizing the chemical ignition delay in diesel combustion systems where the use of species detection by laser excitation is not possible. EXPERIMENTAL SECTION Test Fuels. The biodiesel (Peter Cremer NEXSOL BD-0100) selected for these experiments was produced from soybeans in strict accordance with ASTM D6751 (EPA 4627) guidelines. In order to compare biodiesel ignition behavior to a conventional hydrocarbon fuel, ultra-low sulfur diesel (ULSD) produced by ExxonMobil was selected as the baseline fuel sample. Additionally, blends of the two fuels at volumetric ratios (Table 1) were tested to capture any potential synergistic combustion behaviors. A summary of fuel properties is listed in Table 2.

ACS Paragon Plus Environment

7

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 36

Table 1. Biodiesel blend compositions

Fuel Name B100 B80 B60 B40 B20 ULSD

Soybean Biodiesel (Volumetric Fraction) 1.00 0.80 0.60 0.40 0.20 0.00

ULSD (Volumetric Fraction) 0.00 0.20 0.40 0.60 0.80 1.00

Table 2. Chemical properties of fuel samples

Property

ULSD

Lower Heating Value, MJ/kg Density @ 20°C, kg/m3 Kin. Viscosity @ 40°C, mm2/sec Cetane Number T90, °C

42.6 831.8 1.7-4.1 45.1a 312.1d

a

Soybean Biodiesel 37.4 880 4.0c 48b 353.0e

ASTM D6890; bASTM D613; cASTM D445; dASTM D86; eASTM D1160

An analytical analysis (published by Mueller et al.25) found Peter Cremer soybean biodiesel to primarily contain methyl palmitate (C16:0), methyl stearate (C18:0), methyl oleate (C18:1), methyl linoleate (C18:2), and methyl linolenate (C18:3); of which methyl linoleate (C18:2) and methyl oleate (C18:1) accounted for over 75% of the mole fraction. The structures of these molecules are illustrated in Figure 2.

ACS Paragon Plus Environment

8

Page 9 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 2. Structures of major molecular compounds present in Peter Cremer soy-derived biodiesel

The total aromatic content of the ULSD used in these studies was determined to be 31.54% by weight (ASTM D5186). The remaining components included n-alkanes, iso-alkanes, and cycloalkanes at unknown concentrations. Constant Volume Combustion Chamber. A commercially available Cetane Ignition Delay (CID510) instrument manufactured by PAC L.P. (Houston, TX) was selected for these ignition studies due to its ease of control and constant volume combustion chamber (CVCC) which mimics the combustion chamber from an on-road diesel engine. The CVCC was equipped with a Bosch light-duty diesel injector that delivered fuel at 1000 bar, a pressure similar to an automotive common-rail injection system. Modifications to the CVCC were made to add optical access for investigations into the physical and chemical ignition behavior of liquid fuels. Three access ports were welded to the bottom of the chamber for an optimal view of the spray during fuel injection. Two independent sub-systems occupied these ports: a high speed camera system for physical spray characterization and a photomultiplier tube system for chemiluminescence detection of intermediate chemical species. To provide a reduced-oxygen atmosphere for ignition studies, a gas mixing system developed by Polycontrols Technologies, Inc. (Brossard, Québec, Canada) was installed upstream of the CVCC. The gas mixer used PID-controlled mass flow controllers (MFCs) to dilute air with nitrogen (N2) and carbon dioxide (CO2). A simplified illustration of the overall experimental apparatus is displayed in Figure 3.

ACS Paragon Plus Environment

9

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 36

Figure 3. Modified CVCC of CID510 experimental apparatus

Chemiluminescence Detection System. A chemiluminescence detection system (CDS) was designed to observe the presence of CH2O*, OH*, CH*, and C2* light emissions during the ignition delay period. An air-cooled, UV/Vis optical probe manufactured by SMETec was mounted to an access port at the bottom of the CVCC. The probe was equipped with a 90° wideangle, quartz observation lens for global measurements. The photons emitted during ignition were transferred through an optical fiber and arranged into a homogeneous light beam by a collimating lens at the CDS. The photons within the light beam were sorted out according to their wavelength by passing through a system of dichroic mirrors and band-pass filters before

ACS Paragon Plus Environment

10

Page 11 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

contacting photomultiplier tube (PMT) modules. Two dichroic mirrors in series segregated the light beam according to their cut-off wavelengths of 340 and 460nm. The three resulting channels of light were individually filtered by Edmund Optics band-pass filters at center wavelength values of 307±5nm, 430±5nm, and 515±5nm representing excited OH*, CH* or CH2O*, and C2* photon emission, respectively. To decipher between CH* and CH2O* emissions on the 430nm channel, signals captured before the onset of OH* emission were identified as CH2O* and after as CH*. A PMT module (Hamamatsu, 230-700nm) was positioned after each band-pass filter which generated a 0-5V signal whenever photons came in contact with its anode. A National Instruments high-speed data acquisition (DAQ) card (USB-6366) simultaneously acquired 40,000 samples at 1MHz from each of the three PMT signal channels. The 40 millisecond acquisition period was triggered by the 5V TTL pulse created by the CID circuit board to initiate fuel injection. Each data set was saved to the test computer for postprocessing. The PMT signals were post-processed in a Matlab program written to derive the physical (IDphys) and chemical ignition delay (IDchem) periods. Each PMT signal was smoothed using a 2,000Hz low-pass filter. IDphys was defined as the time period between the start of injection (SOI) and the first significant positive deviation in the smoothed 430nm signal (onset of CH2O* chemiluminescence); an example measurement is shown in Figure 4. The SOI used in this work was defined as the leading edge of the electronic pulse sent to the solenoid of the fuel injector. It is important to note fuel actually entered the chamber 0.240-0.308 ms after the electronic SOI, as confirmed by high speed imaging. IDchem was defined as the time period between the end of IDphys and the first significant 307nm signal (onset of OH* chemiluminescence); similarly, this measurement is detailed in Figure 5. Although IDphys and IDchem contain over-lapping

ACS Paragon Plus Environment

11

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 36

processes,26 the selection of formaldehyde as a transition indicator was made due to its significance in low temperature chemistry. As discussed in greater detail by Musculus,1 formaldehyde is one of the key species formed following the exothermic decomposition of ketohydroperoxide (KHP) species; the reactions that contribute significantly to the onset of firststage ignition. This transition point to significant exothermic reaction chemistry marks the end of the physical delay period.27,

28

For the purposes of this paper, the first instance of observed

formaldehyde chemiluminescence provided a good diagnostic indicator for the apparent “end” of IDphys and “beginning” of IDchem.

Figure 4. Example 430nm voltage signal used to derive IDphys

Figure 5. Example 307nm voltage signal used to derive IDchem

High-Speed Spray Imaging System. Optical access to the CVCC was achieved with a rigid Karl Storz endoscope and custom liquid light guide, each outfitted with liquid-cooled AVL

ACS Paragon Plus Environment

12

Page 13 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

“Aquashaft” mounting sheaths to control endoscope and light guide temperature. Each probe entered the chamber through access ports welded to the bottom external surface of the chamber. A Vision Research Phantom v7.1 monochrome camera equipped with a 50mm f/1.2 lens was mounted to the eye-piece of the endoscope to capture spray images at a rate of 29,197 frames per second (resolution of 128 x 384 pixels, exposure of 20μs). A Karl Storz NOVA 300 Xenon light source supplied high-intensity illumination to the fuel jet at a direct angle for light-scatter studies of the transient spray behavior. Spray images were acquired using Vision Research’s PCC v2.3 software, triggered via the 5V TTL fuel-injection pulse created by the CID’s internal circuit board. The first 100 spray-images were exported in uncompressed AVI format for further analysis. After the spray images were captured, they were post-processed in a Matlab program written to measure liquid length and cone angle of a single liquid fuel jet. An image analysis algorithm (adapted from Pastor et al.13) was used to calculate an optimal image threshold for spray and background identification. Since images were captured with a monochrome camera, each image represented an independent matrix of values ranging 0-255 that corresponded to the grayscale with 0 and 255 representing black and white, respectively. Each image was pre-processed to normalize the background and remove artifacts such as reflections on the chamber walls. A histogram of the pixel values for each image revealed a logarithmic shape of which spray and background pixels could be identified (see example in Figure 6). One-dimensional log-likelihood ratio testing was used to calculate an optimal threshold value for image segmentation, as developed by Pastor et al.13 Each image was converted to a binary matrix in reference to the calculated threshold value. Only one complete spray jet was of interest, therefore identified spray pixels that belonged to the surrounding partial spray jets were removed via spatial selection of

ACS Paragon Plus Environment

13

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 36

known coordinates. Edge detection was accomplished by converting all inner spray pixels values from 1 to 0. The identified spray edges (values of 1) were used to calculate spray penetration length (S) and cone angle (θ), as defined in Figure 7.

Figure 6. Example spray image histogram

Figure 7. Definitions of spray penetration length (S) and cone angle (θ)

Apparent Heat Release Rate Calculation. Apparent heat release rate (AHRR) was derived from the pressure trace data captured by the CID’s pressure transducer at the bottom of the combustion chamber. The large pressure increase associated with ignition often produced acoustic waves that were observed as fluctuations in the pressure data near peak combustion pressures, as illustrated in Figure 8. In order to calculate AHRR more accurately, a 2,500Hz lowpass (Butterworth) filter was used to smooth the pressure signal, as illustrated in Figure 9. This

ACS Paragon Plus Environment

14

Page 15 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

type of signal smoothing has been shown to significantly reduce the noise present in the CID’s pressure trace data without altering the data significantly.29 The term “apparent” in AHRR indicates that the values are approximate, since the exact values could not be quantified solely from the pressure trace data captured in this work. AHRR was derived using thermodynamic principles and combustion equations from Heywood;30 as explained in greater detail in Appendix A. An example AHRR profile is shown in Figure 10 to illustrate the two-stage ignition behavior observed during this work. The AHRR data was used to identify the thermal behavior of the reaction system as it transitioned through the major temperature regimes (low, intermediate, and high) of hydrocarbon oxidation.

Figure 8. Typical pressure trace data illustrating signal fluctuations from an ignition event

ACS Paragon Plus Environment

15

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 36

Figure 9. Smoothed pressure trace data from Figure 8 using a 2,500Hz low-pass filter

Figure 10. Example of a two-stage ignition profile through low, intermediate, and high temperature regimes

Test Procedure. Nitrogen gas (10 bar) forced liquid fuel (160mL) from a reservoir through a 5μm particulate filter and a one-way valve to a pneumatic pressure multiplier. The air inlet solenoid valve opened to fill the CVCC with a premixed O2/CO2/N2 gas mixture from the 1.0 liter reserve tank of the Polycontrols gas mixer. The chamber air pressure was regulated to 20±0.1 bar by actuation of the CVCC inlet and exhaust solenoid valves. The process stalled for approximately 80 seconds to allow for the air to equalize to the specified operating temperature (600°C), monitored by two type-K thermocouples embedded in the side walls of the CVCC. When the temperature and air pressure were met, the hydraulic multiplier pressurized the fuel supply to 1000 bar. Once all conditions were verified by digital sensors, the internal computer initiated fuel injection by sending a 2.5ms 5V TTL pulse to the fuel injector that subsequently opened a solenoid valve, unseating a needle inside the tip of the injector. This allowed fuel to flow around the needle and exit through six orifices (0.1mm diameter) into the chamber. The resulting high velocity fuel jets mixed with the heated air and ignited via physical and chemical processes. A piezoelectric transducer located at the bottom of the CVCC recorded 250

ACS Paragon Plus Environment

16

Page 17 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

milliseconds of pressure trace data at a rate of 50,000 samples/second. The raw data was automatically smoothed by the CID’s internal computer using a 9-point moving average and then reduced to a 40μs time step resolution before they were exported to an attached USB thumbdrive. In addition to a pressure-trace, light emission curves and spray videos were captured via the CDS and camera optical systems. After the combustion process completed, the exhaust solenoid valve opened to release the hot pressurized air to venting. The chamber was quickly refilled with fresh air (10 bar) and then evacuated to flush out any remaining combustion products that may contaminate the next injection. This injection process was repeated 15 times per test. Given the statistically small sample size (15) for each data set, experimental uncertainty was assessed using the Student’s t-test in order to calculate a 95% confidence interval. Finally, each reduced-oxygen gas mixture tested and the corresponding oxygen equivalence ratios (φΩ) are reported in Table 3. Oxygen equivalence ratios were calculated following the procedure discussed by Mueller.31 Table 3. Experimental gas mixture compositions and corresponding oxygen equivalence ratios (φΩ)

Gas Mixture # 1 2 3 4 5

O2

CO2

N2

(% by volume)

(% by volume)

(% by volume)

φΩ,

φΩ,

ULSD

B100

20.8 15.9 12.8 10.7 9.6

0.0 3.7 6.3 8.3 9.3

79.2 80.4 80.9 81.0 81.1

0.407 0.513 0.620 0.725 0.800

0.375 0.471 0.567 0.662 0.728

RESULTS AND DISCUSSION This section is broken down into three sub-sections that cover the physical and chemical behavior of spray ignition for each biodiesel/diesel blend. The first sub-section focuses on the physical aspects of liquid spray development during injection as observed by the high speed camera system. This is followed by a section that analyses the overall ignition behavior by characterizing the physical and chemical ignition delay periods identified by the CDS. Lastly the

ACS Paragon Plus Environment

17

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 36

thermodynamic behavior of ignition is discussed in terms of fuel composition and operating conditions. Fuel Spray Behavior. The Bosch light-duty fuel injector delivered pressurized fuel (1000 bar) into the chamber for 2.5 milliseconds through 6 orifices of 0.1 mm diameter. Due to access constraints imposed by the compact design of the CID CVCC only one liquid jet was analyzed. Since fully-developed diesel spray jets have been studied extensively,7, 14, 32 the aim of this subsection was to focus the investigation on the transient behavior following the initial penetration of liquid spray into the combustion chamber. More specifically, three macroscopic characteristics of transient liquid spray behavior are discussed: shape, penetration length, and cone angle. The effect of ambient gas composition on spray structure was not significant under the conditions tested; therefore all physical spray comparisons were conducted on images captured under the standard air condition (Gas Mixture#1). Each set of images captured by the high speed camera was post-processed using the 1Dloglikelihood test to identify the spray edges; the results of edge detection (highlighted in green) for ULSD and B100 are displayed in Figure 11. Here it was observed that the shape of each fuel spray experienced fluctuations in structure throughout the transient period. Qualitatively, B100 appeared to maintain a more stable “plume” shape than ULSD, most likely due to its higher density and viscosity. Furthermore, the higher viscosity and surface tension likely supported a more uniform spray structure due to the greater mean droplet distribution of B100, an observation detailed by Lee.33 Since the B100 spray extended further into the chamber, it was likely a function of larger droplet size creating greater momentum throughout injection.

ACS Paragon Plus Environment

18

Page 19 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 11. Spray images of (a) ULSD and (b) B100 under standard 20.8% oxygen atmospheres at 600°C

The mean liquid penetration length results for each fuel blend (Figure 12) revealed an interesting behavior. Fuel blends with higher concentrations of biodiesel resulted in longer liquid penetration lengths across the entirety of transient spray development. A common observation across fuel blends was that initially the spray jet grew in length at a constant rate, but was quickly followed by a brief moment of quasi-steady behavior before regaining momentum to grow further in length. These fluctuations have been observed by Eagle and discussed to be the result of internal injector geometries and possibly asymmetrical, time-dependent structures within the spray.9 The liquid penetration length fluctuations were experienced at different time points, depending on the fuel composition. Blends with higher concentration of biodiesel experienced the initial drop in growth rate significantly later than ULSD. In general, the tip of the liquid spray jet for fuel blends with higher concentration of biodiesel maintained a higher velocity later into the injection process before experiencing a sharper slow-down, as summarized in Figure 13. The higher velocity and density of biodiesel enabled the spray tip to penetrate the combustion chamber further since these parameters drive momentum. In fact, biodiesel spray

ACS Paragon Plus Environment

19

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 36

reached a greater maximum penetration length than ULSD once fully-developed. Furthermore, researchers have found a strong correlation between T90 distillation temperature and maximum penetration length, suggested to be the function of boiling point on the evaporation process under these conditions.34 Therefore, in addition to the slowed atomization, the higher T90 of biodiesel supported a longer penetration length. In order to ensure adequate combustion efficiency when developing diesel engine injection strategies for higher viscosity/density liquid fuels (in this case biodiesel), greater emphasis should be placed on achieving adequate fuel atomization and mixing.

ACS Paragon Plus Environment

20

Page 21 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 12. Liquid penetration length throughout transient spray development

Figure 13. Liquid spray tip velocity throughout transient spray development

As expected, cone angle experienced fluctuations throughout the period of transient spray development (Figure 14). The un-seated needle motion and fuel viscosity were considered to impact the flow of fuel as it exited the injector tip. The greater viscosity associated with blends of primarily biodiesel generated the widest initial cone angles; proposed to be a function of the energy required to overcome internal liquid-wall interactions. The general trend for all fuels was that cone angle decreased as the spray jet developed towards steady-state. Cone angle oscillations were greatest during the first stages of injection, however they quickly reduced in amplitude as the penetration length reached its maximum value. Each oscillation in cone angle appeared to occur at a different point between fuel types, similar to the behavior seen with penetration length. Again, viscosity and surface tension were expected to play a role in this result.

ACS Paragon Plus Environment

21

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 36

Figure 14. Liquid cone angle measurements throughout transient spray development

To investigate the shape fluctuations observed more closely, the direct relationship of penetration length to cone angle was investigated. In contrast to the independent comparisons of penetration length and cone angle previously covered, the ratio between the two measurements (Figure 15) revealed a unique fluctuation pattern throughout spray development, shared across all fuel blends. Researchers at Argonne National Laboratory studied the needle movement within a multi-hole diesel fuel injector via x-ray diagnostics and found that the needle oscillated as it was unseated from the sac; a movement that was hypothesized to drive modulation of the spray structure as a result of the internal flow-field dynamics.35 Given that the shape oscillations were very similar between fuel types, it was postulated that needle oscillations in the injector tip were the cause of this unique behavior.

ACS Paragon Plus Environment

22

Page 23 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 15. Cone angle to penetration length ratio throughout transient spray development

Ignition Behavior. Characterization of ignition behavior was achieved through investigations into the physical and chemical processes, carried out with reference to the pressure trace and PMT signal data. The chemiluminescence data captured by the CDS provided key insights into the timing of chemical reactions throughout the overall ignition delay period (IDoverall). Additionally, the pressure trace data captured by the pressure transducer provided the opportunity to investigate the atmospheric and fuel compositions effects on AHRR. The physical ignition delay (IDphys) occupied the time spent between start of injection (SOI) and the first sign of excited formaldehyde (430nm PMT signal). Generally, physical processes such as atomization, mixing, and evaporation occupied this time duration, as observed visually by the high speed camera system. Although the prevailing processes occupying IDphys were physical, low temperature chemistry leading to formation of formaldehyde was present. Figure 16 displays each fuel’s IDphys across the atmospheric oxygen concentration span. The effect of oxygen concentration on IDphys raised questions about whether this linear trend was the result of physical or chemical sensitivities. As mentioned earlier, liquid length and cone angle appeared unchanged across varying atmospheric conditions, suggesting that mass entrainment rates were

ACS Paragon Plus Environment

23

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 36

not significantly affected. Considering that the end of IDphys was marked by the first observation of excited formaldehyde, this linear trend may be due to the early-stage low temperature reaction kinetics leading to formaldehyde; a process not measured by the CDS in these studies. Researchers have found that changes in atmospheric oxygen levels have a much greater effect on the reaction rates of ketohydroperoxide (KHP) chemistry leading to formaldehyde than the physical mixing rates of fuel with air.1 Further investigations are needed to fully understand the positive linear trend between IDphys and decreasing-O2 / increasing-CO2 concentration. Overall, Figure 16 suggests that greater biodiesel content resulted in longer IDphys times, but the trend is not statistically significant. It was expected that greater biodiesel content would lead to extended atomization and evaporation processes due to the higher density and viscosity of the biodiesel fuel, which was observed by the high speed camera system. However, Nerva et al. observed in an optically accessible spray chamber that spray penetration and spreading angle differences between diesel and biodiesel were modest and near the boundary of statistical significance.4

Figure 16. Physical ignition delay (IDphys) at varying atmospheric oxygen concentrations

The chemical ignition delay (IDchem) was defined as the time between the end of IDphys and the first significant signal of OH chemiluminescence. Across all fuel blends tested, an exponential

ACS Paragon Plus Environment

24

Page 25 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

relationship between IDchem and oxygen concentration was observed as shown in Figure 17; however, the degree of sensitivity was unique to fuel composition. Blends with greater concentration of biodiesel maintained a short IDchem as less ambient oxygen was available. Researchers have suggested that biodiesel can tolerate reduced-oxygen atmospheres due to its available fuel oxygen.36 It was worth considering the effect of long unbranched hydrocarbon structures predominately available within biodiesel. Higher concentration of straight chain paraffins increases reactivity in the high temperature combustion regime, leading to shorter ignition delays; in contrast, aromatic structures impose the opposite effect.37 The trends observed in Figure 17 may suggest that the high sensitivity of ULSD to oxygen reduction is primarily due to the reaction rate dependence of aromatic species on ambient oxygen concentration.

Figure 17. Chemical ignition delay (IDchem) at varying atmospheric oxygen concentrations

Apparent Heat Release Rate. The apparent heat release rate (AHRR) was derived from pressure trace data captured by the pressure transducer at the bottom of the combustion chamber. Figure 18 summarizes the AHRR results for each fuel blend under varied air composition. This

ACS Paragon Plus Environment

25

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 36

data provided insight into the effect of fuel and air composition on the thermal ignition behavior in a diesel-like combustion system. The AHRR profiles were most different between ULSD and B100 at the standard 20.8% oxygen condition. Here ULSD produced a two-stage ignition profile, a phenomenon that is commonly observed in diesel combustion systems.1, 38 Interestingly, as biodiesel content in the fuel blends increased in concentration, the ignition profile transitioned to a single-stage event; generally biodiesel ignition follows a two-stage ignition behavior.39 Even though the pressure transducer only captured a single stage ignition, the CDS data revealed both low and high temperature chemistry, indicating a rapid transition between first and second-stage reaction kinetics. Further investigations into the transition between single-stage and two-stage ignition of biodiesel are needed to understand this phenomenon. Figure 18 also indicated that the effect of atmospheric oxygen concentration on low and high temperature heat release rates was sensitive to fuel type. The timing of the low temperature heat release (LTHR) profile was relatively unchanged under each gas mixture, regardless of fuel type; however, the high temperature heat release (HTHR) shifted later for blends with a higher fraction of ULSD. The blends with greater biodiesel fractions maintained a sharper, more advanced HTHR as atmospheric oxygen content decreased. If we consider the molecular structure of biodiesel alone, the oxygen content and long chain paraffin characteristics of the alkyl group on the ester likely played a strong role during high temperature reaction chemistry. ULSD is much more dependent on its interactions with molecular O2, which may explain part of its HTHR sensitivity. Shibata et al. concluded that the HTHR event for a given fuel occurs earlier as the concentration of these hydrocarbon species (ordered from the greatest to least effect) increases: n-paraffins > iso-paraffins > olefins > naphthenes = aromatics.37 Under these guidelines,

ACS Paragon Plus Environment

26

Page 27 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

biodiesel’s high concentration of long alkyl chains favored an advanced HTHR. Additionally, ULSD was composed of ~30% aromatic content of which was expected to retard HTHR; this relationship appeared to be amplified by the reduction of atmospheric oxygen.

Figure 18. Apparent heat release rate profiles for ULSD and biodiesel blends under reduced oxygen atmospheres at 600°C

The maximum LTHR rates were calculated for each fuel and plotted against oxygen concentration in Figure 19. The maximum LTHR values for ULSD/biodiesel blends decreased marginally as oxygen concentration was decreased. Blends with higher concentrations of biodiesel resulted in greater LTHR than for neat ULSD. This result was expected because the

ACS Paragon Plus Environment

27

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 36

biodiesel had a higher cetane number than ULSD, a property that correlates to greater low temperature reactivity.40

Figure 19. Maximum low temperature heat release rates for all fuel blends across reduced oxygen atmospheres at 600°C

Maximum HTHR for all fuels was plotted against oxygen concentration in Figure 20. At 20.8% oxygen, the order of fuels with maximum HTHR rates from greatest to least was: ULSD, B20, B40, B60, B80, and then B100. As ambient oxygen concentration reduced past 18%, the ranking order reversed; the impact of atmospheric oxygen on high temperature reactivity appeared much more significant for ULSD than for biodiesel. Zhang et al. also noted the unique reversal in HTHR reactivity between ULSD and biodiesel at about 18% ambient oxygen.41 The increased maximum HTHR benefit of biodiesel was lost as maximum HTHR for all fuel blends began to converge to 400 J/ms around 9% ambient oxygen. Given the greater high temperature reactivity of biodiesel in atmospheres below 18% oxygen, its use in diesel engines that employ exhaust gas recirculation (EGR) may result in greater combustion stability, assuming a premixed condition.

ACS Paragon Plus Environment

28

Page 29 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 20. Maximum high temperature heat release rates for all fuel blends across reduced oxygen atmospheres at 600°C

CONCLUSIONS In this study, the physical and chemical aspects of ULSD/biodiesel spray ignition under reduced oxygen atmospheres were explored. Using an optical CVCC liquid spray behavior was characterized via high speed imaging and analysis. Additionally, chemical behavior including chemiluminescence and heat release profiles were examined. This work uses an ASTM standard instrument for measurement of the derived cetane number (DCN) as the basis for the experiments, and represents a novel approach to the study of spray and ignition behavior. The following conclusions were made: 1. Biodiesel spray jets penetrate deeper into the combustion chamber than ULSD due to a lessened degree of atomization and evaporation; considered to be a function of higher fuel viscosity and density. 2. The liquid spray jet cone angle and penetration length fluctuations seen through transient spray development are possibly a function of the unique injector needle movement upon

ACS Paragon Plus Environment

29

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 36

unseating from the fuel sac. Furthermore, the time delay of each fuel spray’s structural fluctuations is a function of the fuel viscosity and density. 3. Biodiesel maintained shorter ignition delays and higher heat release profiles than ULSD under atmospheres below 18% oxygen concentration, most likely due to a function of fuel oxygen and straight-chain alkyl hydrocarbon structure. The insights captured in this work indicated that biodiesel may be well-suited to perform under LTC conditions. As countries continue to promote energy independence, those capable of producing biodiesel may benefit from combusting it in engines that utilize LTC technology. In order to further optimize biodiesel for LTC, additional studies are needed to develop improved injection and mixing strategies to overcome the increased atomization and evaporation process intrinsic to higher viscosity and density fuel types.

AUTHOR INFORMATION Corresponding Author *Tel.: (734) 764-6995, email: boehman@umich.edu Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding Sources This research was funded by Volvo Group and the U.S. Department of Energy under DOE award number DE-EE0004232 via subcontract from Volvo Group Truck Technology through the "SuperTruck - Volvo Energy Efficient Vehicle Program.”

ACS Paragon Plus Environment

30

Page 31 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ACKNOWLEDGMENTS We would like to express sincere gratitude to PAC, L.P. (Houston, TX) for the reliable technical support during the design and troubleshooting stages of the CID’s modification. We are grateful to the research groups of Prof. Margaret Wooldridge and Prof. Volker Sick for providing the high speed cameras needed to complete the spray visualization study. Lastly, we are thankful to Volvo Group Truck Technology and Volvo Technology of America for their support of this work through the “SuperTruck” program, and in particular to support and guidance from Sam McLaughlin, Arne Anderssen and Pascal Amar. ABBREVIATIONS AHRR, Apparent heat release rate; CDS, chemiluminescence detection system; CID, Cetane Ignition Delay (instrument manufactured by PAC, L.P.); CVCC, constant volume combustion chamber; HTHR, high temperature heat release; IDchem, chemical ignition delay; IDoverall, overall ignition delay; IDphys, physical ignition delay; LTHR, low temperature heat release; PMT, photomultiplier tube; SOI, start of injection; TTL, Transistor–transistor logic. REFERENCES 1. Musculus, M. P. B.; Miles, P. C.; Pickett, L. M., Conceptual models for partially premixed low-temperature diesel combustion. Progress in Energy and Combustion Science 2013, 39, (2–3), 246-283. 2. Dec, J. E., Advanced compression-ignition engines—understanding the in-cylinder processes. Proceedings of the Combustion Institute 2009, 32, (2), 2727-2742. 3. Imtenan, S.; Varman, M.; Masjuki, H.; Kalam, M.; Sajjad, H.; Arbab, M.; Fattah, I. R., Impact of low temperature combustion attaining strategies on diesel engine emissions for diesel and biodiesels: a review. Energy Conversion and Management 2014, 80, 329-356. 4. Nerva, J.-G.; Genzale, C. L.; Kook, S.; García-Oliver, J. M.; Pickett, L. M., Fundamental spray and combustion measurements of soy methyl-ester biodiesel. International Journal of Engine Research 2013, 14, (4), 373-390. 5. Genzale, C. L.; Pickett, L. M.; Kook, S. Liquid penetration of diesel and biodiesel sprays at late-cycle post-injection conditions; SAE Technical Paper: 2010.

ACS Paragon Plus Environment

31

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 36

6. Yanowitz, J.; Ratcliff, M.; McCormick, R.; Tay, J.; Murphy, M., Compendium of experimental cetane numbers. In NREL Report: NREL/TP-5400-61693: 2014. 7. Siebers, D. L., Liquid-Phase Fuel Penetration in Diesel Sprays. In SAE International: 1998. 8. Zhang, L.; Tsurushima, T.; Ueda, T.; Ishii, Y.; Itou, T.; Minarni, T.; Yokota, K., Measurement of Liquid Phase Penetration of vaporating Spray in a DI Diesel Engine. Training 1997, 1999, 09-27. 9. Eagle, W. E.; Morris, S. B.; Wooldridge, M. S., High-speed imaging of transient diesel spray behavior during high pressure injection of a multi-hole fuel injector. Fuel 2014, 116, (0), 299-309. 10. Pickett, L. M.; Kook, S.; Williams, T. C., Visualization of Diesel Spray Penetration, Cool-Flame, Ignition, High-Temperature Combustion, and Soot Formation Using High-Speed Imaging. SAE Int. J. Engines 2009, 2, (1), 439-459. 11. Pickett, L. M.; Genzale, C. L.; Manin, J.; Malbec, L.; Hermant, L., Measurement uncertainty of liquid penetration in evaporating diesel sprays. ILASS2011-111 2011. 12. Macian, V.; Payri, R.; Garcia, A.; Bardi, M., Experimental Evaluation of the Best Approach for Diesel Spray Images Segmentation. Experimental Techniques 2012, 36, (6), 26-34. 13. Pastor, J. V.; Arrègle, J.; García, J. M.; Zapata, L. D., Segmentation of diesel spray images with log-likelihood ratio test algorithm for non-Gaussian distributions. Applied Optics 2007, 46, (6), 888-899. 14. Naber, J. D.; Siebers, D. L., Effects of Gas Density and Vaporization on Penetration and Dispersion of Diesel Sprays. In SAE International: 1996. 15. Pastor, J. V.; Arrègle, J.; Palomares, A., Diesel Spray Image Segmentation With a Likelihood Ratio Test. Appl. Opt. 2001, 40, (17), 2876-2885. 16. Docquier, N.; Candel, S., Combustion control and sensors: a review. Progress in Energy and Combustion Science 2002, 28, (2), 107-150. 17. Docquier, N.; Belhalfaoui, S.; Lacas, F.; Darabiha, N.; Rolon, C., Experimental and numerical study of chemiluminescence in methane/air high-pressure flames for active control applications. Proceedings of the Combustion Institute 2000, 28, (2), 1765-1774. 18. Hwang, W.; Dec, J.; Sjöberg, M., Spectroscopic and chemical-kinetic analysis of the phases of HCCI autoignition and combustion for single- and two-stage ignition fuels. Combustion and Flame 2008, 154, (3), 387-409. 19. Sheinson, R. S.; Williams, F. W., Chemiluminescence spectra from cool and blue flames: Electronically excited formaldehyde. Combustion and Flame 1973, 21, (2), 221-230. 20. Gradstein, S., Uber die Fluoreszenz des gasförmigen Formaldehyds. Zeitschr. f. Physikal. Chemie 1933, 384-394. 21. Pöschl, M.; Sattelmayer, T., Influence of temperature inhomogeneities on knocking combustion. Combustion and Flame 2008, 153, (4), 562-573. 22. Higgins, B.; McQuay, M. Q.; Lacas, F.; Rolon, J. C.; Darabiha, N.; Candel, S., Systematic measurements of OH chemiluminescence for fuel-lean, high-pressure, premixed, laminar flames. Fuel 2001, 80, (1), 67-74. 23. Dandy, D. S.; Vosen, S. R., Numerical and Experimental Studies of Hydroxyl Radical Chemiluminescence in Methane-Air Flames. Combustion Science and Technology 1992, 82, (16), 131-150.

ACS Paragon Plus Environment

32

Page 33 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

24. Tinaut, F. V.; Reyes, M.; Giménez, B.; Pastor, J. V., Measurements of OH* and CH* Chemiluminescence in Premixed Flames in a Constant Volume Combustion Bomb under Autoignition Conditions. Energy & Fuels 2010, 25, (1), 119-129. 25. Mueller, C. J.; Boehman, A. L.; Martin, G. C., An Experimental Investigation of the Origin of Increased NOx Emissions When Fueling a Heavy-Duty Compression-Ignition Engine with Soy Biodiesel. SAE Int. J. Fuels Lubr. 2009, 2, (1), 789-816. 26. Zheng, Z.; Badawy, T.; Henein, N.; Sattler, E., Investigation of Physical and Chemical Delay Periods of Different Fuels in the Ignition Quality Tester. Journal of engineering for gas turbines and power 2013, 135, (6), 061501-061501. 27. Ryan, T. W. Correlation of physical and chemical ignition delay to cetane number; 01487191; SAE Technical Paper: 1985. 28. Ryan, T. W.; Stapper, B. Diesel fuel ignition quality as determined in a constant volume combustion bomb; SAE Technical Paper: 1987. 29. Lapuerta, M.; Sanz-Argent, J.; Raine, R., Heat release determination in a constant volume combustion chamber from the instantaneous cylinder pressure. Applied Thermal Engineering 2014, 63, (2), 520-527. 30. Heywood, J. B., Internal combustion engine fundamentals. McGraw-Hill: 1988. 31. Mueller, C. J. The quantification of mixture stoichiometry when fuel molecules contain oxidizer elements or oxidizer molecules contain fuel elements; SAE Technical Paper: 2005. 32. Bruneaux, G., Liquid and vapor spray structure in high-pressure common rail diesel injection. Atomization and Sprays 2001, 11, (5). 33. Lee, C. S.; Park, S. W.; Kwon, S. I., An experimental study on the atomization and combustion characteristics of biodiesel-blended fuels. Energy & Fuels 2005, 19, (5), 2201-2208. 34. Canaan, R. E.; Dec, J. E.; Green, R. M.; Daly, D. T. The influence of fuel volatility on the liquid-phase fuel penetration in a heavy-duty DI diesel engine; SAE Technical Paper: 1998. 35. Kastengren, A.; Powell, C.; Tilocco, F.; Fezzaa, K. Initial evaluation of engine combustion network injectors with X-ray diagnostics; Advanced Photon Source (APS), Argonne National Laboratory (ANL), Argonne, IL (US): 2012. 36. Cheng, A. S.; Upatnieks, A.; Mueller, C. J., Investigation of Fuel Effects on Dilute, Mixing-Controlled Combustion in an Optical Direct-Injection Diesel Engine. Energy & Fuels 2007, 21, (4), 1989-2002. 37. Shibata, G.; Oyama, K.; Urushihara, T.; Nakano, T., Correlation of Low Temperature Heat Release With Fuel Composition and HCCI Engine Combustion. In SAE International: 2005. 38. Dec, J. E. A conceptual model of di diesel combustion based on laser-sheet imaging*; SAE technical paper: 1997. 39. Szybist, J. P.; Boehman, A. L.; Haworth, D. C.; Koga, H., Premixed ignition behavior of alternative diesel fuel-relevant compounds in a motored engine experiment. Combustion and Flame 2007, 149, (1), 112-128. 40. Westbrook, C. K., Biofuels Combustion*. Annual review of physical chemistry 2013, 64, 201-219. 41. Zhang, J.; Jing, W.; Roberts, W. L.; Fang, T., Effects of ambient oxygen concentration on biodiesel and diesel spray combustion under simulated engine conditions. Energy 2013, 57, (0), 722-732.

ACS Paragon Plus Environment

33

Energy & Fuels

Page 34 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

34

Page 35 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

APPENDIX A: Detailed Apparent Heat Release Rate Calculation To calculate AHRR, the air temperature throughout the chamber (Tbulk) was assumed to be uniform at each time step of ignition; calculated using Equation 1. Volume (V) was specified as the internal CVCC volume to be 0.473 liters. Instantaneous chamber pressure (P) was taken from the smoothed pressure data and the specific gas constant (Rspecific) of air was assumed to be 0.287 kJ/kg-K. Air and fuel mass (mair+fuel) was calculated using the ideal gas law equation and unit conversion equations, respectively. The fuel injection volume was held constant at 125 mm3 for each fuel. 𝑇bulk =

𝑃𝑉 𝑅specific 𝑚air+fuel

(1)

In order to account for changing heat capacities of the gas mixture with temperature, the ratio of specific heat capacities (gamma, γ) was included in the AHRR calculation; a value that is commonly between 1.3 and 1.4 in diesel combustion systems. Often gamma is left constant in AHRR calculations although this is rarely true under real environments; specified as γ = cp/cv, where cp and cv are constant pressure and constant volume heat capacities, respectively. For this study, a correlation of gamma to bulk temperature (Equation 2) was used to correct for changes in heat capacity ratio as the air mixture transitioned from an unburned to burned fraction. 1

𝛾= 1−

(2)

1 𝑎1 + 𝑎2 𝑇 + 𝑎3 𝑇 2 + 𝑎4 𝑇 3 + 𝑎5 𝑇 4

where, a1 = 3.6359 a2 = -1.33736 × 10-3 a3 = 3.29421 × 10-6

ACS Paragon Plus Environment

35

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 36

a4 = -1.91142 × 10-9 a5 = 0.275462 × 10-12 The first derivative of smoothed pressure data was calculated using a five-point method as shown in Equation 3. 𝑑𝑃 −𝑃(𝑡 + 2ℎ) + 8𝑃(𝑡 + ℎ) − 8𝑃(𝑡 − ℎ) + 𝑃(𝑡 − 2ℎ) ≈ 𝑑𝑡 12ℎ

(3)

where, h = time step Finally, AHRR (ΔQ/dt) was calculated (Equation 4) as a function of gamma (γ), volume (V), and change in pressure (dP/dt). 𝛥𝑄 1 𝑑𝑃 = 𝑉 𝑑𝑡 𝛾 − 1 𝑑𝑡

(4)

ACS Paragon Plus Environment

36