Clean combustion of n-butanol as a next generation biofuel for diesel engines

Applied Energy xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Clean combustion of n-butanol as a next generation biofuel for diesel engines Xiaoye Han ⇑, Zhenyi Yang, Meiping Wang, Jimi Tjong, Ming Zheng Department of Mechanical, Automotive and Materials Engineering, University of Windsor, Ontario, Canada

h i g h l i g h t s
Neat n-butanol is studied and applied to replace diesel fuel for clean combustion.
Emission benefits and technical challenges of n-butanol combustion are identified.
Advanced combustion controls are studied to suit n-butanol’s distinctive properties.
An innovative split-combustion strategy is developed to enable full load operation.
n-Butanol combustion delivers diesel-like efficiency with reduced harmful emissions.

a r t i c l e

i n f o

Article history: Received 23 September 2016 Received in revised form 28 November 2016 Accepted 11 December 2016 Available online xxxx Keywords: n-Butanol Next generation biofuel Engine efficiency CO2 reduction Clean combustion Full load capability

a b s t r a c t This work investigates the applicability of n-butanol as a next generation biofuel to replace diesel in compression ignition engines for efficient operation, pollutant mitigation, and CO2 reduction. A high compression ratio (18.2:1) diesel research engine is configured to run on neat n-butanol. Due to the fuel property departure from diesel, n-butanol combustion exhibits striking combustion characteristics. Alternative combustion strategies, including via partially premixed compression ignition and homogeneous charge compression ignition, are enabled efficiently owing to distinctive fuel properties of n-butanol. The compression ignition of the (partially) premixed n-butanol and air mixture is capable of producing diesel-like engine efficiency and significant nitrogen oxide and smoke reductions. As the engine load increases, however, such neat n-butanol combustion exhibits rapid burning and suffers abrupt pressure rise. Thereby the engine load is generally limited below 50% of the baseline capability. A split-combustion strategy, which employs multiple event fuel injections, is found to be effective to modulate the noise of n-butanol clean combustion, thereby enabling neat n-butanol application across the full engine load range. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Biofuels have raised growing interest in recent years on energy security, greenhouse gas (GHG) mitigation, and socioeconomic harmonization, while the fossil based petroleum fuels remain dominating in the world energy supply [1,2]. Policy-makers are also paying close attention to the use of biofuels; in the United States, the renewable fuel standard is set to sustain the growth of the biofuel industry [3]. Biofuels are among the leading contenders to replace petroleum fuels in the transportation sector for best using the existing powertrain designs and re-fuelling infrastructures. Currently, fuel blends rather than neat biofuels are commonly used for internal combustion engine (ICE) applications, for instance, 5–20% biodiesel (B5-B20) in compression ⇑ Corresponding author. E-mail address: [email protected] (X. Han).

ignition (CI) engines, and 10, 20 and 85% ethanol (E10, E20, and E85) in spark ignition (SI) engines to partially replace diesel and gasoline fuels. The use of 100% biofuels (e.g. B100) is largely in the research and development stage, except for the flex-fuel vehicles capable of running up to 100% ethanol fuel (E100) for Brazilian applications [4]. As suggested by the research findings, the use of biofuels has shown substantial benefits in reducing GHG emissions of the engine systems and on the life-cycle basis [5–10]. In order to circumvent the competition between food and fuel, the next generation biofuels, or the second generation biofuels, are produced from non-food feedstock (e.g. lignocellulose feedstock) and/or food crops that have already fulfilled the food purpose (e.g. vegetable oil waste), which separates them from the first generation biofuels and enables the potential for sustainable, affordable, and environmental friendly fuel supply [11–16]. Butanol is deemed as one of the next generation biofuels for transportation and combustion engine applications [17–19]. Traditionally, bio-

http://dx.doi.org/10.1016/j.apenergy.2016.12.059 0306-2619/Ó 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

2

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

Nomenclature ABE BMEP CA CA5 CA50 CA95 CI CO COV CO2 DI EGR EOC FTIR FSN GHG HC HCCI HCLD HFID

Acetone-Butanol-Ethanol brake mean effective pressure crank angle crank angle of 5% total heat release crank angle of 50% total heat release crank angle of 95% total heat release compression ignition carbon monoxide coefficient of variation carbon dioxide direct injection exhaust gas recirculation end of combustion Fourier Transform Infrared Spectroscopy filter smoke number green house gas hydrocarbon homogeneous charge compression ignition heated chemiluminescence detector heated flame ionization detector

butanol can be synthesized through the Acetone-Butanol-Ethanol (ABE) fermentation process using feedstocks such as cereal grains and sugar, but this process usually suffers from low yields of butanol and receives criticism of causing food shortage [20]. Recent developments in the fermentation process, such as fedbatch, continuous syngas fermentation and immobilized cell fermentation, have substantially improved bio-butanol production with respect to substrate costs, low productivity and downstream process cost and, particularly, the use of non-food feedstock such as lignocellulose materials tackles the confliction with food supply [20–24]. Compared to the most commonly used bio-alcohol fuel, i.e. ethanol, butanol has several advantages for combustion engine applications. Butanol is less corrosive and less prone to water contamination, and thus it is considered as a ‘‘drop-in” fuel for the existing fuel distribution infrastructure. Butanol has around 25% higher energy density than that of ethanol, and lower fuel consumption and better mileage can be achieved. In addition, butanol can blend with both gasoline and diesel fuels to be used in SI and CI engines for a wide range of applications. Butanol possesses similar physical properties to those of gasoline, and many detailed investigations have been performed to study the fuel consumption, combustion characteristics, engine performance, and exhaust emissions of using butanol-gasoline blends on stoichiometric-burn SI engines [25–36]. Applications of butanol-gasoline blends and neat butanol have also been demonstrated on production engines without modifications [37,38]. In these research studies, the similarity in fuel properties is taken as the advantage of butanol to substitute gasoline with minimum compromise on engine performance. In contrast, the use of butanol in CI engines is often researched to utilize butanol’s distinctive fuel properties, e.g. fuel-borne oxygen, for emission reduction of diesel engines. The majority of engine experiments use butanol-diesel blends of different fuel ratios to study the engine performance and exhaust emissions [39–46]. These research studies cover a wide range of diesel engine application and combustion modes, including steady state and transient cycles, naturally aspirated and turbocharged air induction, automotive and stationary engines, and conventional high temperature combustion and premixed low temperature

HFRR HRR ICE IMEP LHV LTC MFB NDIR NOx PFI ppm PPCI pinj pint PRRmax SOC SOI TDC THC WSD

high frequency reciprocating rig heat release rate internal combustion engine indicated mean effective pressure lower heating value low temperature combustion mass fraction burned non-dispersive infrared detector nitrogen oxides port fuel injection parts per million partially premixed compression ignition pressure of injection pressure of intake maximum pressure rise rate start of combustion start of injection top dead centre total hydrocarbons wear scar diameter

combustion. A consensus is that the addition of butanol noticeably reduces the soot emissions owing to added fuel-borne oxygen and enhanced volatility from butanol. The nitrogen oxides (NOx) in the exhaust emissions can decrease or slightly increase depending on the engine operating conditions, while the use of exhaust gas recirculation (EGR) can effectively suppress the NOx emissions. Another butanol application in CI engines is the recent development of dual fuel engine operations. In butanol-diesel dual fuel combustion, the two fuels are delivered individually instead of using fuel blends, and the fuel blending occurs inside the cylinder [47–51]. A secondary intake port fuel injection (PFI) is added to a direct injection (DI) diesel engine, and butanol is therefore delivered via PFI similar to its application in SI engines but usually ignited via the direct injection of diesel. A major advantage of dual fuel combustion is that the fuel ratio can be adjusted on the fly by controlling the fuelling rate of each fuel, rather than a fixed fuel ratio as the fuel blend in the fuel tank. In general, these dual fuel test results show the benefits of adding an oxygenated fuel to diesel; the better mixing and oxygen availability result in soot reduction and thus permit higher EGR ratios to reduce NOx emissions. The dynamic control of fuel blending ratio offers additional flexibility in combustion control to better accommodate different engine operating conditions. Despite the butanol fuel blends and dual fuel applications are actively investigated, a relatively limited number of studies have been performed to explore the neat butanol application that completely replaces diesel in CI engines. The combustion characteristics and fuel spray of neat n-butanol have been investigated through fundamental studies in constant volume chambers that simulate diesel combustion environments [52–54]. Engine tests have also shown encouraging potentials of neat n-butanol to produce comparable engine efficiency to its diesel counterpart while emitting substantially lower NOx and smoke emissions [49,55– 58]. In the meantime, these research studies also manifest the engine load limitations of butanol combustion in diesel engines. A primary advantage of diesel engines is its high power density, and thus the use of a biofuel to replace diesel becomes less attractive if the fuel change results in a large degradation in engine performance and power density. Therefore, the authors intend to evaluate the benefits and challenges of neat n-butanol application

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

in a diesel engine due to fuel property departure, and explore innovative combustion strategies to expand the load capability of n-butanol combustion to match the diesel performance, thereby providing viable solutions to the practical fuel shifting from petroleum diesel to n-butanol, a next generation biofuel. 2. Experimental setup 2.1. Research platform Experiments are conducted on an advanced engine combustion research platform that is capable of systematic testing, comprehensive measurements, and detailed data analyses. A modern production four-cylinder diesel engine is re-configured to perform single cylinder research, and the original engine specifications are given in Table 1. Instrumentations and modifications are implemented on the engine for high quality combustion research. A schematic of the experimental setup for the research engine is illustrated in Fig. 1. An optical encoder (resolution of 0.1°CA) is used to determine the piston position and engine crank angle. A piezoelectric pressure transducer (AVL GU13P) is installed in place of the glow plug for cylinder pressure measurement. An external compressor with a conditioning unit supplies clean and dry air to simulate the engine intake boost. The air supply temperature is near the room temperature, usually at 25 °C. The air flow rate is measured by a volumetric air flow meter (Roots Meter 2M175) and, in order to attain reliable air flow measurement, an intake surge tank is placed between the air flow meter and the engine intake to dampen the pressure fluctuations created by the opening and closing events of engine valves. A pneumatic control valve is installed to restrict the exhaust outflow and thus regulate the exhaust backpressure. An EGR cooler and an EGR valve are used to control the EGR temperature and flow rate. During the engine testing, the exhaust backpressure is usually controlled at levels slightly higher (e.g. 2–5 kPa) than the intake boost to drive the EGR flow. An eddy current dynamometer is connected to the research engine for the control and measurement of engine speed and power. Two fuel carts are fabricated to deliver n-butanol, namely the low pressure port fuel injection and the high pressure direct injection. An in-house developed injection control system, consisted of real-time controllers with field programmable gate array devices, is used to command the injection timing and duration. When running n-butanol, an oxygenated fuel, potential fuel corrosion problems may occur in the fuelling systems. An alcohol compatible fuel flow meter (Ono Sokki FP-213) is used to measure the n-butanol fuel flow rate. Port fuel injectors capable of running ethanol (E100) are selected to deliver n-butanol at the intake runners. As the lubricity of n-butanol is beyond the permissible limit of the high pressure injection system, lubricity improver is added to avoid excessive wear. The major fuel properties of the test fuels are tabulated in Table 2.

Table 1 Research engine specifications. Engine type Displacement Bore Stroke Compression ratio Rated power Rated torque Maximum BMEP Maximum cylinder pressure Direct-injection system Additional port-injection system

4-Cylinder, 4-stroke 1998 cm3 86 mm 86 mm 18.2:1 97 kW @ 3800 rpm 320 N m @ 1800 rpm 20.7 bar @ 1800 rpm 180 bar Common-rail, up to 1600 bar One injector in each intake runner

3

2.2. Engine performance evaluation and combustion analysis The engine performance is evaluated on the indicated basis from cylinder pressure measurement and the combustion process is analysed through apparent heat release analysis based on the 1st law of thermodynamics. The net thermal efficiency is calculated for the entire engine cycle. These calculations follow Eqs. (1)–(3), where HRR is the apparent heat release rate [J/°CA], c is the ratio of specific heats for the cylinder charge, h is the engine crank angle [°CA], p is the measured cylinder pressure [N/m2] at crank angle h, V is the cylinder volume [m3] at crank angle h, Vd is the engine displacement [m3], IMEP is indicated mean effective _ f is the measured fuel flow rate [mg/cycle], pressure [N/m2], m and LHV is the lower heating value [kJ/g] of the test fuel.

HRR ¼


 1 dV dp þV  cp ðc  1Þ dh dh Z

IMEP ¼

ð1Þ

720

pðhÞdV=V d

ð2Þ

gind ¼ IMEP  V d =ðm_ f  LHVÞ  100%

ð3Þ

0

Based on the heat release analysis, critical timings of the combustion process are defined, including the start of combustion (SOC) represented by the crank angle when 5% of the total heat is released (CA5), the combustion phasing represented by the crank angle when 50% of the total heat is released (CA50), and the end of combustion (EOC) defined as the crank angle when 95% of the total heat is released (CA95). The combustion duration is the time period between CA5 and CA95, and the ignition delay is calculated using SOC and the commanded start of injection (SOI). The engine gaseous emissions are measured using a dual-bank analyser rack (analysers summarized in Table 3). The smoke emissions are measured by an AVL-415S smoke meter. By convention, the EGR ratio is the ratio between intake CO2 concentration and exhaust CO2 concentration. The combustion efficiency is defined as the energy percentage of burned fuel over the total fuel input. The amount of burned fuel is indirectly obtained from the measured fuel flow rate and measured incomplete combustion products in the exhaust (i.e. HC and CO).

3. Results and discussion Extensive engine tests have been carried out to investigate the engine performance and combustion characteristics of the diesel research engine running on neat n-butanol. The impacts of fuel replacement are first highlighted by comparing between nbutanol and diesel combustion processes under the same engine operating conditions. Subsequently, the partially premixed compression ignition (PPCI) and homogeneous charge compression ignition (HCCI) are enabled by taking advantages of n-butanol’s distinctive fuel properties to explore alternative combustion strategies for neat n-butanol application in the diesel research engine. A split-combustion strategy is thereafter implemented to achieve practical engine operations up to the rated engine load, which demonstrates the possibility of replacing diesel with nbutanol without compromising the engine performance. Finally, the fuel consumption of neat n-butanol combustion is discussed by comparing with diesel combustion on the same test engine to evaluate the tank-to-wheel CO2 emission of neat n-butanol combustion.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

4

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

Fig. 1. Schematic of the research engine setup.

600

1400

400

250

200

200

0 11

10

0

90

80

70

60

50

10

00

300

00

600

0

800

350

0

400

0

1000

0

450

0

1200

0

500

40

C4H9OH 810 25 87 33.1 21.62 117.66 28.8 343.3 591 2.6@25 °C

0

CH1.89 846 46.5 25 43.5 0 229–337 (T5–T95) 73.3 254–285 300 2.5

30

Formula Density @ 20 °C (kg/m3) Cetane number [–] Octane number [–] Lower heating value [MJ/kg] Oxygen content by mass [%] Boiling temperature [°C] Flash point [°C] Auto-ignition temperature @ 1 bar abs [°C] Lubricity, HFRR WSD @60 °C [lm] Kinematic viscosity @40 °C [cSt]

550

0

n-Butanol

20

Diesel

Opening Delay [µs]

Fuel

1600 Butanol Diesel

Closing Delay [µs]

Table 2 Major properties of test fuels [59–62].

Command Duration [µs] Table 3 Summary of emission analysers.

Fig. 2. n-Butanol vs. diesel injector opening and closing delays.

Analyzer type

Measured emissions

Model

Paramagnetic Flame ionization (HFID) Non-dispersive infrared (NDIR) Chemiluminescence (HCLD) FTIR

O2 [%] THC [ppm] CO [ppm] CO2 [%]

602P 300M HFID 200/300 NDIR

NO & NO2 [ppm] NO, light HC species, CO [ppm] Smoke [FSN]

600 HCLD MKS 2030HS

Smoke meter

AVL 415S

3.1. Changes in combustion characteristics due to fuel property departure Prior to engine experiments, the high pressure fuel injectors are tested with both diesel and n-butanol on an advanced injection bench to examine the fuel injection characteristics. These tests provide information about the actual fuel injection process compared with the injection command, such as the injector opening and closing delays. The injector opening delay is defined as the time difference between the start of the injection command and the initial appearance of the fuel exiting the injector nozzle and a similar definition applies to the injector closing delay. An example of these injector delays are shown in Fig. 2 for diesel and n-butanol under 900 bar injection pressure. The injector opening delay is nearly identical for the two fuels, while the injector closing delay data shows a slightly scattered pattern. Nonetheless, the fuel change

exhibits minor differences in the injector delays under the same injection conditions. In Fig. 3, fuel spray images obtained by high speed direct imaging are compared between diesel and n-butanol. The fuel injections are commanded for 1000 ls at 1200 bar injection pressure into an optical chamber charged with nitrogen at 40 bar pressure and room temperature. The two fuel sprays are very similar except that n-butanol exhibits a slightly narrower cone angle than that of the diesel spray. It is noted that the evaporation effect is not fully captured as the temperature in the optical chamber is substantially lower than the compression temperature in an actual diesel engine where n-butanol would evaporate faster than diesel owing to its lower boiling temperature and thus improve the fuel-air mixing. Although the offline injection testing shows minor differences in the injection processes of the two tests fuels, significant changes in combustion characteristics are observed during engine testing. In Fig. 4, the n-butanol combustion is compared with two diesel combustion cases, which match the combustion phasing (Diesel CA50) and injection timing (Diesel SOI) respectively. It is noted that misfire would occur if n-butanol is delivered at 353°CA to match diesel SOI in the ‘Diesel CA50’ case. The engine operating conditions and major combustion parameters are listed in Table 4. The n-butanol combustion presents a very sharp heat release profile and primarily a premixed burning process; whereas ‘Diesel CA50’ clearly shows a combustion process predominated by

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

5

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

Diesel 10 mm

n-Butanol 10 mm 0 µs

96 µs

144 µs

240 µs

336 µs

Fig. 3. Examples of high speed imaging n-butanol and diesel sprays – pinj 1200 bar, background pressure 40 bar, room temperature.

350 n-Butanol Diesel_CA50 Diesel_SOI

150

300 250 200

100

150 100 50

50

0 0 330

340

350

360

370

380

Heat Release Rate [J/°CA]

Cylinder Pressure [bar]

200

-50 390

Crank Angle [°CA] Fig. 4. n-Butanol vs. diesel combustion – heat release rate and cylinder pressure.

Table 4 Baseline comparison between n-butanol and diesel combustion. Operating conditions Engine speed [rpm] pint [bar abs] pinj [bar gauge] EGR [%]

1500 2 600 Zero

Characteristics

n-Butanol

Diesel SOI

Diesel CA50

Fuel injection Fuelling rate [mg/cycle] CA50 [°CA] IMEP [bar] dp/dhmax [bar/°CA] pmax [bar] Ignition delay [ms] Combustion duration [ms] gind [%] gcomb [%] NOx [ppm] Smoke [FSN] HC [ppm] CO [ppm]

900 ls@338°CA 21.9

650 ls@338°CA 17.1

650 ls@353°CA 17.0

366 6.6 20.5 144 2.83 1.04

350 6.0 16.6 166 1.00 2.09

366 6.7 5.1 137 0.79 2.98

45.3 97.51 32 0.007 93 1439

40.0 99.90 1376 0.126 16 34

45.4 99.85 652 0.249 30 49

diffusion burning. It is noticeable that the cylinder compression pressure is lower after n-butanol is injected compared with that in the ‘Diesel CA50’ case, as the heat absorption of n-butanol evaporation counteracts the pressure and temperature increase due to cylinder compression. The combustion of diesel starts much earlier when SOI is advanced to match the ‘n-Butanol’ case. The entire diesel combustion event (Diesel SOI) occurs before the top dead centre

(TDC) and the combustion pressure approaches 170 bar. It is noted that the ‘Diesel SOI’ case is impractical for normal engine operation but useful to demonstrate combustion differences resulted by the fuel change. As suggested by the previous work [54], n-butanol has a similar rate of injection as diesel does under the same conditions. In order to deliver comparable fuel energy, longer n-butanol injection duration is required due to its lower energy density. Ultimately, it leads to higher volumetric and mass fuel consumption to operate on nbutanol. In the CA50 matching cases, the fuelling rates are 21.9 mg/cycle for n-butanol and 17.0 mg/cycle for diesel to achieve similar IMEP levels (6.6 and 6.7 bar respectively); in the ‘Diesel SOI’ case, the diesel injection duration is kept the same as that in the ‘Diesel CA50’ case, and the fuelling rate is 17.1 mg/cycle. The excessive early diesel combustion leads to reduction in energy efficiency and thus the IMEP in the ‘Diesel SOI’ case drops to 6.0 bar. The fast burning of n-butanol results in much higher pressure rise rates compared with its diesel counterparts. Such a rapid upsurge of cylinder pressure not only produces loud audible combustion noise but also places repetitive and periodical stress on engine components (e.g. pistons and connecting rods), which can potentially cause severe engine damage. The fast burning phenomenon stems from the long ignition delay. In the presented cases, the ignition delay of n-butanol is more than double of the diesel ignition delay, during which n-butanol has prolonged time period to mix with in-cylinder air. The compression ignition of the (partially) premixed cylinder charge occurs at multiple locations across the combustion chamber, and thus the fuel energy is released within very short combustion duration (1/3 or 1/2 of that in the diesel cases), resulting in a rough and rapid burning process. The rapid heat release within a short duration causes high pressure rise rates, but it also leads to near constant-volume combustion that approaches the ideal Otto cycle and improves thermal efficiency. The indicated thermal efficiency of n-butanol is about the same (45.3% vs. 45.4%) as its diesel counterpart of the same combustion phasing; the combustion efficiency is slightly lower due to lower fuel reactivity and thus less complete burning. It is important to note that n-butanol combustion produces substantially lower NOx and near-zero smoke emissions, resulted by the lean burn of a highly premixed air-fuel mixture. As demonstrated by the baseline comparison, the fuel change apparently alters the combustion processes; n-butanol combustion is largely premixed burning in contrast to the predominant diffusion burning in diesel combustion, thereby resulting in remarkable differences in combustion characteristics, including challenges of excessive pressure rise rates, changes of injection timing, and advantages of low NOx and smoke emissions. It is also encouraging that the diesel test engine produces comparable thermal efficiency when operating on neat n-butanol.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

1500 rpm, Zero EGR Boost: 2 bar abs n-Butanol PPCI mf: 18.1 mg/cycle IMEP: 2.3~4 bar

4.0

Ign. Delay [ms]

3.5 3.0 2.5

Diesel ignition delay less than 1 ms 0.5 324

328

332

336

340

344

348

SOI [°CA] Fig. 5. n-Butanol PPCI SOI sweep – ignition delay (Ign. Delay).

352

90

10

85

0 140 130 120 110 100 90 376

80 25 20 15 10 5 0 24 n-Butanol PPCI, mf: 18.1 mg/cycle IMEP: 2.3~4 bar, Boost: 2 bar abs

372

18

368

12

364

6 0 328

332

336

340

344

348

COV_IMEP [%] PRRmax [bar/°CA] Comb. Eff. [%]

95

20

324

352

SOI [°CA] Fig. 6. n-Butanol PPCI SOI sweep – combustion phasing (CA50), co-efficient of variation of IMEP (COV_IMEP), maximum cylinder pressure (p_max), maximum pressure rise rate (PRRmax), indicated efficiency (Ind. Eff.), and combustion efficiency (Comb. Eff.).

A significant reduction from 90% to 85% is observed in combustion efficiency with retarded combustion phasing. Due to the incomplete and off-phasing combustion, the indicated thermal efficiency substantially decreases. As suggested by the test results, optimal engine efficiency and stability can be obtained within a relatively narrow SOI window. It is noted that the maximum cylinder pressure (p_max) and maximum pressure rise rate (PRRmax) are at their peaks in this SOI window. The levels of p_max and PRRmax during this test are within the engine hardware limits, as the test is performed at low engine load (2.3–4 bar IMEP). The combustion control will ultimately encounter challenges of excessive cylinder pressure and pressure rise rate when the engine load is raised. The n-butanol PPCI combustion offers significant benefits in NOx and smoke emissions, as shown in Fig. 7. The lean burn of a premixed cylinder charge has low flame temperatures and thus produces ultralow NOx (<20 ppm) and near-zero smoke. Similar to most low temperature combustion strategies, n-butanol PPCI has relatively high unburned HC and CO emissions. However, the combustion efficiency is expected to improve at higher engine

NOx [ppm]

1.0

30

360

2.0 1.5

100

2000 1600 1200 800 400 0 20

6000 5400 4800 4200 3600 3000 0.08

n-Butanol PPCI, mf: 18.1 mg/cycle IMEP: 2.3~4 bar, Boost: 2 bar abs

15

0.06

10

0.04

5

0.02

0

CO [ppm]

4.5

40

Smoke [FSN]

3.2.1. Neat n-butanol PPCI The fuel injection control of modern diesel engines equipped with high pressure direct injection systems typically modulates the injection timing and quantity of each injection event. These injection events play a critical role in determining the combustion process, especially the combustion timing and phasing. When diesel is replaced by neat n-butanol, the same injection control method can still be applied; however, the fuel change has substantial impacts on the correlation between the fuel injection and combustion process. The ignition delay results of an SOI sweep are shown in Fig. 5, where the n-butanol fuelling rate is kept constant but the fuel injection is commanded at different crank angles. The ignition delay shortens as n-butanol is delivered into a hotter environment at a later timing in the compression stroke. The n-butanol ignition delay ranges between 2.4 and 4 ms in this test, whereas the diesel ignition delay is usually less than 1 ms under the same test conditions. Therefore, n-butanol has substantially prolonged time to mix with in-cylinder air prior to the onset of combustion, thereby enabling partially premixed compression ignition (PPCI). The combustion characteristics of the same n-butanol SOI sweep are plotted in Fig. 6. Both early and late fuel injection timings result in retarded combustion phasing represented by later CA50, and unstable combustion process represented by higher coefficient of variations of IMEP (COV_IMEP). In the early injection timing cases, the injected n-butanol has adequate mixing time (e.g. 4 ms) and the cylinder charge reaches a high degree of homogeneity with an overall air excess ratio of 5.4 in this test. As a result, the ultra-lean and well-mixed cylinder charge exhibits slow and unstable burning. In the late injection timing cases, the combustion phasing is pushed further into the expansion stroke despite of the relatively shortened ignition delay. The combustion events occurring at late phasing become unstable and incomplete.

p_max [bar]

Alternative combustion strategies are investigated to accommodate n-butanol’s distinctive fuel properties. Owing to longer mixing time and better fuel atomization, the enhanced fuel-air mixing essentially enables partially premixed compression ignition (PPCI); the high volatility allows practical n-butanol port fuel injection, thereby providing the possibility to run homogeneous charge compression ignition (HCCI). These combustion strategies are therefore studied for neat n-butanol applications on the diesel test engine.

CA50 [°CA]

3.2. Alternative combustion strategies for burning neat n-butanol in diesel engines

Ind. Eff. [%]

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

THC [ppm]

6

0.00 324

328

332

336 340 SOI [°CA]

344

348

352

Fig. 7. n-Butanol PPCI SOI sweep – exhaust emissions, NOx, smoke, THC, and CO.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

7

40 30

Zero EGR, pinj: 600 bar Boost: 2 bar abs

20

mf: 26.9 mg/cycle 23.0 18.0

13.9

10

Maximum Pressure Rise Rate [bar/°CA]

0 40 30

pinj: 900 bar

Zero EGR mf: 23 mg/cycle Boost: 2 bar abs

20

pinj: 600 bar

10 0 40 30 20

Zero EGR mf: 23 mg/cycle pinj: 900 bar 1.75

10

Boost: 1.5 bar abs

0 40 30

2.0

Zero EGR

20 10

Boost: 2 bar abs mf: 23 mg/cycle pinj: 900 bar

Zero EGR

EGR: 30%

0 325

329

333

337

341

345

349

SOI [°CA] Fig. 8. n-Butanol PPCI SOI sweeps under different engine operating conditions – maximum pressure rise rate vs. EGR, boost, injection pressure (pinj), and fuelling rate (mf).

200 150 100 50 0 -50 200 150 100 50 0 -50 200 150 100 50 0 -50 200 150 100 50 0 -50

mf: 26.9 mg/cycle

SOI: 345°CA

23.0

18.0 13.9

SOI: 345°CA pinj: 900 bar pinj: 600 bar

Boost: 2 bar abs SOI: 345°CA

1.75, 345

1.5, 338

Zero EGR

SOI: 331°CA

30% EGR

330

340

350 360 370 Crank Angle [°CA]

380

450 350 250 150 50 -50 450 350 250 150 50 -50 450 350 250 150 50 -50 450 350 250 150 50 -50

Heat Release Rate [J/°CA]

loads and/or lower k where higher flame temperature can be achieved. As suggested by n-butanol test results and the comparison with diesel combustion, the fast burning process of n-butanol and resultant high pressure rise rates present a major concern to run n-butanol PPCI on the diesel test engine. In order to develop deeper understanding of n-butanol PPCI in this regard, investigations are carried out to study the effects of combustion control measures under different engine operating conditions. In Fig. 8, the results of maximum pressure rise rate are plotted across SOI sweeps for different fuelling rates, injection pressures, intake boost levels, and EGR ratios; representative traces of corresponding cylinder pressure and heat release rate under these test conditions are shown in Fig. 9. In all the presented cases, the fuel injection timing control shows a dominant influence on the level of maximum pressure rise rates. In the case of increasing fuelling rates to achieve higher engine load, the maximum pressure rise rates tend to escalate as the combustion starts earlier and releases more energy for a given fuel injection timing (Fig. 9). In addition, the maximum pressure rise rates at higher fuelling rates become more sensitive to a fixed increment (e.g. 1°CA) of fuel injection timing. The practical injection timing window narrows, where early injection timing is limited by excessive pressure rise rates and late injection timing is limited by unstable combustion. Such a high sensitivity and a narrow feasible window make the fuel injection control very challenging to be implemented in real-world engine operations. The highest IMEP levels achieved with the 4 different fuelling rates are 2.1, 4.2, 6.1, and 7.6 bar respectively.

Cylinder Pressure [bar]

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

390

Fig. 9. n-Butanol PPCI under different engine operating conditions – examples of cylinder pressure and heat release rate.

Other key combustion control techniques that modulate injection pressure, intake boost, and EGR ratios can be applied to mitigate the maximum pressure rise rates. Similar to their effects in diesel combustion, a lower n-butanol injection pressure reduces the peak of heat release during the premixed burning; less intake boost and more EGR slow down the ignition and combustion processes. In summary, the principles of diesel combustion control can still be applied to n-butanol PPCI. However, in stark contrast to robust diesel combustion, the efficient, safe and stable n-butanol PPCI usually occurs within a narrow combustion phasing window; as a result, the feasible range and sensitivity of these combustion control measures are noticeably different. 3.2.2. Neat n-butanol HCCI As discussed in the previous subsection, the changes in the combustion mode and control effectiveness primarily stem from the difference of fuel ignition resistance when replacing diesel with n-butanol. In point of fact, n-butanol is more similar to gasoline instead of diesel, especially the physical properties (e.g. volatility). Therefore, n-butanol is suitable to be delivered at the engine intake port as in conventional gasoline engines. The ignition of the premixed air and n-butanol mixture is initiated by the compression as in diesel engines, thereby enabling the homogeneous charge compression ignition (HCCI) operation. When enabling HCCI combustion, intake heating is commonly used in a research environment to facilitate fuel evaporation in the intake manifold and promote ignition of low reactivity fuels. However, the test engine is essentially a production diesel engine without any intake heating devices. In order to enhance the cylinder charge homogeneity, the engine coolant and oil are fully warmed up using external conditioning units prior to experiments. Since the boiling temperature of n-butanol is higher than the engine coolant temperature, open valve injection is adopted to avoid fuel pooling in the intake port. The fuel evaporation relies on the high compression temperature

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

150

1250

100

1000

50

n-Butanol HCCI IMEP: 2.1 ~ 6.7 bar

750

0 150

500 1.00

100

0.75

50

0.50

0

0.25

-50

0.00 350

360

370

360

370

380

Crank Angle [°CA] Fig. 10. n-Butanol HCCI load sweep – cylinder pressure (Cyl. Pressure), cylinder bulk gas temperature (Cyl. Temp.), heat release rate (HRR), and mass fraction burned (MFB).

pint: 2.0 bar abs pint: 1.5 bar abs

100 90 80 70 60

Ind. Eff. [%]

50 40 30 20 10

Comb. Eff. [%]

Lambda [-]

16 12 8 4 0

n-Butanol HCCI Zero EGR

0

1

2

3

4

5

6

7

8

IMEP [bar] Fig. 11. n-Butanol HCCI load sweep at different intake boost levels – Lambda, combustion efficiency (Comb. Eff.), indicated efficiency (Ind. Eff.).

n-Butanol HCCI Zero EGR

16 12 8 4 0

370 366 362 358

pint: 1.5 bar abs

COV_IMEP [%]

CA50 [°CA]

PRRmax [bar/°CA]

40 30 20 10 0

pint: 2.0 bar abs

0

1

2

3

4 5 IMEP [bar]

6

7

8

PRRmax [bar/°CA]

Fig. 12. n-Butanol HCCI load sweep at different intake boost levels – maximum pressure rise rate (PRRmax), co-efficient of variation of IMEP (COV_IMEP), and CA50.

40 30 20 10 0

n-Butanol HCCI pint: 1.5 bar abs mf: 24 mg/cycle IMEP: 6.4~7.5 bar

20 15 10 5 0

372

Comb.Dura. [°CA]

1500

Cyl. Temp. [K]

200

MFB [-]

HRR [J/°CA]

Cyl. Pressure [bar]

so that the injected n-butanol can evaporate and mix with air before combustion. In Fig. 10, demonstration tests are shown for n-butanol HCCI at different fuelling rates and thus different engine load levels (2.1–6.7 bar IMEP). The intake boost is 2 bar abs and EGR is not applied. The traces of cylinder pressure and heat release rate exhibit a similar trend as those in the n-butanol PPCI tests that the combustion releases more energy at an earlier phasing as the engine load increases. The calculated mass fraction burned (MFB) also shows a faster fuel burn rate at a higher engine load. The calculated cylinder gas temperature is used as an estimation of the flame temperature in HCCI combustion. Even though the cylinder gas temperature noticeably rises with increased engine load, the temperature levels are still below the threshold (e.g. 1800 K) for thermal NOx generation, and thus the NOx emissions of these n-butanol HCCI combustion tests are generally less than 10 ppm. In addition, nearly zero smoke is produced owing to the high degree of homogeneity, low flame temperature, and fuel-borne oxygen. Different from n-butanol PPCI where the fuel injection plays a critical role in combustion control, n-butanol HCCI largely relies on the air path control to modulate the combustion process, as the port fuel injection events almost completely separate from the combustion process and thus have nearly negligible impacts. The cylinder charge dilution, by either boost or EGR, essentially has the primary control over the compression ignition of the homogeneous n-butanol and air mixture. The effects of these two control measures are demonstrated by the test results shown in Figs. 11–15. As shown in Fig. 11, the n-butanol HCCI runs extremely lean in the low load engine test. As discussed previously, the ultra-lean burn of a well-mixed cylinder charge suffers from a high degree of incomplete combustion. The test results show that a lower intake boost helps in improving the combustion efficiency and thus the indicated thermal efficiency. When the intake boost is reduced, air flow rate becomes less, resulting in a lower excess air ratio (k) for a given fuelling rate; higher flame temperatures are therefore expected as a similar amount of fuel energy is released to heat up less cylinder charge, and hence the fuel burns more completely in a hotter environment. However, as suggested by previous studies under similar test conditions with n-butanol [57] and gasoline [63], higher intake boost promotes and facilitates the compression ignition of a homogeneous air-fuel mixture. As a result, the n-butanol HCCI with a lower intake pressure burns slower and exhibits later combustion phasing, even though producing higher flame temperatures. Such a phenomenon is clearly manifested by the test results shown in

CA50 [°CA]

8

366 360 0

5

10

15

20

25

30

35

40

45

50

EGR Ratio [%] Fig. 13. n-Butanol HCCI EGR sweep – maximum pressure rise rate (PRRmax), combustion duration (Comb. Dura.), and CA50.

Fig. 12. At low load (e.g. 1–2 bar IMEP), the lower intake pressure and late combustion also present higher cycle-to-cycle variations. It is important to note that the maximum pressure rise rates are very close between the two cases of different intake pressures. Similar to the in-cylinder temperature rise during combustion, the pressure rise rate should be higher when the intake pressure is lower and the air charge is less but the same fuel energy is

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

9

47 45 43 41 39

100 98 96 94 92

n-Butanol HCCI pint: 1.5 bar abs mf: 24 mg/cycle IMEP: 6.4~7.5 bar

5

10

15

20 25 30 35 EGR Ratio [%]

40

45

50

Fig. 14. n-Butanol HCCI EGR sweep – indicated efficiency (Ind. Eff.), combustion efficiency (Comb. Eff.), and co-efficient of variation of IMEP (COV_IMEP).

Cylinder Pressure [bar]

120 90

500

n-Butanol HCCI pint: 1.5 bar abs mf: 24 mg/cycle IMEP: 6.4~7.5 bar EGR Sweep

400 300

60

200 EGR: 5%

30 0 -30

23%

38% 40% 42%

100 0

Heat Release Rate [J/°CA]

150

-100 340 345 350 355 360 365 370 375 380 385 390 Crank Angle [°CA]

Fig. 15. n-Butanol HCCI EGR sweep – cylinder pressure and heat release rate.

released; however, such an effect happens to be counterpoised by the later combustion phasing and slower burning. As a result, the change in boost pressure shows minor influence on the pressure rise rates under the tested engine operating conditions. In contrast, the dilution by EGR has noticeable impacts in mitigating the maximum pressure rise rates, as shown in Fig. 13. At a constant n-butanol fuelling rate, increasing EGR ratio delays combustion phasing, prolongs combustion duration, and substantially reduces the maximum pressure rise rates. Whereas intake boost creates leaner air-fuel mixture by increasing the air amount, EGR dilutes the cylinder charge with re-cycled exhaust while keeping the amount of overall cylinder charge approximately constant. As the re-cycled exhaust (containing CO2 and water) replaces a part of the fresh air intake, the cylinder charge has reduced reactivity and increased specific heat capacity, which result in later combustion, slower burning, and lower flame temperatures. Fig. 14 shows engine efficiency and combustion stability results across the EGR sweep, and Fig. 15 shows the corresponding cylinder pressure and heat release traces. At low EGR ratios, e.g. 5%, the n-butanol HCCI combustion occurs at a very early combustion phasing for the given fuelling rate. As additional EGR postpones the combustion towards 367°CA that is deemed as the optimal combustion phasing for thermal efficiency of the test engine [64], the indicated thermal efficiency improves, with minor reduction in combustion efficiency. It is however noted that the n-butanol HCCI combustion becomes very sensitive to EGR ratio change from 38% to 42%, where the combustion transitions into slow and long

3.2.3. Neat n-butanol split-combustion The PPCI and HCCI combustion strategies are adopted to better accommodate n-butanol’s distinctive fuel properties. Trade-off is observed among excessive pressure rise rates, limited engine load capability, ultralow NOx and smoke emissions, and efficient and stable combustion. It is import to note that, the engine performance should not be significantly compromised when replacing the petroleum diesel fuel with a biofuel, n-butanol in this case. Moreover, the maximum pressure rise rates of modern production diesel engines range about 5–8 bar/°CA for smooth engine operation and combustion noise control, whereas the n-butanol PPCI and HCCI easily produce pressure rise rates near 20 bar/°CA. Therefore, the split combustion strategy, which employs multiple combustion events within a single engine cycle, is investigated to avoid excessive pressure rise rates and attain full load capability by modulating the n-butanol combustion process. Two examples of n-butanol split combustion are shown in Fig. 17, along with n-butanol HCCI and PPCI examples. The comparison is made at an engine load of 8 bar IMEP. High EGR ratios are used for HCCI and PPCI to mitigate pressure rise rates while maintaining relatively efficient and stable combustion. The presented split combustion deploys two combustion events where the second combustion events are enabled by late direct injections. The first combustion event is essentially low load n-butanol PPCI in the ‘Split 1’ case and low load n-butanol HCCI in the ‘Split 2’ case, depending on the fuel delivery method. In either split combustion case, the peak of heat release rate is much lower than those of PPCI and HCCI while delivering the same indicated engine power output. In addition, the two heat release events of n-butanol split

IMEP [bar]

0

burning processes with worsened combustion stability and reduced combustion efficiency. To a great extent, the HCCI tests resemble the previous PPCI tests in that, at an engine load 7 bar IMEP, the trade-off between pressure rise rates and efficient combustion presents a major challenge of running neat n-butanol in the diesel test engine. Fig. 16 shows the results of n-butanol HCCI tests aiming to raise the engine load levels by increasing fuel flow rate and adding EGR simultaneously. Ultimately, the engine load is increased up to 10 bar IMEP with an EGR ratio of 55%. It is clear that the PRRmax vs. EGR curves are steeper at higher engine loads, suggesting that the n-butanol HCCI combustion becomes more sensitive to the EGR change and more challenging to be practically implemented.

PRRmax [bar/°CA]

8 6 4 2 0

Comb.Eff. [%]

COV_IMEP [%]

Ind.Eff. [%]

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

11 10 9 8 7 6 5 26

33.4

n-Butanol HCCI pint: 2 bar abs EGR / Load Sweep

29.3 24.5

26.0

mf: 23 mg/cycle

22 18 14 10 0

5 10 15 20 25 30 35 40 45 50 55 60 EGR Ratio [%]

Fig. 16. n-Butanol HCCI EGR and load sweep – IMEP, PRRmax vs. EGR ratio.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

n-Butanol Speed: 1500 rpm Boost: 2 bar abs IMEP: 8.1~8.2 bar

150 120

mf [mg/cycle] 27.7, 27.5, 28.6, 30 EGR [%] 50, 40, 30, Zero

90

320 280 240 200

60

HCCI

160

PPCI

30

120

0

80

Split 2

Split 1

-30

40

-60

0

-90

-40 330

340

350 360 370 380 Crank Angle [°CA]

390

400

Fig. 17. n-Butanol split combustion compared to HCCI and PPCI – cylinder pressure and heat release rate.

0 1.00 MFB [-]

0.75 0.50

40 35 30 25

95 90 85 80

Split 2

HCCI Split 1 355

150

360

365 370 375 Crank Angle [°CA]

n-Butanol Split combustion Speed: 1500 rpm Boost: 2 bar abs IMEP [bar]: 13.9, 16.5, 19.0

100

50

150

100

50

0

PPCI

0.00 350

A commonly held concern about alcohol fuels and other biofuels is the possibly increased fuel consumption. A greater amount of an alcohol fuel, e.g. n-butanol, needs to be burned to deliver the same amount of fuel energy, since alcohol fuels generally have lower energy density (usually represented by the lower heating

100

n-Butanol Speed: 1500 rpm Boost: 2 bar abs IMEP: 8.1~8.2 bar

0.25

3.3. Considerations of fuel consumption and CO2 emission

Cylinder Pressure [bar]

10

45

Comb. Eff. [%]

20

Ind. Eff. [%]

Split 2

PPCI

HCCI

30

Split 1

PRRmax [bar/°CA]

combustion can be individually controlled by adjusting the fuel injection amount and scheduling, thereby offering desired controllability over the n-butanol combustion process. The maximum pressure rise rates, engine efficiencies and mass fraction burned are compared in Fig. 18 between HCCI, PPCI, and split combustion cases. By splitting the combustion process into two stages (as shown by MFB curves), the maximum pressure rise rates are substantially reduced towards the levels of modern production diesel engines. The indicated thermal efficiency of nbutanol split combustion is slightly lower as the phasing of each individual combustion event departs from the optimum (i.e. 367°CA). In ‘Split 1’, the second combustion event initiates prior to the end of the first, and n-butanol starts to burn as it is injected into the hot combustion chamber. Such a combustion process resembles the conventional diesel diffusion burning and achieves a high degree of combustion completeness. In ‘Split 2’, the combustion efficiency suffers from both the port fuel injection and the late second combustion event. Based on observations in the previous HCCI and PPCI tests, the thorough mixing for the first HCCI-like combustion event results in an ultra-lean air-fuel mixture throughout the combustion chamber and ultimately leads to high levels of unburned and partially burned fuel in the exhaust; the second combustion event can burn a certain amount of incomplete combustion products from the first HCCI-like combustion event but the second combustion itself tends to generate HC and CO

emissions when it occurs at a late phasing. It is noted that the second combustion event in ‘Split 2’ is intentionally placed at a late phasing for demonstration purpose and it can be advanced for better engine performance. A more important significance of the split combustion strategy is to safely raise the engine load to much higher levels than those of PPCI and HCCI combustion. In Fig. 19, examples of n-butanol split combustion at high engine loads are illustrated by the cylinder pressure and heat release rate traces along with the injection commands. An engine load of 13.9 bar IMEP is achieved with the two-stage split combustion, and a third combustion event is switched on to further increase the engine load, to 16.5 and ultimately 19.0 bar IMEP. As the third combustion event occurs late after the first two, it normally has no impact on the peaks of cylinder pressure and pressure rise rate. A drawback from the late phasing of the third combustion is the generally lower thermal efficiency, because less cylinder expansion can be utilized to produce useful work. Test results of neat n-butanol split combustion at mid-high engine loads are shown in Fig. 20. Across the tested engine loads, the maximum pressure rise rates are controlled between 5 and 8 bar/°CA by following the criterion for modern production diesel engines. The results of both two-stage and three-stage split combustion strategies are overlaid on the same plots. The comparison also includes results of different fuel injection methods (PFI and DI) for enabling the first combustion event in the n-butanol split combustion. When PFI is used to generate the HCCI-like combustion for the first combustion event, lower combustion efficiency is observed. Nonetheless, as the flame temperature increases at higher engine loads, the combustion efficiency improves and approaches 99% which is comparable to that of diesel combustion. The indicated thermal efficiency exhibits an increasing trend as the engine load is raised from low-mid load to mid-high load, but appears to reach a plateau at 16–19 bar IMEP when three-stage split combustion is employed. As discussed earlier, the third combustion event in the n-butanol split combustion suffers from late phasing and resultant lower energy efficiency, which in turn causes negative impacts on the overall indicated thermal efficiency.

0

Heat Release Rate [J/°CA]

180

Heat Release Rate [J/°CA]

Cylinder Pressure [bar]

10

Injection command

380

385

390

-50

-50 330

345

360

375

390

405

420

Crank Angle [°CA] Fig. 18. n-Butanol split combustion compared to HCCI and PPCI – mass fraction burned (MFB), maximum pressure rise rate (PRRmax), indicated efficiency (Ind. Eff.), and combustion efficiency (Comb. Eff.).

Fig. 19. n-Butanol split combustion at high engine loads – two-stage and threestage cylinder pressure, heat release rate, and injection command.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

15 10 5 0

Ind. Eff. [%]

100 96

Two-stage DI Two-stage PFI+DI Three-stage PFI+DI Three-stage DI

50

92

Comb. Eff. [%]

PRRmax [bar/°CA]

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

45 40 9

11

13

15

17

19

21

4. Conclusions

IMEP [bar] Fig. 20. n-Butanol split combustion – maximum pressure rise rate (PRRmax), indicated efficiency (Ind. Eff.), and combustion efficiency (Comb. Eff.) at mid-high engine loads.

Diesel

300 275 250 225 200 175 150

350 300 250 200 150

670 650 630 610 590 570 550 3

5

7

9 11 IMEP [bar]

13

ISFC_vol [L/kWh]

50 48 46 44 42 40 38

n-Butanol

Ind. Eff. [%]

value) compared with gasoline and diesel fuels. In Fig. 21, the indicated thermal efficiency, indicated specific fuel consumption (ISFC), and indicated specific CO2 emission are compared across an engine load sweep between neat n-butanol combustion and diesel conventional combustion on the same test engine. The test results indicate that neat n-butanol combustion is capable of producing comparable engine efficiency as diesel, with slightly lower efficiency at mid-high loads. However, the specific fuel consumption of n-butanol is much higher on either mass or volume basis, which means vehicles operating on n-butanol instead of diesel require larger fuel capacity in order to deliver the same tonmiles of freight, or the customers need to fill the tank more frequently.

ISFC [g/kWh]

When comparing two different fuels, however, higher fuel consumption does not necessarily equate to more CO2 emission in the engine exhaust. In point of fact, the carbon content by mass is 64.9% for n-butanol (C4H9OH) compared with 86.4% for a common diesel fuel (CH1.89), and the carbon content on energy basis is 19.6 mg/kJ and 20.1 mg/kJ respectively. As a result, when the two fuels produce the same engine efficiency, n-butanol combustion emits relatively less specific CO2 (ISCO2) that is converted from the fuel-borne carbon. This is considered as the Tank-to-Wheel benefits in CO2 mitigation by replacing diesel with n-butanol, in addition to the well-known advantages of biofuels over petroleum during Well-to-Tank processes.

35 7

ISCO2 [g/kWh]

11

15

Fig. 21. n-Butanol combustion vs. diesel conventional combustion – indicated efficiency (Ind. Eff.), indicated specific fuel consumption (ISFC), indicated specific volumetric fuel consumption (ISFC_vol), and indicated specific CO2 emission (ISCO2).

A comprehensive experimental study has been carried out to assess the feasibility of applying n-butanol as a next generation biofuel to replace diesel in compression ignition engines. The fuel property departure results in different air-fuel mixing processes and distinct combustion characteristics. These technical challenges are evaluated and addressed through empirical investigations on the engine performance, exhaust emissions, and combustion control strategies. The engine tests in this study have demonstrated the possibility of replacing diesel with neat n-butanol across a wide engine load range. The research findings and conclusions are summarized as follows: 1. In conventional diesel combustion, part of the fuel-air mixing usually occurs concurrently with combustion, forming the traditional heterogeneous diffusion burning process. In stark contrast, n-butanol tends to burn as a premixed cylinder charge that is formed prior to the onset of combustion, because the greater ignition resistance of n-butanol allows longer mixing time and the higher volatility facilitates faster atomization and mixing. 2. Two types of premixed combustion, PPCI and HCCI enabled with direct and port injections respectively, are found to be feasible to run neat n-butanol on the diesel test engine. n-Butanol’s distinctive fuel properties are well utilized in these two combustion strategies to create an ultra-lean and highlyhomogeneous cylinder charge, thereby enabling engine operations in low temperature combustion. As shown by the test results, both PPCI and HCCI of n-butanol are capable of producing low NOx and near-zero smoke emissions while achieving diesel-like engine efficiencies. 3. Similar to most combustion strategies of burning a premixed cylinder charge, n-butanol PPCI and HCCI are achieved only in the low-mid engine load range; higher load operations are prohibited by excessive pressure rise rates. Test results suggest that the systematic modulation of air dilution and fuel injection helps in controlling the combustion phasing and mitigating the pressure rise rates, yet fairly short to enable high load operations. 4. The split-combustion strategy, which employs 2–3 combustion events per engine cycle via multi-event butanol injections, is effective to control the rate of n-butanol combustion, thereby enabling neat n-butanol application across the engine load range up to the rated engine power density as when fuelled with conventional diesel. 5. Compared with diesel, neat n-butanol combustion delivers comparable indicated engine efficiency and emits similar specific CO2 across a wide engine load range, despite the noticeably higher specific fuel consumption due to n-butanol’s lower energy density.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

12

X. Han et al. / Applied Energy xxx (2016) xxx–xxx

Acknowledgements The research in the Clean Combustion Engine Laboratory at the University of Windsor is sponsored by the BioFuelNet, NSERC-CRD, CFI, OIT, AUTO21, NSERC-DG, NSERC-RTI, the Canada Research Chair program, the University of Windsor, and the automotive OEMs.

References [1] Demirbas A. Political, economic and environmental impacts of biofuels: a review. Appl Energy 2009;86(S1):S108–17. [2] International Energy Agency. 2015 Key world energy statistics; 2015. p. 6–8. [3] Environmental Protection Agency. Renewable fuel standard program: standards for 2014, 2015, and 2016 and biomass-based diesel volume for 2017, US; 2015. [4] Joseph H. Flex fuel vehicles in Brazil, Global Bioenergy Partnership Week; 2013. [5] Mofijur M, Rasul M, Hyde J, et al. Role of biofuels on IC engines emission reduction. Energy Proc 2015;75:886–92. [6] Cherubini F, Bird ND, Cowie A, et al. Energy- and greenhouse gas-based LCA of biofuel and bioenergy systems: key issues, ranges and recommendations. Resour Conserv Recycl 2009;53(8):434–47. [7] Koponen K, Soimakallio S, Tsupari E, et al. GHG emission performance of various liquid transportation biofuels in Finland in accordance with the EU sustainability criteria. Appl Energy 2013;102:440–8. [8] Hennecke A, Faist M, Reinhardt J, et al. Biofuel greenhouse gas calculations under the European Renewable Energy Directive – a comparison of the BioGrace tool vs. the tool of the roundtable on sustainable biofuels. Appl Energy 2013;102:55–62. [9] Nagy K, Körmendi K. Use of renewable energy sources in light of the ‘‘New Energy Strategy for Europe 2011–2020”. Appl Energy 2012;96:393–9. [10] Cho H, Kim J, Ahmed F, et al. Life-cycle greenhouse gas emissions and energy balances of a biodiesel production from palm fatty acid distillate (PFAD). Appl Energy 2013;111:479–88. [11] Tilman D, Socolow R, Foley JA, et al. Beneficial biofuels — the food, energy, and environment trilemma. Science 2009;325(5938):270–1. [12] Fairley P. Introduction: next generation biofuels. Nature 2011;474:S2–5. [13] Eijck J, Batidzirai B, Faaij A. Current and future economic performance of first and second generation biofuels in developing countries. Appl Energy 2014;135:115–41. [14] Heyne S, Harvey S. Assessment of the energy and economic performance of second generation biofuel production processes using energy market scenarios. Appl Energy 2013;101:203–12. [15] Sims RE, Mabee W, Saddler JN, et al. An overview of second generation biofuel technologies. Bioresour Technol 2010;101(6):1570–80. [16] Naik SN, Goudb VV, Rout PK, et al. Production of first and second generation biofuels: a comprehensive review. Renew Sustain Energy Rev 2010;14 (2):578–97. [17] Green E. Fermentative production of butanol—the industrial perspective. Curr Opin Biotechnol 2011;22(3):337–43. [18] Pereira L, Dias M, Mariano A, et al. Economic and environmental assessment of n-butanol production in an integrated first and second generation sugarcane biorefinery: fermentative versus catalytic routes. Appl Energy 2015;160:120–31. [19] Daroch M, Geng S, Wang G. Recent advances in liquid biofuel production from algal feedstocks. Appl Energy 2013;102:1371–81. [20] Jin C, Yao M, Liu H, et al. Progress in the production and application of nbutanol as a biofuel. Renew Sustain Energy Rev 2011;15:4080–106. [21] Nigam PS, Singh A. Production of liquid biofuels from renewable resources. Prog Energy Combust Sci 2011;37:52–68. [22] Kumar M, Gayen K. Developments in biobutanol production: new insights. Appl Energy 2011;88:1999–2012. [23] García V, Päkkiläa J, Ojamo H, et al. Challenges in biobutanol production: how to improve the efficiency? Renew Sustain Energy Rev 2011;15:964–80. [24] Ezeji T, Qureshi N, Blaschek HP. Butanol production from agricultural residues: impact of degradation products on Clostridium beijerinckii growth and butanol fermentation. Biotechnol Bioeng 2007;97(6):1460–9. [25] Deng B, Yang J, Zhang D, et al. The challenges and strategies of butanol application in conventional engines: the sensitivity study of ignition and valve timing. Appl Energy 2013;108:248–60. [26] Wei H, Feng D, Pan M, et al. Experimental investigation on the knocking combustion characteristics of n-butanol gasoline blends in a DISI engine. Appl Energy 2016;175:346–55. [27] Irimescu A. Performance and fuel conversion efficiency of a spark ignition engine fueled with iso-butanol. Appl Energy 2012;96:477–83. [28] Alasfour FN. NOx emission from a spark-ignition engine using 30% iso-butanol – gasoline blend: part 1: preheating inlet air. Appl Therm Eng 1998;18 (5):245–56. [29] Alasfour FN. NOx emission from a spark-ignition engine using 30% iso-butanol – gasoline blend: part 2: ignition timing. Appl Therm Eng 1998;18(8):609–18.

[30] Williams J, Goodfellow C, Lance D, et al. Impact of butanol and other biocomponents on the thermal efficiency of prototype and conventional engines. SAE technical paper 2009-01-1908; 2009. [31] Regalbuto C, Pennisi M, Wigg B, et al. Experimental investigation of butanol isomer combustion in spark ignition engines. SAE technical paper 2012-011271; 2012. [32] Niass T, Amer A, Xu W, et al. Butanol blending – a promising approach to enhance the thermodynamic potential of gasoline – part 1. SAE Int J Fuels Lubr 2012;5(1):265–73. [33] Broustail G, Halter F, Seers P, et al. Comparison of regulated and non-regulated pollutants with iso-octane/butanol and iso-octane/ethanol blends in a portfuel injection spark-ignition engine. Fuel 2012;94:251–61. [34] Georgios K, Daniel S, Diep V, et al. Evaluating the regulated emissions, air toxics, ultrafine particles, and black carbon from SI-PFI and SI-DI vehicles operating on different ethanol and iso-butanol blends. Fuel 2014;128:410–21. [35] Venugopal T, Ramesh A. Effective utilization of butanol along with gasoline in a spark ignition engine through a dual injection system. Appl Therm Eng 2013;59(1):550–8. [36] Venugopal T, Ramesh A. Experimental studies on the effect of injection timing in a SI engine using dual injection of n-butanol and gasoline in the intake port. Fuel 2014;115:295–305. [37] Szwaja A, Naber J. Combustion of n-butanol in a spark-ignition IC engine. Fuel 2010;89:1573–82. [38] Wigg B, Coverdill R, Lee C, et al. Emissions characteristics of neat butanol fuel using a port fuel-injected, spark-ignition engine. SAE technical paper 2011-010902; 2011. [39] Huang H, Zhou C, Liu Q, et al. An experimental study on the combustion and emission characteristics of a diesel engine under low temperature combustion of diesel/gasoline/n-butanol blends. Appl Energy 2016;170:219–31. [40] Choi B, Jiang X, Gilsung K, et al. Effect of diesel fuel blend with n-butanol on the emission of a turbocharged common rail direct injection diesel engine. Appl Energy 2015;146:20–8. [41] Zhang Z, Balasubramanian R. Influence of butanol addition to diesel–biodiesel blend on engine performance and particulate emissions of a stationary diesel engine. Appl Energy 2014;119:530–6. [42] Rakopoulos C, Dimaratos A, Giakoumis E, et al. Investigating the emissions during acceleration of a turbocharged diesel engine operating with bio-diesel or n-butanol diesel fuel blends. Energy 2010;35:5173–84. [43] Ballesteros R, Hernandez JJ, Guillen-Flores J. Carbonyls speciation in a typical European automotive diesel engine using bioethanol/butanol-diesel blends. Fuel 2012;95:136–45. [44] Valentino G, Corcione FE, Iannuzzi SE, et al. Experimental study on performance and emissions of a high speed diesel engine fueled with nbutanol diesel blends under premixed low temperature combustion. Fuel 2012;92:295–307. [45] Merola SS, Tornatore C, Lannuzzi SE, et al. Combustion process investigation in a high speed diesel engine fueled with n-butanol diesel blend by conventional methods and optical diagnostics. Renew Energy 2014;64:225–37. [46] Siwale L, Kristof L, Adam T, et al. Combustion and emission characteristics of nbutanol/diesel fuel blend in a turbo-charged compression ignition engine. Fuel 2013;107:409–18. [47] Zou X, Wang H, Zheng Z, et al. Numerical study of the RCCI combustion processes fuelled with methanol, ethanol, n-butanol and diesel. SAE technical paper 2016-01-0777; 2016. [48] Soloiu V, Muinos M, Harp S. Investigation of dual fuel PCCI (PFI of n-butanol and DI-ULSD) compared with DI of binary mixtures of the same fuels in an omnivorous diesel engine. SAE technical paper 2015-01-0857; 2015. [49] Han X, Zheng M, Tjong J, et al. Suitability study of n-butanol for enabling PCCI and HCCI and RCCI combustion on a high compression-ratio diesel engine. SAE technical paper 2015-01-1816; 2015. [50] Wang X, Liu H, Zheng Z, et al. Development of a reduced n-butanol/biodiesel mechanism for a dual fuel engine. Fuel 2015;157:87–96. [51] Soloiu V, Rivero-Castillo A, Muinos M, et al. Simultaneous reduction of NOX and soot in a diesel engine through RCCI operation with PFI of n-butanol and DI of cottonseed biodiesel. SAE technical paper 2014-01-1322; 2014. [52] Han W, Karthik N, Timothy L, et al. Spray and combustion characteristics of neat acetone-butanol-ethanol, n-butanol, and diesel in a constant volume chamber. Energy Fuels 2014;28(10):6380–91. [53] Li H, Lee C, Huang M, et al. Combustion characteristics and soot distributions of neat butanol and neat soybean biodiesel. Energy Fuels 2011;25(7):3192–203. [54] Gao T, Xie K, Yu S, et al. Characterization of n-butanol high pressure injection from modern common rail injection system. In: ASME 2015 internal combustion engine division fall technical conference, Paper No. ICEF20151129; 2015. [55] Zheng M, Li T, Han X. Direct injection of neat n-butanol for enabling clean low temperature combustion in a modern diesel engine. Fuel 2015;142:28–37. [56] Yanai T, Dev S, Han X, et al. Impact of fuelling techniques on neat n-butanol combustion and emissions in a compression ignition engine. SAE Int J Eng 2015;8(2):735–46. [57] Han X, Wang M, Zheng, M. An enabling study of neat n-butanol HCCI combustion on a high compression-ratio diesel engine. SAE technical paper 2015-01-0001; 2015. [58] Han X. Study of fuels and fuelling strategies for enabling clean combustion in compression ignition engines Electronic Theses and Dissertations. University of Windsor; 2014.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059

X. Han et al. / Applied Energy xxx (2016) xxx–xxx [59] Yaws CL. Yaws’ transport properties of chemicals and hydrocarbons. Oxford: Gulf Publishing Company; 2014. [60] Totten GE, Liang H. Surface modification and mechanisms: friction, stress, and reaction engineering. CRC Press; 2004. [61] Wadumesthrige K, Ng KS, Salley SO. Properties of butanol-biodiesel-ULSD ternary mixtures. SAE Int J Fuels Lubr 2010;3(2):660–70.

13

[62] Brownstein AM. Renewable motor fuels – the past, the present and the uncertain future. Boston: Butterworth-Heinemann; 2015. [63] Han X, Xie K, Zheng M, et al. Ignition control of gasoline-diesel dual fuel combustion. SAE technical paper 2012-01-1972; 2012. [64] Zheng M, Tan Y, Mulenga MC, et al. Thermal efficiency analyses of diesel low temperature combustion cycles. SAE technical paper 2007-01-4019; 2007.

Please cite this article in press as: Han X et al. Clean combustion of n-butanol as a next generation biofuel for diesel engines. Appl Energy (2016), http://dx. doi.org/10.1016/j.apenergy.2016.12.059