Akhilendra Pratap SinghEngine Research Laboratory,

Department of Mechanical Engineering,

Indian Institute of Technology Kanpur,

Kanpur 208016, India

e-mail: akhips@iitk.ac.in

Nikhil BajpaiEngine Research Laboratory,

Department of Mechanical Engineering,

Indian Institute of Technology Kanpur,

Kanpur 208016, India

e-mail: sujeet20186@gmail.com

Avinash Kumar Agarwal1Engine Research Laboratory,

Department of Mechanical Engineering,

Indian Institute of Technology Kanpur,

Kanpur 208016, India

e-mail: akag@iitk.ac.in

Combustion Mode SwitchingCharacteristics of a Medium-Duty Engine Operated inCompression Ignition/PCCICombustion ModesPremixed charge compression ignition (PCCI) combustion is a novel combustion con-cept, which reduces oxides of nitrogen (NOx) and particulate matter (PM) emissionssimultaneously. However, PCCI combustion cannot be implemented in commercialengines due to its handicap in operating at high engine loads. This study is focused onthe development of hybrid combustion engine in which engine can be operated in bothcombustion modes, namely, PCCI and compression ignition (CI). Up to medium loads,engine was operated in PCCI combustion and at higher loads, the engine control unit(ECU) automatically switched the engine operation to CI combustion mode. These com-bustion modes can be automatically switched by varying the fuel injection parametersand exhaust gas recirculation (EGR) by an open ECU. The experiments were carried outat constant engine speed (1500 rpm) and the load was varied from idling to full load(5.5 bar brake mean effective pressure (BMEP)). To investigate the emission and particu-late characteristics during different combustion modes and mode switching, continuoussampling of the exhaust gas was done for a 300 s cycle, which was specifically designedfor this study. Results showed that PCCI combustion resulted in significantly lower NOx

and PM emissions compared to the CI combustion. Lower exhaust gas temperature(EGT) in the PCCI combustion mode resulted in slightly inferior engine performance.Slightly higher concentration of unregulated emission species such as sulfur dioxide(SO2) and formaldehyde (HCHO) in PCCI combustion mode was another importantobservation from this study. Lower concentration of aromatic compounds in PCCI com-bustion compared to CI combustion reflected relatively lower toxicity of the exhaust gas.Particulate number-size distribution showed that most particulates emitted in PCCI com-bustion mode were in the accumulation mode particle (AMP) size range, however, CIcombustion emitted relatively smaller sized particles, which were more harmful to thehuman health. Overall, this study indicated that mode switching has significant potentialfor application of PCCI combustion mode in production grade engines for automotivesector, which would result in relatively cleaner engine exhaust compared to CI combus-tion mode engines. [DOI: 10.1115/1.4039741]

Keywords: partially premixed charge compression ignition, mode switching, unregulatedemissions, particulate emissions, open ECU

Introduction

Due to concerns of rapid depletion of fossil fuel reserves, envi-ronmental degradation, and global warming, internal combustion(IC) engines with higher efficiency and lower emissions are beingexpeditiously developed. Diesel engines are becoming increas-ingly popular for automotive applications due to their inherentlysuperior fuel economy, reliability, durability, lower carbon diox-ide (CO2) emission, and higher specific power output compared toany other power plant in that size range. However, the presence oflocalized fuel-rich regions and high-temperature regions in theconventional compression ignition (CI) combustion leads to highlevels of emissions of oxides of nitrogen (NOx) and particulatematter (PM). Massive efforts have been undertaken in last fourdecades to reduce these pollutants using in-cylinder control meas-ures as well as exhaust gas after-treatment systems; however,

these techniques cannot be implemented in all production gradeengines due to several operational and system complexities.

In last two decades, the low temperature combustion (LTC) hasgained significant attention of the IC engine researchers due to itspotential to reduce the PM and NOx emissions simultaneously.The absence of localized fuel rich zones results in homogeneouscombustion, which leads to lower peak in-cylinder temperature,resulting in ultralow emissions of PM and NOx with satisfactoryengine performance. Among various LTC derivatives, namely,premixed charge compression ignition (PCCI) combustion, homo-geneous charge compression ignition combustion, and partiallypremixed charge compression ignition combustion, PCCI com-bustion has been extensively explored because of its superiorcombustion control potential and practicality for production gradeengines [1]. In PCCI combustion, a fraction of fuel is injectedvery early in the compression stroke (as pilot injection), whichresults in homogeneous fuel–air mixture formation, however, theremaining fuel is injected in the same way, as is done in conven-tional CI combustion. Weiskirch and Mueller [2] reported thatPCCI combustion characteristics lie in between conventional CIcombustion and homogeneous charge compression ignition

1Corresponding author.Contributed by the Internal Combustion Engine Division of ASME for

publication in the JOURNAL OF ENERGY RESOURCES TECHNOLOGY. Manuscript receivedOctober 23, 2017; final manuscript received March 15, 2018; published online April19, 2018. Editor: Hameed Metghalchi.

Journal of Energy Resources Technology SEPTEMBER 2018, Vol. 140 / 092201-1Copyright VC 2018 by ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

combustion. Despite extremely low exhaust emissions, PCCIcombustion could not be applied to production grade engines dueto lack of direct combustion control, which leads to crude orknocking combustion at high loads. Efforts have been made toexpand the operating range of PCCI combustion by employingturbochargers, superchargers, and multiple fuel injection strategy;however, it is still difficult to achieve PCCI combustion for theentire load range of a commercial diesel engine [3]. Therefore,researchers explored the “mode switching” or “mixed-modecombustion” strategy for practical implementation of PCCI com-bustion in commercial engines. In mode switching strategy,engine operates in PCCI combustion mode up to medium engineloads and it switches back to CI combustion mode at higher loadsin order to deliver full power (rated load) under realistic drivingconditions. Many researchers implemented mode switching strat-egy in gasoline engines by controlling the hot residual gas fractionby employing the variable valve actuation (VVA) mechanism[4–6]. Narayanaswamy and Rutland [7] suggested that modeswitching can be achieved by controlled variations in engine oper-ating parameters such as fuel injection timings, injection quantity,fuel injection pressure (FIP), and exhaust gas recirculation (EGR)rate. They suggested the indicated mean effective pressure(IMEP) fluctuations to be the main concern during mode switch-ing, which was required to be controlled precisely for stableengine operation. The presence of relatively lower in-cylinderpressure during advanced fuel injection timing causes wallimpingement of the fuel spray. This leads to participation of lesserfuel quantity in combustion, which results in IMEP variations.This could be avoided by varying the fuel quantity per enginecycle in addition to adopting complicated fuel injection strategy.Relatively higher sensitivity of in-cylinder conditions (tempera-ture and fuel–air mixture quality) was another factor behind IMEPfluctuations during PCCI-to-CI combustion mode switching,which resulted in abnormal combustion at high engine loads. Fanget al. [8] developed a hybrid configuration mode switching strat-egy in which an integrated starter generator was used to eliminatethe IMEP fluctuations by providing compensating torque. Wang[9] designed a robust hybrid nonlinear control system to achievesmoother and faster transient response during mode switching byproper coordination of variable nozzle turbocharger and EGRrate. However, these systems cannot be used in production gradeengines due to system complexities.

Some researchers adopted “fuel compensation strategy” toeliminate the IMEP fluctuations during mode switching. In thisstrategy, appropriate fuel–air mixture was supplied to combustionchamber using optimized fuel-injection parameters. Burton et al.[10] and Busch et al. [11] observed spikes in power output as wellas emissions due to difference in response time between the fueland air handling systems. They reported HC emissions spike andpoor combustion during PCCI-to-CI combustion mode switching.This was mainly due to charge dilution because of high residualgas quantity and relatively lower fuel–air ratio, which led to mis-fire. Similarly, CI-to-PCCI combustion mode switching resultedin a NOx emissions peak and extremely high rate of pressure rise(RoPR) in the combustion chamber due to high fuel–air ratios.Black et al. [12] found high dependence of NOx emissions onengine load during a transient engine cycle, while Hagena et al.[13] reported that the rate of change of transient engine load sig-nificantly affected the spike in NOx emissions. Fuel injectionstrategies during mode switching were also investigated by sev-eral researchers. Tanabe et al. [14] used optimized fuel injectionstrategy and air management to reduce NOx emissions and smokeduring mode switching between PCCI-to-CI combustion. Bane-rjee and Rutland [15] computed the effect of multiple injections ina mode switching cycle and reported that multiple injections sig-nificantly reduced the excessive RoPR due to charge stratification.Han et al. [16] found rapid changes in power output, IMEP, maxi-mum RoPR, combustion noise, and HC emissions during modeswitching from LTC-to-CI combustion. In another study con-ducted by Han et al. [17], they reported that smooth mode

switching could be achieved by employing short-route EGR andfuel injection control. Kim et al. [18] and Rohani et al. [19] inves-tigated the fuel injection strategies for achieving smooth modeswitching between PCCI and CI combustion and verified theobservations of Han et al. [17].

These studies demonstrated that transient fuel injection strategyhas significant potential to reduce IMEP fluctuations during com-bustion mode switching. Bae et al. [20] studied the fuel injectionstrategies during combustion mode switching cycles and reportedthat IMEP of some cycles were significantly higher than steady-state PCCI and CI combustion cycles. Ouyang et al. [21] investi-gated the effects of different control parameters such as injectiontiming, FIP, and EGR rate and developed an optimized modeswitching strategy based on predefined step-by-step fuel–air coor-dination. Shi et al. [22] reported that incomplete combustion offuel during transitional cycles caused a reduction in IMEP. Theysuggested that increased fuel injection quantity during early tran-sitional cycles could compensate the IMEP reduction. Deng et al.[23] investigated fuel compensation strategy to improve smooth-ness of IMEP during mode switching, however, they did notreport emission characteristics during mode switching.

The previous studies showed that fuel injection parameters andEGR rate effectively controlled the fuel–air mixture formation andits chemical kinetics in the combustion chamber. Therefore in thisstudy, optimized fuel injection strategy for mode switching betweenconventional CI and PCCI combustion modes has been developed.Experiments were carried out in a suitably instrumented productiongrade engine using an open engine control unit (ECU). Measure-ment of particle number-size distribution and unregulated gaseousemission species are two important and unique aspects of this study,which are not investigated thoroughly in the previous studies.

Experimental Setup

In this study, PCCI combustion was achieved in a productiongrade engine using an open ECU. Experiments were carried out toinvestigate the combustion mode switching characteristics betweenCI and PCCI combustion modes. The experimental setup consistsof three main systems, namely, test engine, emission measurementsystem, and open ECU. In this study, a two-cylinder, four-stroke,medium-duty transportation direct injection compression ignitionengine (Mahindra and Mahindra; 0.9 L 2CY CRDe NA) was usedfor investigating the mode switching characteristics between CI andPCCI combustion modes. The engine was coupled to an eddy currentdynamometer (Dynomerk Controls; EC100). The engine had a com-mon rail direct injection system capable of generating FIP up to1200 bar. Fuel injection parameters were controlled by an ECU(Bosch; EDC17C55, India), supplied by the original equipment man-ufacturer (OEM). For PCCI combustion, a fraction of exhaust gaswas supplied to the intake manifold through a proportional–integral–derivative controlled EGR valve. A laminar flow element(Meriam, 50MC2-2F) was installed to measure the volumetric airflow rate. Fuel consumption was measured using a graduated buretteby determining the time required for consumption of a given volumeof fuel. Schematic of the experimental setup is shown in Fig. 1.Technical specifications of the test engine are given in Table 1.

OEM ECU supplied along with the production grade engineperforms actions based on predetermined logic (map) stored in itsmicroprocessor. This system does not offer any flexibility for con-trolling various parameters such as FIP, start of pilot injection,start of main injection (SoMI), fuel injection quantity, and EGRrate for achieving combustion mode switching. Therefore, OEMECU was replaced with an open ECU (Nira Controls; NIRAi7r),which was duly connected to various sensors and actuators in theengine. Open ECU is a sophisticated electronic engine manage-ment system, which is custom-configured for this study. The sche-matic of the open ECU configuration is shown in Fig. 2. Thetechnical specifications of open ECU are given in Table 2.

Open ECU controlled the amount of fuel to be injected intoeach cylinder with the help of a “demand torque map.” Injected

092201-2 / Vol. 140, SEPTEMBER 2018 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

fuel quantity was varied depending on engine speed and load.Open ECU controlled the fuel quantity for pilot injection with thehelp of “injection mass pilot map.” FIP was controlled by a “fuelpressure controller set point map,” which was dependent on twovariables, namely, engine speed and fuel mass request. This cre-ated the desired set point for the proportional–integral–derivativecontroller. Fuel injection timings were controlled by two fuelinjection timing maps, which took inputs from cycle reference sig-nal and generated pulse output in terms of crank angle degreebefore the top dead center. EGR rate was controlled with the helpof EGR actuator control map, which controlled the opening andclosing of the EGR valve through pulse width modulated signals.

To measure the regulated gaseous emissions, a raw exhaust gasemission analyzer (Horiba, EXSA-1500) was used. This analyzerprovided the concentrations of regulated gaseous species, namely,HC, CO, NOx, and CO2. Measurement principle and range of rawexhaust gas emission analyzer for each gaseous species is given inTable 3.

Fourier transform infrared (FTIR) emission analyzer (Horiba;MEXA-6000FT-E) was used to measure concentrations ofunregulated gaseous species using the FTIR method combined

with a multivariate analysis algorithm [24]. For number-size dis-tribution of particulates, an engine exhaust particle sizer (EEPS)(TSI, EEPS-3090) was used, which could measure particle sizesranging from 5.6 to 560 nm with maximum particle number den-sities up to 108 particles/cm3 of the exhaust gas at 10 Hz fre-quency. EEPS spectrometer provided both high temporalresolution and reasonable size resolution using multiple sizedetectors working in parallel. EEPS spectrometer is therefore anideal instrument for measuring particulates emitted in the engineexhaust under transient engine operating conditions. Other impor-tant details of EEPS spectrometer can be seen in our previouspublication [25].

Experimental Methodology

Mode switching strategy was based on variations of fuel injec-tion parameters and EGR rate using the open ECU. CI and PCCIcombustion modes were optimized, and optimum fuel injectionparameters and EGR rates were determined for the entire set ofengine loads at fixed speed of 1500 rpm. On these engine operat-ing points, performance and emissions including regulated,unregulated, and particulates were determined. Results of thesebaseline experiments were used to calibrate the open ECU maps,which in-turn controlled the fuel injection parameters and EGRrate, when such demands were placed on the ECU.

Mode switching experiments were performed at a constantengine speed of 1500 rpm as well. In mode switching experiments,the maximum fuel injection quantity (at 100% throttle) was fixed,which resulted in �5.5 bar brake mean effective pressure (BMEP)in CI combustion mode. A customized test cycle of 300 s was exe-cuted for the mode switching experiments, which included engineidling, CI combustion, PCCI combustion, and mode switchingbetween CI and PCCI combustion modes. In mode switchingexperiment, low load limit was defined by high IMEP variationand high load limit was defined by knocking combustion [26].

Fig. 1 Schematic of the mode switching experimental setup: 1—test engine, 2—eddy currentdynamometer, 3—dynamometer controller, 4—laminar flow element, 5—open ECU, 6—ECUinterface system, 7—fuel tank, 8—fuel pump, 9—fuel rail, 10—EGR valve, 11—EGR loop, 12—thermodiluter, 13—engine exhaust particle sizer, 14—EEPS data logger, 15—FTIR emissionanalyzer, and 16—raw exhaust gas emission analyzer

Table 1 Technical specifications of the test engine

Manufacturer/Model Mahindra & Mahindra Ltd. India/0.9 L 2CY CRDe

Engine type Four stroke, naturally aspiratedNumber of cylinders TwoCompression ratio 18.5Injection system Direct injectionBore/stroke 83/84 mmSwept volume 909 ccPower 20 kW at 3600 rpmCooling system Water cooledValve train type Double overhead camshaft (DOHC)

Journal of Energy Resources Technology SEPTEMBER 2018, Vol. 140 / 092201-3

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

Strategy of mode switching experiments to achieve both combus-tion modes is shown in Fig. 3.

In a test-cycle of the mode switching experiment, engine wasstarted in CI combustion mode and the minimum fuel quantitywas supplied to each cylinder. Engine was started in CI combus-tion mode to avoid misfire at low loads, however this led to higherHC emissions. After engine warm-up, mode transition from CI-to-PCCI combustion mode was done by altering the fuel injectionparameters. During PCCI combustion, start of pilot injection tim-ing was varied from 30 deg to 35 deg CA bTDC and pilot injectionquantity was varied from 5% to 12% of the main injection quan-tity, depending on the engine load. In PCCI combustion region,

FIP was varied from 550 to 750 bar. FIP was decided based onSoMI timing, which was varied from 15 deg to 20 deg CA bTDC.At the starting of PCCI combustion, EGR was 0%, and after com-plete mode switching from CI-to-PCCI combustion, EGR wasmaintained at 15% during the entire duration of PCCI combustion.In PCCI combustion mode, engine was loaded up to 55% throttle(BMEP �2.95 bar). At 55% throttle position, engine showed sig-nificantly higher combustion noise due to excessive RoPR. Athigher engine loads, relatively faster chemical kinetics of fuel-airmixture led to severe knocking. Earlier SoMI timing used forPCCI combustion resulted in slightly inferior fuel–air mixing dueto lower in-cylinder pressure and temperature, leading to higherHC and carbon monoxide (CO) emissions. Beyond 55% throttle,engine was switched back to CI combustion mode by altering fuelinjection parameters and deactivating EGR valve gradually. Pilotinjection was also deactivated and SoMI was retarded to 6 deg CAbTDC. At higher engine loads, FIP was decreased to 400–450 barin order to avoid fuel spray impingement on the cylinder walls.Using optimized fuel injection parameters, engine achieved up to�5.5 bar BMEP in CI combustion mode and catered to the ratedload.

Results and Discussion

In this study, experimental results were divided into four cate-gories, namely (i) performance characteristics, (ii) regulated emis-sion characteristics, (iii) unregulated emission characteristics, and(iv) particulate characteristics. Engine performance analysisincluded exhaust gas temperature (EGT) and brake thermal effi-ciency (BTE) analyses. Regulated emission analysis included CO,NOx, and HC emissions. Unregulated emission analysis includedtrace species such as benzene (C6H6), Toluene (C7H8), sulfurdioxide (SO2), formaldehyde (HCHO), formic acid (HCOOH),and iso-cyanic acid (HNCO). Particulate characteristics includedparticle number-size distribution, total particle number (TPN)concentration, nucleation mode particle (NMP) number concen-tration, and accumulation mode particle (AMP) number concen-tration. All experiments were repeated thrice and an average of

Table 2 Specifications of the open ECU

Make/model Nira controls/NIRA i7rsOperating voltage 24 VNumber of connectors 02 (A and K)Number of pins 196 (105þ 91)Type Open (configurable via PC)Analog/Digital inputs 30/15Pulse width modulated outputs 18Relay outputs 4 (low side)H-bridge 4Injector driver Solenoid (8)Communication CAN Bus

Table 3 Measurement principle and range of the exhaust gasemission analyzer

Species Measurement principle Measurement range

CO Nondispersive infra-red 0–5000 ppmCO2 Nondispersive infra-red 0–10/20% (v/v)THC Flame ionization detector 0–100/1000/5000/10,000/50,000 ppmNOx Chemiluminescence

detector0–100/500/1000/5000 ppm

Fig. 2 Open ECU system configuration

092201-4 / Vol. 140, SEPTEMBER 2018 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

these results was plotted along with error bars, representing possi-ble experimental errors.

Figure 4 shows the variations in the engine performance charac-teristics, namely, EGT, BTE, and BMEP with respect to time. Allperformance parameters were measured once in every 20 s. In thisfigure, different background colors represented the applicablecombustion mode. Figure 4 shows that BMEP was very low (�0.7bar) at the time of engine starting. During the mode switchingfrom CI-to-PCCI combustion, BMEP was changed gradually inorder to avoid sudden changes in fuel injection parameters, whichcould possibly result in rough engine operation. At the end of thetransition period, fuel injection quantity was increased graduallyand EGR was employed to operate the engine in PCCI combustionmode up to �2.9 bar BMEP. At the end of PCCI combustionregion (at t¼ 150 s), termination of pilot injection resulted inslightly inferior fuel–air mixing, leading to reduction in BMEP.To avoid further reduction in BMEP during PCCI-to-CI modeswitching, fuel injection quantity was slightly increased at thestart of mode switching from PCCI-to-CI combustion. This wassimilar to the “fuel compensation strategy” adopted by Deng et al.[23]. After achieving CI combustion (t¼ 200–300 s), BMEP

increased gradually due to increase in fuel quantity up to the ratedengine load.

At lower engine loads, slightly lower BTE (�17%) wasobserved, which increased gradually with increasing engine load.At lower engine loads, engine misfiring due to low in-cylindertemperature was the main reason for this behavior. During PCCIcombustion, application of pilot fuel injection promoted fuel-airmixing, therefore BTE increased rapidly up to �33%. However inthe remaining region of PCCI combustion, BTE remained con-stant (flat curve between t¼ 100 and 150 s). This was due to twocounter effects, namely (i) increasing in-cylinder temperature athigher engine loads, which increased the BTE, and (ii) increasedHC and CO emissions reduced the BTE [27]. Further increase inBTE was observed during mode switching from PCCI-to-CI com-bustion. This was mainly due to deactivation of EGR, whichresulted in slightly higher peak in-cylinder temperature, leading tohigher BTE. In CI combustion mode, increasing engine load ledto higher BTE. Maximum BTE in CI combustion mode was�36%. EGT variation with respect to engine load showed signifi-cantly lower EGT in PCCI combustion mode compared to CIcombustion mode. During mode switching from PCCI-to-CI

Fig. 4 Variations in BTE and EGT with BMEP during different combustion modes and modeswitching

Fig. 3 Variation of fuel injection parameters and EGR rate during mode switching test-cycle

Journal of Energy Resources Technology SEPTEMBER 2018, Vol. 140 / 092201-5

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

mode, a sudden increase of �100 �C in EGT was observed whichincreased continuously in CI combustion mode with increasingengine load. The presence of diffusion combustion phase in CIcombustion mode was the main reason for higher EGT. Termina-tion of EGR during PCCI-to-CI combustion mode switching wasanother important reason for such a sudden increase in the EGT.

Figure 5 shows the variations in CO, HC, and BMEP withrespect to time. Higher CO and HC emissions from PCCI combus-tion reflected high degree of incomplete combustion, which wasmainly due to lower bulk in-cylinder temperatures [28]. Regulatedgaseous species (CO and HC) are presented in brake specific val-ues in Fig. 5.

During idling, HC emissions were very high due to milder con-ditions prevailing in the engine combustion chamber. HC emis-sions decreased with increasing engine load since the chambertemperature increased. HC emissions increased with increasingengine load during PCCI combustion and reached up to 2 g/kWh.Relatively lower in-cylinder temperature during PCCI combustionwas the prime reason for this, which was responsible for incom-plete combustion in the absence of diffusion combustion phase.During mode switching from PCCI-to-CI combustion, substantialreduction in HC emissions were observed due to higher bulk in-cylinder temperatures, which promoted fuel oxidation. During CI

combustion mode, HC emissions further decreased with increas-ing engine load. This was another reason for higher EGT duringCI combustion mode. CO emission showed significantly differenttrend during mode switching. The literature shows that CO emis-sion was higher in LTC mode [29,30]. However, mode switchingfrom PCCI-to-CI combustion resulted in higher CO emissionfrom CI combustion compared to PCCI combustion mode. Thepresence of relatively richer fuel–air mixture at higher engine loadmay be a possible reason, where lack of oxygen prevented oxida-tion of CO in to CO2. Peak CO emission concentration in PCCIcombustion mode was �18 g/kWh, however, for CI combustionmode, peak CO concentration went up to� 60 g/kWh.

Figure 6 shows the variations in TPN, NOx, and BMEP withrespect to time. Particulate characteristics were measured usingEEPS. EEPS was continuously operated for 300 s, which helpedin measurement of particulate data during mode switching fromconventional CI-to-PCCI combustion.

Figure 6 shows significantly lower TPN and NOx emissionsduring PCCI combustion mode compared to CI combustion mode.TPN in CI combustion mode was an order of magnitude highercompared to PCCI combustion mode. In PCCI combustion region,the engine yielded �4� 108 particles/cm3 of the exhaust, how-ever, in CI combustion region, the yield was �1.75� 109

Fig. 5 Variations in CO and HC emissions with BMEP during different combustion modes andmode switching

Fig. 6 Variations in TPN and NOx emissions with BMEP during different combustion modesand mode switching

092201-6 / Vol. 140, SEPTEMBER 2018 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

particles/cm3 of the exhaust. In both combustion modes, increas-ing engine load resulted in slightly higher TPN due to higherinjected fuel quantity. However, these increments were not signifi-cant compared to increase in TPN during mode switching fromPCCI-to-CI combustion mode. This was an important observationof this study. Retarded SoMI timing and deactivation of pilotinjection were the two important factors for higher particulateemissions from CI combustion compared to PCCI combustion.Retarded SoMI timings resulted in lesser time available forfuel–air mixing and the absence of pilot injection led to relativelyinferior in-cylinder conditions for main fuel injection. Combinedeffect of these two factors resulted in localized fuel-rich zones,which promoted soot nuclei formation and resulted in higher par-ticulate emissions during CI combustion mode compared to PCCIcombustion mode.

Significantly lower NOx emissions in PCCI combustion com-pared to CI combustion region was another advantage of PCCIcombustion mode over the CI combustion mode. In PCCI com-bustion mode, the maximum NOx concentration was �150 ppm,however, it increased up to �575 ppm in CI combustion mode.During PCCI combustion, NOx emissions were almost constant;however, at the end of the PCCI combustion region, NOx emis-sions increased rapidly (�300 ppm). This was an important crite-rion for determining the upper load limit of PCCI combustion.The maximum NOx increase was observed during PCCI-to-CIcombustion mode switching, where retarded SoMI timing, and theabsence of EGR and pilot injection led to heterogeneous fuel–airmixing, leading to higher in-cylinder temperatures [31]. At theend of mode switching from PCCI-to-CI combustion, NOx emis-sions reduced slightly due to thermal stabilization after suddenincrease in peak cylinder temperature, however, NOx emissionsremained almost constant during the CI combustion mode.Although temperature increased during the CI combustion, how-ever the presence of relatively richer fuel–air mixture led to oxy-gen deficiency, which might be a possible reason for this trend.This fact also supports the CO emission trend (Fig. 5).

Figure 7 shows the variations of particle number-size distributionwith respect to time. Particle number-size distribution was also splitinto five regions in the figure: CI (idle); CI-to-PCCI region; PCCIregion; PCCI-to-CI region; and CI region. All these regions weredivided based on BMEP (shown on left vertical surface).

Figure 7 clearly indicates the variations in particle numbersemitted from mineral diesel fueled engine during different

combustion modes and mode switching. Smooth variations in par-ticle number-size distribution curve also reflect the combustionstability during different combustion modes and mode switching.At the time of engine starting, particle number concentration wasvery low, which remained almost constant during mode switchingfrom CI-to-PCCI combustion mode. During PCCI combustionmode, increasing load resulted in higher particle number concen-tration. During PCCI combustion mode, peak of particle number-size distribution shifted toward the accumulation mode(50 nm<Dp< 1000 nm) particles, which indicated that the engineemitted relatively larger particles during PCCI combustion mode.The absence of nanoparticles (Dp< 10 nm) during PCCI combus-tion was another important observation of this study. Concentra-tion of nanoparticles became significant during mode switchingfrom PCCI-to-CI combustion mode. This reflected the presence offuel-rich zones, which promoted the nuclei formation. Duringmode switching from PCCI-to-CI combustion mode, peak of par-ticle number-size distribution shifted toward smaller particles.This was mainly due to higher in-cylinder temperature (due to dif-fusion combustion), which prevented the condensation of gaseousspecies onto particulate surfaces and led to relatively smaller par-ticulate. During CI combustion, increasing engine load shifted theparticle number-size distribution curve upward but toward theright side. This showed that particle number concentrationincreased with increasing engine load and relatively larger par-ticles were emitted due to higher fuel quantity injected. Nanopar-ticle concentration was significantly higher in CI combustionmode compared to PCCI combustion mode, which reflected sig-nificantly higher health risk potential of CI combustion mode.Comparison of particle number-size distribution curves of PCCIcombustion region and CI combustion region clearly showedwider size distribution of particles emitted in CI combustionmode. Highest particle number concentration in PCCI combustionmode was �2� 107 particles/cm3 of the exhaust, which increasedup to 1.6� 108 particles/cm3 of the exhaust in CI combustionmode, i.e., it increased by an order of magnitude.

Figure 8 shows the variations in concentrations of NMP(Dp< 50 nm), AMP, and BMEP with respect to time. Figure 8shows that NMPs were almost constant throughout the test-cycle(in both combustion modes) except during the mode switchingperiod from PCCI-to-CI combustion mode. This was mainly dueto reduction in FIP, which led to slightly inferior fuel–air mixingand resulted in the formation of higher number of NMPs [32]. The

Fig. 7 Variations in particle number-size distributions with respect to time and BMEP fordifferent combustion modes and mode switching

Journal of Energy Resources Technology SEPTEMBER 2018, Vol. 140 / 092201-7

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

effect of inferior fuel-atomization was also observed in AMP con-centration, which increased rapidly during PCCI-to-CI combus-tion mode switching. Few researchers also reported this behaviorand suggested that solid carbonaceous particles also exist in CIcombustion mode due to fuel-air mixture inhomogeneity [33].After mode switching from PCCI-to-CI combustion, AMP con-centration further increased due to combined effect of the two fac-tors, namely, heterogeneous fuel–air mixing and higher fuelquantity. These trends showed that the differences in TPN concen-trations emitted during CI combustion mode was mainly due todominance of AMPs in the exhaust gas.

Figures 9–11 show the variations in different unregulated gase-ous species emitted from different combustion modes and modeswitching. These unregulated gaseous species were measuredusing FTIR emission analyzer, which was capable of continuousmeasurement of 31 unregulated hydrocarbon species quite pre-cisely. During mode switching, six gaseous emission species wereobserved to be present in significant quantities, which were classi-fied as harmful to the human health.

Figure 9 shows the variations in benzene (C6H6), toluene(C7H8), and BMEP with respect to time. These species were a partof emission group of benzene, toluene, ethyl-benzene, and xylene

(BTEX), however, ethyl-benzene and xylene were not detected.The presence of unsaturated hydrocarbons in mineral diesel acts asprecursor, which is responsible for the formation of aromatic spe-cies such as benzene and toluene. Benzene is a known carcinogen;therefore its emission and human exposure is undesirable [34–36].

Figure 9 shows that benzene concentration was almost constantin PCCI combustion mode and it increased drastically in CI com-bustion mode. In IC engines, benzene forms mainly due to incom-plete combustion of higher carbon chain length hydrocarbonspresent in fuel and lubricants. In the presence of high in-cylindertemperature and pressure, fuel oxidizes completely; however,pyrolysis of fraction of lubricating oil results in higher concentra-tion of benzene in the CI combustion region. These results aresimilar to the results reported by Berm�udez et al. [37], who sug-gested that aromatic species such as benzene form due to fuelpyrolysis at higher temperature and pressure conditions prevailingin the engine combustion chamber. Insufficient oxygen availabil-ity was another factor, which partially oxidized the fuel carboninto several unburnt hydrocarbon products, including benzene. InPCCI combustion mode, maximum benzene concentrationreached up to �3 ppm, and in CI combustion mode, it reached upto �8 ppm. Results of FTIR analysis show that minor

Fig. 8 Variations in concentrations of nucleation mode and AMPs with BMEP during differentcombustion modes and mode switching

Fig. 9 Variations in benzene and toluene with BMEP during different combustion modes andmode switching

092201-8 / Vol. 140, SEPTEMBER 2018 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

concentration (�2–4 ppm) of toluene was also present in theengine exhaust. Toluene has a cyclic structure (similar to benzene)with one methyl group attached to benzene ring and its propertiesare nearly same as that of benzene. However, the toxicity of tolu-ene is relatively lower than benzene. Toluene is also a product ofincomplete combustion (pyrolysis of heavier fractions of mineraldiesel and lubricating oil). The present study showed slightly ran-dom behavior of toluene emission in different combustion modesand mode switching. During PCCI combustion and PCCI-to-CImode switching, toluene concentration was slightly higher, whichdecreased in CI combustion mode. This was due to relativelylower in-cylinder temperature in the PCCI combustion mode,which promoted incomplete combustion of fuel and lubricatingoil. The toluene emission levels were within permissible limitbecause current standard for a permissible exposure limit for tolu-ene vapors stands at 100 ppm [38].

Figure 10 shows the variations in HCHO, HCOOH, and BMEPwith respect to time. These species are categorized as carbonylspecies. The term carbonyl refers to the carbonyl functional group,

which is a divalent group consisting of a carbon atom, double-bonded to an oxygen atom.

Figure 10 shows that variations in HCOOH were not signifi-cant. In entire operating region, HCOOH was almost constant(�0.5 ppm), therefore, HCOOH emission was not discussed anyfurther. However, HCHO emissions were significant in both com-bustion modes. HCHO was a result of partial combustion andpyrolysis of fuel and lubricating oil in the engine combustionchamber. HCHO emission was higher in PCCI combustion modeand it reduced with increasing engine load in the CI combustionmode. In PCCI combustion mode, in-cylinder temperature wasrelatively lower, which led to relatively lower evaporation andhigher degree of incomplete combustion of fuel droplets. CI com-bustion showed slightly lower HCHO emission due to higher in-cylinder temperatures; however lack of oxygen during relativelyricher fuel–air mixture conditions restricted HCHO oxidation.HCHO was produced in CI combustion mode and was destroyedin the later stages of combustion. In PCCI combustion mode, rela-tively shorter combustion duration was another important factor

Fig. 10 Variations in HCHO and HCOOH with BMEP during different combustion modes andmode switching

Fig. 11 Variations in SO2 and HNCO with BMEP during different combustion modes andmode switching

Journal of Energy Resources Technology SEPTEMBER 2018, Vol. 140 / 092201-9

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

for higher HCHO emissions, where lesser time availabilityreduced the oxidation of HCHO [39]. Highest HCHO emissionlevels reached up to 20 ppm during mode switching from PCCI-to-CI combustion mode, which reduced to �16 ppm during CIcombustion mode.

Figure 11 shows the variations in SO2, HNCO, and BMEP withrespect to time. Figure 11 shows that SO2 emissions were lessthan 6 ppm in both combustion modes. The presence of sulfur inmineral diesel led to formation of SO2 during combustion. In CIcombustion mode, SO2 concentration was relatively lower thanthe PCCI combustion mode. This was mainly due to formation ofsulfur trioxide (SO3), which resulted in formation of sulfuric acid(H2SO4) in the presence of moisture. The presence of sulfuric acidvapors in the exhaust acts as a site for condensation of volatilematerials, which promotes particulate formation with higheradsorbed organic content. Therefore, conversion of SO2 into par-ticulates led to higher particulate concentration in CI combustionmode and subsequently lower SO2 in the gas phase in the engineexhaust. HNCO emission was another important aspect of thisstudy. The literature states that HNCO forms in higher quantitieswhen NO, CO, and H2/NH3 react over precious metal catalysts.Recent studies have shown that IC engines are also one of themain sources of HNCO emissions in the atmosphere [40,41].HNCO has been known to be a highly toxic gaseous substanceand is a potential health risk due to its dissociation at physiologi-cal pH. Roberts et al. [42] reported that inhalation of even verylow concentrations of HNCO (� 1 ppbv) may be sufficient tocommence carbamylation reactions in the human body. In thepresent study, it was observed that HNCO emission increasedwith increasing engine load during PCCI combustion mode, how-ever, CI combustion mode showed slightly lower HNCO emis-sion. Brady et al. [41] reported that lower in-cylinder temperature(during cold-start) might be a possible reason of HNCO formationin the engines. Maximum HNCO concentration reached up to�23 ppm during PCCI combustion mode and reduced to �18 ppmduring the CI combustion mode.

Conclusions

In this study, experiments were performed to achieve combus-tion mode switching between conventional CI and PCCI combus-tion modes in a production grade engine. For mode switching,open ECU maps were generated for controlling different engineoperating parameters, depending on the engine load. Resultsshowed significant differences in the performance and emissioncharacteristics during mode switching between CI-to-PCCI com-bustion modes in a customized test-cycle of 300 s. BTE was rela-tively lower at lower engine loads and it increased at higherengine load. CI combustion showed slightly improved BTE com-pared to PCCI combustion. EGT variations with respect to engineload showed significantly lower EGT in PCCI combustion modecompared to CI combustion mode. Regulated emission resultsshowed significantly lower NOx emissions in PCCI combustionmode, however, HC emissions were higher compared to CI com-bustion mode. Higher CO emission during mode switching fromPCCI-to-CI combustion mode was an important observation ofthis study. Particulate emission results revealed significantly lowerTPN during PCCI combustion mode compared to CI combustionmode. The presence of nanoparticles in the exhaust from CI com-bustion mode showed its significantly higher health risk potential.Unregulated emission results showed that the concentrations ofHCHO, SO2, C7H8, and HNCO increased with increasing engineload during PCCI combustion mode, however it reduced slightlyin the CI combustion mode.

Based on overall experimental analysis and results, it can beconcluded that mode switching technique has significant potentialfor commercial application of PCCI combustion in a productiongrade engine. This study indicated that it is suitable to employPCCI combustion at lower-to-medium engine loads and conven-tional CI combustion at higher engine loads by employing mode

switching technology and thus reaping the complete benefits ofLTC for commercial applications such as in automotive sector.

Funding Data

� Technology Systems Group, Department of Science andTechnology (DST), Ministry of Science and Technology,Government of India (Grant No. DST/TSG/AF/2011/144-Gdated 14-01-2013).

Nomenclature

AMP ¼ accumulation mode particlesBMEP ¼ brake mean effective pressure

BTE ¼ brake thermal efficiencyCI ¼ compression ignition

CO2 ¼ carbon dioxideECU ¼ engine control unit

EEPS ¼ engine exhaust particle sizerEGR ¼ exhaust gas recirculationEGT ¼ exhaust gas temperature

FIP ¼ fuel injection pressureFTIR ¼ Fourier transform infrared

HCHO ¼ formaldehydeHCOOH ¼ formic acid

HNCO ¼ iso-cyanic acidIC ¼ internal combustion

IMEP ¼ indicated mean effective pressureLTC ¼ low temperature combustion

NMP ¼ nucleation mode particlesNOx ¼ oxides of nitrogen

PCCI ¼ premixed charge compression ignitionPM ¼ particulate matter

RoPR ¼ rate of pressure riseSO2 ¼ sulfur dioxideTPN ¼ total particle number

References[1] Agarwal, A. K., Singh, A. P., and Maurya, R. K., 2017, “Evolution, Challenges

and Path Forward for Low Temperature Combustion Engines,” Prog. EnergyCombust. Sci., 61, pp. 1–56.

[2] Weiskirch, C., and Mueller, E., 2007, “Advanced in Diesel Engine Combustion:Split Combustion,” SAE Paper No. 2007-01-0178.

[3] Musculus, M. P. B., Miles, P. C., and Pickett, L. M., 2013, “Conceptual Modelsfor Partially Premixed Low-Temperature Diesel Combustion,” Prog. EnergyCombust. Sci., 39(2–3), pp. 246–283.

[4] Santoso, H., Matthews, J., and Cheng, W. K., 2005, “Managing SI/HCCI Dual-Mode Engine Operation,” SAE Paper No. 2005-01-0162.

[5] Milovanovic, N., Dave, B., and Gedge, S., 2005, “Cam Profile Switching andPhasing Strategy Versus Fully Variable Valve Train Strategy for SwitchingBetween Spark Ignition and Controlled Auto Ignition Modes,” SAE Paper No.2005-01-0766.

[6] Zhang, Y., Xie, H., and Zhao, H., 2009, “Investigation of SI-HCCI HybridCombustion and Control Strategies for Combustion Mode Switching in a Four-Stroke Gasoline Engine,” Combust. Sci. Tech., 181(5), pp. 782–799.

[7] Narayanaswamy, K., and Rutland, C., 2006, “A Modeling Investigation ofCombustion Control Variables During DI-Diesel HCCI Engine Transients,”SAE Paper No. 2006-01-1084.

[8] Fang, F. Y., Ouyang, M., Gao, G., and Chen, L., 2013, “CombustionMode Switching Control in a HCCI Diesel Engine,” Appl. Energy, 110, pp.190–200.

[9] Wang, J. M., 2008, “Hybrid Robust Air-Path Control for Diesel Engines Opera-tion Conventional and Low Temperature Combustion and Conventional DieselCombustion Modes,” IEEE Trans Control Syst. Technol., 16(6), pp.1138–1151.

[10] Burton, J., Williams, D., Glewen, W., and Andrie, M., 2009, “Investigation ofTransient Emissions and Mixed Mode Combustion for a Light Duty DieselEngine,” SAE Paper No. 2009-01-1347.

[11] Busch, S., Bohac, S. V., and Assanis, D. N., 2008, “A Study of the TransitionBetween Lean Conventional Diesel Combustion and Lean, Premixed, Low-Temperature Diesel Combustion,” ASME J. Engine Gas Turbines Power,130(5), p. 052804.

[12] Black, J., Eastwood, P. G., Tufail, K., Winstanley, T., and Hardalupas, Y.,2007, “Inter-Correlations Between Smoke Opacity, Legal Particulate Sampling(LPS) and TEOM, During Transient Operation of a Diesel Engine,” SAE PaperNo. 2007-01-2060.

092201-10 / Vol. 140, SEPTEMBER 2018 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use

[13] Hagena, R. J., Filipi, Z. S., and Assanis, D. N., 2006, “Transient Diesel Emis-sions: Analysis of Engine Operation During a Tipin,” SAE Paper No. 2006-01-1151.

[14] Tanabe, K., Komatsu, F., and Nakayama, S., 2011, “A Study on Mode Transi-tion Control Between PCI and Conventional Combustion in a Diesel Engine,”Int. J. Engine Res., 12(1), pp. 69–86.

[15] Banerjee, S., and Rutland, C., 2012, “Numerical Investigation of High PoweredDiesel Mode Transition Using Large Eddy Simulations,” SAE Paper No. 2012-01-0693.

[16] Han, S., Bae, C., and Choi, S. B., 2012, “Effects of Operating Parameters onMode Transition Between Low Temperature Combustion and ConventionalCombustion in a Light Duty Diesel Engine,” Int. J. Engine Res., 14(3), pp.231–246.

[17] Han, S., Kim, H., and Bae, C., 2013, “Strategy for Mode Transition BetweenLow Temperature Combustion and Conventional Combustion in a DieselEngine,” SAE Paper No. 2013-24-0058.

[18] Kim, K., Han, S., and Bae, C., 2011, “Mode Transition Between Low Tempera-ture Combustion and Conventional Combustion With EGR and Injection Modu-lation in a Diesel Engine,” SAE Paper No. 2011-01-1389.

[19] Rohani, B., Park, S., and Bae, C., 2015, “Effect of Injection Strategy on LowTemperature Conventional Diesel Combustion Mode Transition,” SAE PaperNo. 2015-01-0836.

[20] Bae, C., Rohani, B., and Park, S. S., 2016, “Effect of Injection Strategy onSmoothness, Emissions and Soot Characteristics of PCCI-Conventional DieselMode Transition,” Appl. Therm. Eng., 93, pp. 1033–1042.

[21] Ouyang, M., Yang, F., Gao, G., Chen, L., and Yang, Y., 2013, “Research on a Die-sel HCCI Engine Assisted by an ISG Motor,” Appl. Energy, 101, pp. 718–729.

[22] Shi, L., Hu, W., and Deng, K., 2014, “Effects of Fuel Compensation in Transi-tional Cycles on the Smoothness of Combustion Mode Switching in a DieselEngine,” Fuel Process. Technol., 118, pp. 55–63.

[23] Deng, K., Xu, M., and Gui, Y., 2015, “Fuel Injection and EGR Control Strategyon Smooth Switching of CI/HCCI Mode in a Diesel Engine,” J. Energy Inst.,88(2), pp. 157–168.

[24] Agarwal, A. K., Shukla, P. C., Patel, C., Gupta, J. G., Sharma, N., Prasad, R.K., and Agarwal, R. A., 2016, “Unregulated Emissions and Health Risk Poten-tial From Biodiesel (KB5, KB20) and Methanol Blend (M5) Fuelled Transpor-tation Diesel Engines,” Renewable Energy, 98, pp. 283–291.

[25] Singh, A. P., and Agarwal, A. K., 2015, “Diesoline, Diesohol, and DiesoseneFuelled HCCI Engine Development,” ASME J. Energy Resour. Technol.,138(5), p. 052212.

[26] Hou, J., Liu, J., Wei, Y., and Jiang, Z., 2016, “Experimental Study on In-Cylinder Pressure Oscillations of Homogenous Charge CompressionIgnition–Direct Injection Combustion Engine Fueled With Dimethyl Ether,”ASME J. Energy Resour. Technol., 138(5), p. 052211.

[27] Gao, T., Divekar, P., Asad, U., Han, X., reader, G. T., Wang, M., Zheng, M.,and Tjong, J., 2013, “An Enabling Study of Low Temperature CombustionWith Ethanol in a Diesel Engine,” ASME J. Energy Resour. Technol., 135(4),p. 042203.

[28] Bureau of Indian Standards, 1999, “Automotive Vehicles—Exhaust emis-sions—Gaseous Pollutants From Vehicles Fitted With Compression IgnitionEngines—Method of Measurement,” Bureau of Indian Standards, New Delhi,India, Indian Standard No. 14273.

[29] Singh, A. P., Jain, A., and Agarwal, A. K., 2017, “Fuel-Injection Strategy forPCCI Engine Fueled by Mineral Diesel and Biodiesel Blends,” Energy Fuels,31(8), pp. 8594–8607.

[30] Nord, A. J., Hwang, J. T., and Northrop, W. F., 2017, “Emissions From a DieselEngine Operating in a Dual-Fuel Mode Using Port-Fuel Injection of HeatedHydrous Ethanol,” ASME J. Energy Resour. Technol., 139(2), p. 022204.

[31] Dumitrescu, C. E., Cheng, A. S., Kurtz, E., and Mueller, C. J., 2017, “A Com-parison of Methyl Decanoate and Tripropylene Glycol Monomethyl Ether forSoot-Free Combustion in an Optical Direct-Injection Diesel Engine,” ASME J.Energy Resour. Technol., 139(4), p. 042210.

[32] Agarwal, A. K., Gadekar, S., and Singh, A. P., 2018, “In-Cylinder Flow Evolu-tion Using Tomographic Particle Imaging Velocimetry in an Internal Combus-tion Engine,” ASME J. Energy Resour. Technol., 140(1), p. 012207.

[33] Fang, W., Kittelson, D. B., and Northrop, W. F., 2017, “Dilution Sensitivity ofParticulate Matter Emissions From Reactivity-Controlled Compression IgnitionCombustion,” ASME J. Energy Resour. Technol., 139(3), p. 032204.

[34] Grosjean, E., Rasmussen, R. A., and Grosjean, D., 1998, “Ambient Levels ofGas Phase Pollutants in Porto Alegre,” Brazil Atmos. Environ., 32(20), pp.3371–3379.

[35] Ellenhorn, M. J., and Barceloux, D. G., 1988, Medical Toxicology: Diagnosisand Treatment of Human Poisoning, Elsevier, New York, pp. 609–610.

[36] Agarwal, A. K., Shukla, P. C., Gupta, J. G., Patel, C., Prasad, R. K., andSharma, N., 2015, “Unregulated Emissions From a Gasohol (E5, E15, M5, andM15) Fuelled Spark Ignition Engine,” Appl. Energy, 154, pp. 732–741.

[37] Berm�udez, V., Lujan, J. M., Pla, B., and Linares, W. G., 2011, “ComparativeStudy of Regulated and Unregulated Gaseous Emissions During NEDC in aLight-Duty Diesel Engine Fuelled With Fischer Tropsch and Biodiesel Fuels,”Biomass Bioenergy, 35(2), pp. 789–798.

[38] Benignus, V. A., 1981, “Health Effects of Toluene: A Review,” Neurotoxicol-ogy, 2(3), pp. 567–588.

[39] Held, T. J., and Dryer, F. L., 1998, “A Comprehensive Mechanism for Metha-nol Oxidation,” Int. J. Chem. Kinet., 30(11), pp. 805–830.

[40] Wentzell, J. J. B., Liggio, J., Li, S. M., Vlasenko, A., Staebler, R., Lu, G., Poi-tras, M. J., Chan, T., and Brook, J. R., 2013, “Measurements of Gas PhaseAcids in Diesel Exhaust: A Relevant Source of HNCO?,” Environ. Sci. Tech-nol., 47(14), pp. 7663–7671.

[41] Brady, J. M., Crisp, T. A., Collier, S., Kuwayama, T., Forestieri, S. D., Perraud,V., Zhang, Q., Kleeman, M. J., Cappa, C. D., and Bertram, T. H., 2014, “Real-Time Emission Factor Measurements of Isocyanic Acid From Light Duty Gaso-line Vehicles,” Environ. Sci. Technol., 48(19), pp. 11405–11412.

[42] Roberts, J. M., Veres, P. R., Cochran, A. K., Warneke, C., Burling, I. R., Yokel-son, R. J., Lerner, B., Gilman, J. B., Kuster, W. C., Fall, R., and de Gouw, J.,2011, “Isocyanic Acid in the Atmosphere and Its Possible Link to Smoke-Related Health Effects,” Proc. Natl. Acad. Sci. U. S. A., 108(22), pp. 8966–8971.

Journal of Energy Resources Technology SEPTEMBER 2018, Vol. 140 / 092201-11

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 08/10/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use