Reducing Cold Start Emissions from Automotive Diesel Engine at Cold Ambient Temperatures

Cold start performance of diesel engines is determined by engine design, fuel type, fuel injection strategies, lubricant and ambient temperature conditions. Prevailing emissions legislation regarding low temperature emission tests applicable for gasoline vehicles is likely to be implemented to diesel vehicles. The present research work investigates the effects of intake air heating on a Euro 5 diesel engine’s performance and exhaust emission (gaseous and particulate emissions) characteristics during the cold start followed by idling at cold ambient conditions. Heating of intake air entering the engine at cold ambient temperature conditions improved fuel combustion as well as reduced the cranking period and improved the fuel economy. More than 50% reduction in HC and 17% reduction in NOx emissions were achieved by intake air heating. Number count of accumulation mode particulates were higher during cold start compared to idle operation for all the temperature conditions. Intake air heating decreased the particulate number and size that led to reduction in total particulate mass by higher than 50% and 75% during cold start and idle respectively. The intake air heating strategy improved the cold start performance of the diesel engine at cold ambient temperature conditions and thereby would reduce the overall driving cycle emissions.


INTRODUCTION
Diesel engines are the versatile power source and widely used in automotive and power sectors.Advanced engine technologies have been developed to meet the ever stringent automotive exhaust emission norms.An approach to meet the latest emission norms is reducing the exhaust emission generated in combustion and/or optimizing the performance of exhaust after treatment devices.Passenger car vehicles are evaluated for exhaust emissions on regulatory emissions driving cycles such as New European Driving Cycle (NEDC) and Federal Test Procedure (FTP) on chassis dynamometer test benches.Cold start emissions at normal ambient temperatures are several orders higher than that of at warm start.The emission reduction performance of exhaust aftertreatment devices is also poor at start conditions, as the catalyst would not reach the light-off temperature even at normal ambient temperatures as well.Effective year 2000 (Euro 3), the NEDC test procedure was modified to eliminate the first 40 sec engine warm-up period prior to beginning of emission sampling (dieselnet, 2014).These cold start emissions account for the major portion of the total cycle emissions.Hence, investigations on the cold start emissions become a major thrust area of research for the automotive industry, oil industry, environmentalist and academicians.
Diesel fuel combustion is dependent on engine geometry, fuel properties, compression temperature of fuel-air mixture, fuel injection strategies and ambient temperature conditions.Fuel injection at lower ambient temperatures adversely affects the fuel atomization, retards heating of the fuel droplets and its evaporation which make it difficulties in starting of the engine (Chris et al., 2013).Stanton et al. (1998) reported that a large portion of the injected fuel during the cold start, which was not burned that adhered to the cold surfaces and made difficult to vaporize and increased the cold start combustion instability with the drop in the ambient temperatures (Stanton, 1998;Dardiotis et al., 2013).To meet the stringent emission norms particularly, with nitrogen oxides (NO x ) emissions require exhaust after treatment devices such as lean NO x traps or Selective Catalytic Reduction (SCR).Recirculation of exhaust gases is reduced or deactivated in diesel engines at low ambient temperature conditions in order to avoid water condensation.Currently, the low temperature emissions tests (Type VI: at -7C ambient temperature) are mandatory only for gasoline vehicles and extension of this test to diesel vehicles is under consideration (Hara, 1999).
Cold start performance of diesel engines was improved by a high cetane number and high volatility fuels at cold ambient conditions (Hara, 1999;Startck, 2010).The cold start performance of diesel engines is severe when the ambient temperature is reduced to sub-zero.The nominal start time of an engine is typically 1 or 2 sec when the engine is at warm ambient conditions, whereas the start time is greater than 10 sec at sub-zero temperatures (Mann, 1999;Brown et al., 2007).Total hydro carbon (THC) and particulate matter (PM) emissions of the engine during the first 3 minutes of cold start and idle operation was several times greater than that of warm start (Bielaczyc et al., 2001).
Particulate emissions are formed by the condensation of the oxidized and/or pyrolysis products of fuel molecules.Lighter fractions of fuel change its state from atomized liquid aerosol to gas and cooling the air-fuel mixture in the combustion chamber had a contributory effect in preventing the combustion.The exhaust particulates are composed of nucleation and accumulation mode particulates.The nucleation mode particulates composed of soluble or volatile organic fractions (SOF/VOF) which are formed mainly from exhaust dilution and cooling processes by the small amount of fuel or evaporated lubricating oil which escapes the oxidation process.The accumulation mode (larger size) particulates are formed by agglomeration of many fine particles and semi volatiles absorbed on the soot particles (Blackwood, 1998;Peckham, 2011).Euro 4 emissions norms for diesel engines mandate the particulate matter emission measurement by mass only.Euro 5b emission norms onwards mandate the measurement of particle number (PN) in addition to the mass based measurement to limit the ultra fine particles (Oberdorster, 2005;European Commission, 2008, Andrea et al., 2011).This is highly influenced by the particulate size distribution e.g., smaller particulates are more harmful to human health (due to their larger surface to volume ratio) however they do not adversely impact visibility (Srivatsava, 2011;Agarwal, 2013).The diesel particulate emissions are responsible for poor visibility, soiling of buildings and adverse health effects on humans, livestock, etc.
Glow plugs, intake air heater and block heater are the potential cold start aids to reduce the exhaust emissions.Glow plugs are suitable for smaller size engines whereas intake air heaters are suitable for larger size engines.Preheating of intake air increased vaporization of the fuel and reduced CO and THC emissions (Yilmaz, 2012).Broatch et al. (2008) conducted Motor Vehicle Emission Group (MVEG) tests to study the effect of air heating by using a glow plug and intake air heater.They highlighted that the intake air heating has a direct and positive impact on HC and CO emissions and this technology needed to reduce EGR rates for controlling the emissions at the initial instants of the urban driving cycle (UDC) cycle.also It has been also reported that air heating technology with electrical resistances reduced the CO, HC, NOx emissions and increased the particulate emissions.
Although a limited number of investigations on the cold starting of a single cylinder or multi cylinder diesel engines at cold ambient temperature conditions have been published, a very few investigations only reported the effect of the cold start aids on cold start emissions (including particulate emissions) from multi-cylinder diesel engines at sub-zero temperatures.The characterization of exhaust particulates in terms of particle number and size during the cold start at very cold ambient conditions is significant.
The objective of this present research is to investigate the effect of intake air heating on the cold start and idle emissions from a common rail direct injection (CRDI) diesel engine at very cold ambient temperatures.

EXPERIMENTAL SETUP
The cold cell transient dynamometer-engine test facility is designed to operate conventional engine testing at normal ambient conditions, low temperature legislative emissions cycles, and performing the cold start studies down to -20°C.The experimental setup and the test equipment used for this study (Fig. 1) are described as follows.

Engine and Dynamometer
The engine used in the study is a 3.0 L six cylinder, turbo charged common rail direct injection (CRDI) diesel engine for the passenger cars and the specification of the engine is given in Table 1.The test bench includes a DynoDur 290 dynamometer, which is an AC machine capable of full four-quadrant operation.Torque measurement is made with a torque flange.An AVL Puma Open 5.1 test bed automation system controls the dynamometer, fluid controlling systems and emission measuring equipment which is interfaced with the engine test bed.The test cycles are programmed in the Puma for the automatic operation of test run and an Open ECU is used to control the engine.The engine ECU data was recorded by the PUMA control system through ETAS INCA software.

Climatic Control System
The climatic enclosure is constructed with a rigid insulated floor plate, which supports an extruded aluminum frame.Modular insulated panels, made from glass fiber cloth with a sandwich of elastomeric foam, can be clipped onto the frame and secured together with Velcro strips.Fluid services pipes and instrumentation cabling brought in and out of the enclosure between the panels.

Intake Air Heating System
The combustion air supply system is located in the plant room adjacent to the test cell and consists of an AVL ACS1600 unit and an air dryer to reduce the humidity of the incoming air.The system has capacity to supply chilled air up to 500 m 3 h -1 for below 0°C testing and up to 1600 m 3 h -1 for ambient testing.In diesel engines, fuel-air mixture reaches the temperature of about 400°C at the end of compression stroke, which is the minimum temperature of the fuel ignition during cold start phase of the engine when the temperature of intake air is at 20°C.An intake air heater was installed in between intake manifold and turbocharger.Important parameters considered for calculating the energy required for an intake air heater are total intake air mass flow rate during the cold start, thermal power dissipated in the air and temperature of the supply air to the heater.The duration and power required to achieve the desired intake  2993 cc Maximum power@ rpm (kW) 199 kW@4000 RPM Maximum torque@ rpm (Nm) 600 Nm@2000 RPM air temperatures by the air heater was controlled by an external variable voltage source supplied to the air heater.

Coolant Conditioning System
This unit is equipped with two cooling circuits and is capable of conditioning the engine coolant at normal ambient temperatures during conventional testing and also circulating coolant at temperatures down to -20°C through a stationary engine to perform a low temperature soak prior to an emissions test or a cold start.

Oil Conditioning System
This unit is equipped with two cooling circuits and is capable of conditioning the engine oil at normal ambient temperatures (90 to140°C) during conventional testing and also circulating oil at temperatures down to -20°C through a stationary engine to perform a low temperature soak prior to an emissions test or a cold start.The oil is circulated in and out of the oil pan and the oil pump therefore needs to run continuously, even when the engine is not running.To cope with the large viscosity difference between oil at -20 and +140°C, the pump is inverter controlled to run at a speed proportional to oil temperature.

Fuel Measurement and Conditioning System
Fuel metering is performed by the AVL735S, which utilizes the Coriolis principle to measure mass flow of fuel consumed by the engine.AVL 753C along with the Cold fuel booster unit condition the temperature of the fuel upto -20°C.The fuel flow rate from the 753C/Cold boost unit is approximately four times the maximum fuel consumption rate, so most of the fuel flows through the back-pressure regulator and returns to the conditioning system.The high proportion of re-circulated fuel ensures a stable supply temperature.The Puma system monitors a pressure transducer on the return line and adjusts the fuel pump speed to maintain a steady backpressure for fuel leaving the engine.

Emission Measurement
The AVL AMA i60 is an integrated exhaust gas analyzer is used for gaseous emission measurement.The instantaneous particulate emission was measured by the Cambustion Differential Mobility Spectrometer (DMS500).The sample probe is installed ahead of the after treatment devices and the sample probe is thermally insulated to avoid condensation of particulates during measurement in the cold chamber.The sampling probe is connected to the electrically heated sample line.The secondary dilution ratio was set at 250:1 for the cold start studies.The signal from the starter was taken as an analogue input channel into the DMS500 to synchronize the test starting time.

Test Procedure
The engine is soaked at the desired test temperature (-7C environment) for 8 hours in the cold cell and ensured that air, coolant, oil and fuel temperatures are maintained before start of the test.The intake air heater is preheated (~40 sec) to achieve the desired intake air temperature and the engine was started.The heater is further allowed to operate for 20 sec after starting of the engine.For this cold start study, the engine run at idle speed for 3 min after starting and the exhaust emissions were measured at the upstream of the after treatment devices.The cold start period includes the cranking period and the acceleration period to reach peak engine speed.After completion of each test, the engine is cranked without fuel to remove the any residual gases in the engine.For the analysis of the test results, cold start period is considered as 10 seconds from key-on and idle period is 15 seconds from the end of the cold start.The analysis of the cold start behavior of the engine at different ambient conditions presented in this article is within the European Commission (EC) research program investigating the diesel engine cold start ability and transient operation (DECOST).

RESULTS AND DISCUSSION
Engine cold start conducted at different intake air temperatures in the -7°C environment and the test results are discussed as follows.

Engine Performance
The cold start performance of the engine is analyzed in terms of its speed and fuel consumption.The cranking period is the time taken to initiate the combustion of air-fuel mixture from cranking i.e., ignition key-on.Engine speed increased rapidly during initial phases of fuel combustion and then it is dropped down to the idle speed.Figs.2(a) and 2(b) shows a plot for the engine speed and fuel consumption with respect to time at the different intake air temperatures in -7°C environment.
Longer cranking period, higher fuel injection quantity, poor lubrication, partial fuel evaporation and improper combustion conditions are the crucial problems for the cold start.While starting of the engine more quantity of fuel was injected to initiate the combustion, and to overcome the higher friction resistance offered by the engine components and lower combustion chamber temperature at the low ambient temperatures.After several revolution of the crankshaft, the compression temperature and pressure inside the cylinder increased and the accumulated fuel burned abruptly that caused the rapid rise in engine speed to reach a peak value.Then the engine speed dropped down to maintain the idle speed and hence the fuel consumption was also reduced.It is observed that the higher intake air temperature significantly influenced the engine start performance and fuel consumption.The peak speed of the engine for the intake air temperature at 5°C and 15°C were 15% and 16% respectively, higher than that of at -7°C.This is due to higher frictional and pumping forces and incomplete combustion of fuel at very cold ambient temperatures.The warm intake air inducted into the engine helped in quicker vaporization of fuel and there by improved the combustion.Thus, the cranking period and fuel consumption of the engine was drastically decreased with the intake air heating.The total fuel consumption of the engine during the cold start period (for the first 10 sec) at the intake air temperatures of 5° and 15°C were 20% and 25% lower than that of intake air temperature at -7°C.

Gaseous Emissions NO x Emissions
Instantaneous and cumulative nitrogen oxides (NO x ) emissions from the diesel engine during the test conducted at -7°C environment is shown in Figs.3(a The NO x emission spikes were observed during the cold start for all the tests.This is due to more time taken by the engine to reach the fuel combustion stage at the intake air temperature of -7°C (Fig. 2(a)) and hence the accumulated fuel during this period burned suddenly and emitted higher NO x .The cold combustion air, higher amount of fuel injection and nil EGR at low ambient temperatures increased the NO x emissions significantly.In addition, high intake air mass flow for the low ambient scenario also boosted the oxygen availability inside the cylinder.Thus, the cold start NO x emissions reached a high peak value for the intake air temperature at-7°C compared with the intake air heating.In the case of intake air heating, lesser fuel accumulation during the cranking period and the instantaneous fuel combustion lowered the NO x emissions in comparison with the -7°C environment.Intake air temperature of 5°C reduced the NO x emissions by 17%, whereas 2% increase in NO x emission was observed at the intake temperature of 15°C which was due to the warmer in-cylinder conditions has improved the fuel combustion and hence increased the NO x emissions as well.A similar trend was observed during the idle condition for the all intake air temperature conditions also.

HC Emissions
Figs. pumping resistance offered by engine components at the lower ambient conditions.Because of the lower intake or compression air temperatures, the injected fuel undergoes slower evaporation.Lower wall temperatures enhanced the flame quenching and increased the HC emissions.Moreover, the over-lean mixture will not auto ignite, and it can be only oxidized by relative slow thermal-oxidation reactions that will be incomplete.All the above-mentioned factors results in poor fuel-air mixing and incomplete combustion which favored the formation of HC emissions.Preheated intake air to the engine helped in easier the evaporation of the fuel and thereby accelerated the pre-combustion chemical reactions.Thus the intake air heating helped in reduction in ignition delay of the fuel and thereby reduced the HC emissions.The preheated intake air of 5°C and 15°C was inducted into the engine during the cold start reduced the cold start phase emissions by 55% and 75% respectively in comparison with the intake air temperature of -7°C.Also, the idle emissions at 5°C and 15°C intake air temperatures was reduced by 60% and 80% respectively in comparison with the -7°C intake air temperature.

Particulate Emissions Particle Number
Fig. 5 shows exhaust particle number (PN) for the cold start and idle operation of the diesel engine at different intake air temperatures.It is observed that the particle number reached a peak value during the acceleration period of cold start and then dropped down during the idle.A higher quantity of fuel injection and lower in-cylinder temperature at cold ambient temperatures are likely to produce incomplete combustion, eventually leading to excessive particulate emissions.The PN was lower at the idle condition in comparison with that of cold start for the all intake air temperatures.The percentage reduction in the total particle number during the first 10 seconds of cold start was 3% and 10%, and during the next 15 seconds of idle operation was 10% and 30% for the intake air temperatures at 5°C and 15°C respectively compared with the intake air temperature of -7°C.Heated intake air inducted into the engine helped in vaporization of fuel and thereby improved the fuel combustion that led to a reduction in particulate emissions.

Size Spectral Density
Size spectral density (SSD) of exhaust particulates depends on the number and size of the particles.Figs.6(a) and 6(b) shows the SSD of exhaust particulates during the first 10 seconds of cold start operation at different intake air temperatures.The number count of the smaller diameter particles (< 23 nm) increased and bigger diameter particles (> 100 nm) decreased with the increase in intake air temperatures.The particles of diameter in the range 23-100 nm was also increased at the intake air temperature 5°C and then decreased at the intake air temperature 15°C.The lower intake air temperatures and delayed firing cycles caused incomplete combustion and formation of bigger size particles of diameter greater than 100 nm whereas preheated intake air improved the combustion thereby reduced the bigger size particulate emissions.For the 5°C and 15°C intake air temperatures, 35% and 60% reduction in bigger size particles (> 100 nm) was achieved in comparison with the -7°C, that translated in the number count of lesser diameter particulates.
The SSD of the exhaust particulates during the idle period of 15 sec after the cold start at different intake air temperature conditions is shown in Fig. 7.The exhaust particulates of diameter ranges between 10 nm and 100 nm were at higher in numbers for all the temperature conditions.The particulate formation is strongly influenced by the localized temperature distribution and fuel/air ratio which varies greatly inside the combustion chamber.Rise in temperature of combustion chamber during the idle increased the rate of oxidation of fuel-air mixture rapidly than the rate of soot formation and hence the diameter of the particles was decreased.Furthermore, reduced fuel injection quantity and preheated intake air supported the pre-combustion reactions and reduced the particle number concentration.The percentage reduction in the total particle number for the size greater than 100 nm is 80% and 85%; for the size range 23-100 nm is 8% and 35%, and 25% and 35% increase particle number of smaller size (< 23 nm) was achieved for the intake air temperatures of 5°C and 15°C respectively in comparison with the intake air temperature at -7°C.Thus, the intake air heating improved the combustion and thereby reduced the bigger diameter particulates significantly.

Count Mean Diameter
Count mean diameter (CMD) is the average diameter based on the unit number count of particulates.Fig. 8 shows the count mean diameter (CMD) of the particulates emitted during the cold start and idle.It is seen that diameter of the particulates (accumulation) was higher at lower intake air temperatures.A large amount of hydrocarbon was absorbed by accumulation particulates that led to a significant increase of diameter and mass concentration of particulates at low ambient temperatures.The intake air heating improved the fuel combustion and reduced the hydrocarbon and particulate emissions.This led to reduction in the number of accumulation mode particulates and consequently decreased the CMD.The CMD of particulates during cold start was decreased by 10% and 30% respectively; and during the idle was decreased by 20% and 40% respectively for the intake air temperatures of 5°C and 15°C respectively compared to that of at -7°C.

Particulate Mass
The total particulate mass (PM) for the cold start and followed idle operation of the diesel engine is depicted in Figs.9(a) and 9(b).It is observed that the exhaust particulate mass was higher at -7C in comparison with the intake air heating at 5C and 15C.This is due to the intake air heating which improved the fuel combustion and reduced the accumulation mode particulates.The exhaust particulates of diameter 23-100 nm were larger in numbers for all the intake air temperatures at -7°C environment.It is seen from the Fig. 6(b) that bigger diameter particulates was decreased by the intake air heating and hence the overall mass of the particulates was also reduced.The cold start particulate mass at -7°C was decreased by 50% and 60% respectively for the intake air temperature of 5°C and 15°C.Fig. 9(b) depicts that the particulate mass at the idle condition was much lower as compared to the cold start.This was due to the warming up of the engine improved the combustion of fuel and hence reduced the size of particulates that leads to reduction in particulate mass.During the idle, particulate mass was reduced by 70% and 85% for the intake air temperatures of 5°C and 15°C in comparison with -7°C.In summary, increase in the intake air temperature reduced the particulate mass significantly during the cold start and idle.

CONCLUSIONS
An intake air heating strategy was implemented to reduce cold start emissions from a latest generation Euro V diesel engine at cold ambient conditions and to study the effect of the increasing the intake air temperature from -7°C to 5°C and 15°C on the engine performance, gaseous and particulate emissions.The following conclusions can be made based on the analysis of test results.
• Intake air heating reduced the cranking period of the engine and improved the fuel economy during the cold start.
• Intake air temperature of 5°C reduced the NO x emissions by 17% during cold start and idle, and higher intake air temperature increased the NO x emissions marginally.More than 50% reduction in the HC emissions was achieved during the cold start and idle at the intake air temperature of 5°C and further increase in the intake air temperature reduced the HC emissions significantly.• Accumulation mode particulates (> 100 nm) were higher in number for the cold start compared to the idle operation which was due to incomplete combustion of fuel at the cold ambient temperatures.• Particulates of diameter in the range of 23-100 nm were larger in number in the total particulates spectrum for all the intake air temperature conditions.• Intake air heating reduced the bigger size particulates and increased the lesser diameter particulates during cold start and idle.CMD of particulates was reduced remarkably by intake air heating during cold start and idle.• The reduction in particle number and size during the cold start and idle by intake air heating led to reduction in the total particulate mass also.The intake air heating strategies reduced the particulate mass by more than 50% and 75% during the cold start and idle respectively.In summary, the preheating of intake air in the cold ambient temperature conditions improved the combustion and there by reduced the gaseous and particulate emissions from the diesel engine significantly.

ACKNOWLEDGEMENT
First author express his thanks to the European Commission for sponsoring Marie Curie International Incoming Fellowship to carry out the DECOST project under FP7 framework in the Future Engines and Fuels Lab at the University of Birmingham.Authors acknowledge the support of the European Regional Development Fund and Advantage West Midland for the cold cell test facility.The authors would also like to thank Jaguar Land Rover and Shell Global Solutions for their support in progress of the project work.First author also thank the management of Indian Oil Corporation Limited, R&D Centre for their permission to pursue his post-doctoral research.

Fig. 2 .
Fig. 2. Engine parameters during cold start at different intake air temperatures (a) Speed (b) Fuel consumption.

Fig. 4 .
Fig. 3. NO x emissions at different intake air temperatures (a) instantaneous NO x (b) Cumulative NO x .

Fig. 6 .
Fig. 6.Particle size distribution during cold start at different intake air temperatures (a) instantaneous (b) cumulative.

Fig. 8 .
Fig. 8. Count mean diameter of particles at different ambient temperatures.