- Email: [email protected]
An assessment of regulated emissions and CO2 emissions from a European light-duty CNG-fueled vehicle in the context of Euro 6 emissions regulations
Applied Energy 117 (2014) 134–141
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
An assessment of regulated emissions and CO2 emissions from a European light-duty CNG-fueled vehicle in the context of Euro 6 emissions regulations Piotr Bielaczyc ⇑, Joseph Woodburn, Andrzej Szczotka BOSMAL Automotive Research and Development Institute Ltd, Sarni Stok 93, Bielsko-Biala 43-300, Poland
h i g h l i g h t s Compressed natural gas is a promising fuel type for light-duty vehicles. Euro 5 and 6 emissions standards will reduce emissions to very low levels. A bi-fuel Euro 5 vehicle (CNG/gasoline) was tested on a chassis dynamometer. When operating on CNG the vehicle easily met Euro 6 limits. Carbon dioxide emissions were 24–25% lower when running on CNG.
a r t i c l e
i n f o
Article history: Received 26 February 2013 Received in revised form 28 October 2013 Accepted 4 December 2013 Available online 27 December 2013 Keywords: Passenger car Alternative fuel Compressed natural gas Exhaust emission Euro 6
a b s t r a c t Natural gas is one of the most promising alternative fuels to meet the upcoming stringent Euro 6 emissions regulations in the European Union, as well as the planned reductions in CO2 emissions. For sparkignition engines, bi-fuel fuelling equipment is widely available and engine conversion technology for European automobiles is well established, thereby facilitating usage of natural gas in its compressed form (CNG). In light of the promising characteristics and increasing usage of natural gas as a vehicular fuel, this study investigates emissions from a passenger car featuring a spark-ignition engine capable of running on both CNG and standard gasoline. Results from emissions testing of the vehicle on a chassis dynamometer are presented and discussed in the context of the Euro 6 emissions requirements. The test vehicle featured a multipoint gas injection system and was an unmodified, commercially available European vehicle meeting the Euro 5 standard. The results indicated that when fueled with CNG, such a vehicle can comfortably meet Euro 6 emissions limits, with certain differences observed in the emissions according to the fuel type used. Furthermore, when running on CNG the vehicle was observed to emit considerably less carbon dioxide than when fueled with gasoline, with the reduction closely agreeing with the results of other studies. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Worldwide population growth and industrialization have resulted in increases in demand for energy in the transportation Abbreviations: CH4, methane; CARB, California Air Resources Board; CNG, compressed natural gas; CO, carbon monoxide; CO2, carbon dioxide; CVS, constant volume sampling; EC, European commission; ECE, Economic Commission for Europe; ECU, electronic control unit; EU, European Union; EUDC, extra urban driving cycle; GHG, greenhouse gases; HC, hydrocarbons; mpg, miles per gallon; MPI, multipoint [fuel] injection; N2O, nitrous oxide; NEDC, New European Driving Cycle; NMHC, non-methane hydrocarbons, HC–CH4; NOx, oxides of nitrogen; RON, research octane number; SI, spark ignition; TWC, three-way catalyst; UDC, urban driving cycle. ⇑ Corresponding author. Tel.: +48 338130598. E-mail address: [email protected] (P. Bielaczyc). 0306-2619/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2013.12.003
sector, among other sectors. As a result, air pollution and anthropogenic greenhouse gas emissions have become key global problems. Various options are available to militate against these effects, one of which is making use of fuel types with more favorable emissions characteristics. The use of alternative fuels, mainly biodiesel, gasoline–alcohol blends, natural gas and liquefied gasolinium gas in vehicular applications has grown in recent years in European Union countries, the United States, Japan, India, Brazil and many other markets. Legislation is also in place to encourage or effectively force further adoption of these fuel types (e.g. [1]). European Union requirements regarding vehicle emissions for passenger cars and light commercial vehicles were introduced in 2009 for type approval and in 2011 for all new types, specified as Euro 5, with further requirements (Euro 6) planned from 2014
135
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
2,5 2,3
* PM - vehicles with direct injection
CO THC NMHC
2,0
Emission [g/km]
onwards. These regulations set limits for emissions of HC, CO, and NOx; CO2 emissions are covered by separate legislation. While harmful emissions are generally of relatively little direct consequence to a vehicle’s owner/operator, CO2 emissions are inherently linked to fuel consumption, which is of great importance to the owner. Two recent reports have further underlined the discrepancy between declared (i.e. type approval) CO2 emissions and the real values actually observed during vehicle usage [2,3]. In light of ongoing discussion on these discrepancies, fuel types with well-established benefits regarding CO2 emissions and regulated emissions are understandably of great interest to legislators. Such factors are less important to the average vehicle owner, but CNG has the added advantage of potentially lowering refueling costs for light-duty vehicles (compared to gasoline), based on data on pricing submitted by consumers and retailers [4,5]. Pricing data from the USA also indicates the potential for substantial reductions in refueling costs in that country, when running on CNG [6]. It is in this context that usage of natural gas as an automotive fuel seems bound to increase over the next few years – indeed, demand for natural gas for use in European road vehicles is projected to double between 2015 and 2020 [7]. A recent techno–economic analysis of a broad range of fuel types revealed CNG to outperform gasoline in terms of fuel economy, with a vehicle purchase cost that was close to that of a gasoline vehicle [8]. As things stand, it appears likely that fossil fuels will remain the chief source of energy in the transportation sector for the foreseeable future, but the aforementioned factors indicate pressure to move towards increased usage of alternative fuels. Since energy density is an important factor for automotive fuels, gases are inherently at a disadvantage compared to liquid fuels (e.g. gasoline, bioethanol, and diesel), but this unfavorable characteristic can be partially offset by compression of natural gas, creating compressed natural gas (CNG). In light of these factors, CNG is currently the best alternative to conventional transport fuels. Various basic physical and technical parameters of CNG make it a very good fuel for turbocharged SI engines (discussed in Section 2). While the energy content of the fuel is somewhat lower, as an automotive fuel CNG generally has favorable drivability characteristics and has proven relatively popular with consumers in multiple markets. By the end of 2011 the global NGV fleet numbered more than 15 million vehicles, with around 2.5 million vehicles coming into use in 2011 alone [9]. Separate (but qualitatively similar) legislative moves taking place in the EU and the USA can also be considered to be drivers of increased interest in (and usage of) CNG as an automotive fuel. In the EU, the commission’s proposal to enforce a fleet average CO2 emissions limit of 95 g/km over the NEDC by 2020, together with ongoing discussions over aiming for a fleet average figure of 70 g/km by 2025. In the US, the target requiring carmakers to increase fuel economy in new vehicles sold between 2011 and 2025 to finally reach 54.5 mpg as CAFÉ (corporate average fuel economy) by 2025. The proposed CAFÉ rule grants incentives to plug-in electric and hybrid vehicles, with the final rule adding CNG-powered vehicles to the list. Natural gas is mainly obtained from gas wells or is driven off as a by-product during production of crude oil. The gas typically contains 80–99% methane, together with some higher hydrocarbons and impurities [10]. A product broadly equivalent to natural gas can also be produced biogenically (termed ‘biomethane’). Natural gas for automotive spark ignited engines continues to receive considerable attention in the literature (e.g. [11–13]). When compared to usage of gasoline as a vehicular fuel, CNG exhibits significant potential for the reduction of both gaseous emissions [10,14–19], and solid emissions [8,10,17,18,20,21]. Carbon dioxide emissions from a vehicle fueled with methane are typically some 25% lower than the CO2 emissions from a similar vehicle fuelled with gasoline [7], helped by the fact that the carbon
NOx PM
1,5 1,0
1,0
1 ,0
1,0
0,5 0,2
0,15
0,1 0,1
0,08
0,06 0,0050* 0,068 (0,0045)
0,1
0,06 0,0050* 0,068 (0,0045)
0,0 Euro 3
Euro 4
Euro 5
Euro 6
Fig. 1. Progress in European emission regulations for passenger cars fitted with spark ignition engines.
to hydrogen ratio of methane is low – approximately 52% lower than that of gasoline. Usage of biomethane can dramatically reduce fuel life-cycle greenhouse gas emissions, compared to gasoline, particularly when employed in tandem with other technologies such as hybridization [22]. EU legislation admits the possibility of lifecycle greenhouse gas reductions as high as 86% [1]. California Air Resources Board (CARB) legislation makes reference to a list of the standard CO2-equivalent emissions from the quantity of fuel which contains one megajoule of energy, with all types of CNG listed faring substantially better than gasoline [23]. It is also worth highlighting that EU legislation already in force [1] negates the CO2 emissions produced from combustion of biofuels (including biogas) in its fuel life-cycle assessments. Given the potential for synergistic effects between efforts to reduce emissions of harmful pollutants and CO2 [8], research on emissions resulting from usage of CO2 is a priority. The aim of this study was to assess and compare the emissions performance of a Euro 5 vehicle operating on gasoline and CNG, for comparison to the planned Euro 6 limits, shown in Fig. 1. CO2 emissions were also measured and analyzed. Furthermore, instantaneous concentrations of regulated compounds and CO2 were measured undiluted at the vehicle’s tailpipe, in order to gain further insight into emissions phenomena affected by the fuel type in use (i.e. CNG or gasoline). 2. CNG technology for light-duty vehicles Natural gas is already widely used in bi-fuel light-duty vehicles (LDV) and light-commercial-vehicles (LCV) as an alternative to gasoline. The physicochemical properties of CNG are important in all discussions in this area and Table 1 presents typical characteristic properties. At this stage of development, vehicle and engine configTable 1 Typical values of key properties of natural gas from an automotive viewpoint (Bielaczyc [14]). Property (units /conditions)
Value
Carbon to hydrogen ratio Relative density (kg/dm3 at 15 °C/1 bar) Boiling point (°C/1 bar) Flashpoint (°C) Octane number (RON/MON) Methane number Stoichiometric air/fuel ratio by mass Lower heating value (MJ/kg) Methane concentration (volumetric %) Ethane concentration (volumetric %) Nitrogen concentration (volumetric %) Carbon dioxide concentration (volumetric %) Sulfur concentration (ppm, mass) Wobbe index (MJ/m3)
0.25–0.33 0.72–0.81 162 540–650 120–130 80–99 17.2 38–50 80–99 2.7–4.6 0.1–15 1–5 <5 41–58
136
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
urations play a significant role in determining exhaust emissions performance and compliance with applicable emissions legislation. To meet low exhaust emission limits, dedicated CNG fuelling control systems must be employed. Because of the lower energy density of natural gas (compared to gasoline), certain changes to the engine control system are necessary to avoid power loss. Engine technologies such as variable valve timing, EGR or highly refined systems for the direct injection of natural gas can help to enhance the emissions performance and improve fuel efficiency of gas engines. The aforementioned technology for light-duty CNG vehicles is already well established, and nowadays CNG vehicles have comparable performance to their gasoline-powered counterparts. It is advisable to employ spark-ignited stoichiometric combustion for all operating conditions, so that a three-way catalyst (TWC), modified and improved for more effective elimination of CH4 can be used. The TWC must be specially designed for the conversion of CH4which is difficult to achieve. The substantial shift
Fig. 2. Internal view of the climatic chamber within the test laboratory used in this study.
in the chemistry of the exhaust gas which occurs (depending on the fuel type currently in use) presents a major challenge, as a bifuel vehicle must be capable of running on both its designated fuel types without incurring excess emissions or drivability problems. 3. Experimental program A series of laboratory tests were carried out in the climate-controlled exhaust emissions laboratory (Figs. 2 and 3) at BOSMAL Automotive R&D Institute in Bielsko-Biala, Poland as part of a broader research program investigating the influence of alternative fuels on exhaust emissions from automotive vehicles with spark ignition and compression ignition engines. The test laboratory and its analyzers have been described in detail elsewhere [24,25]; all equipment is fully compliant the applicable EU regulations for legislative testing [26]. Testing was performed in accordance with the EU’s legislative exhaust emissions test procedure for vehicles featuring spark ignition engines [26], which involves collecting a sample of diluted exhaust gas in sample bags for accurate chemical analysis, achieved by means of a temperature-controlled constant volume sampler. The analyzers directly measure total hydrocarbons (HC), carbon monoxide (CO), oxides of nitrogen (NOx), carbon dioxide (CO2) and methane (CH4); non-methane hydrocarbons (NMHC) are calculated from (HC–CH4). Concentrations of regulated compounds measured from the sampling bags are converted into mass and divided by the exact distance covered by the vehicle during the emissions test, to produce a result in (mg/km) (but (g/km) for CO2), as used by the legislation. Five tests were performed on each fuel type, with mean values taken for data analysis. Additionally, continuous concentrations of regulated pollutants were also measured during the execution of the emissions tests, to gain further insight into emissions phenomena. These concentrations are converted into instantaneous emissions (units (mg/s) or (g/s)). In line with current EU legislation, the New European Driving Cycle was employed. This test cycle is executed by driving the vehicle on a chassis dynamometer and the cycle consists of two phases: the urban driving cycle (UDC), followed by the extra-urban driving cycle (Fig. 4). Crucially, this driving cycle is performed with a cold
Fig. 3. Schematic of the exhaust emissions testing setup used in this study.
137
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
Table 2 A comparison of emissions from the test vehicle when running on CNG to the relevant Euro 6 limits. Emissions and emissions margins are given to the same number of decimal places as featured in the legislation.
UDC
120
EUDC
100 80 60
Compound
Mean emissions (mg/km)
Euro 6 limit (mg/km)
Margin (%)
HC NMHC CO NOx
28 12 313 13
100 68 1000 60
72 82 69 78
40
0
0
200
400
600
800
1000
1200
Time [s] Fig. 4. The New European Driving Cycle.
engine (cold start; temperature of the lubricating oil and all metallic components of the engine approximately equal to the ambient temperature). The UDC consists of 4 repetitions of the ECE elementary cycle, and as such each repetition of the cycle can be regarded as a sub-cycle (hereafter referred to as ‘1st ECE’. Modal emissions data were integrated over the first 195 s of the UDC (i.e. the 1st ECE) to provide emissions factors for this initial sub-cycle. The ECE is of particular interest as it contains the start-up event, the first time the vehicle begins to move and the first gearshifts – all transient events with significant impacts on emissions – as well as the shift from gasoline to CNG operation (when in CNG mode). All tests were performed under identical test conditions, on both CNG and standard European gasoline (in turn). Measures were taken to pre-condition and heat soak the vehicle before each test, in line with the legislation and good emissions testing practice. The climatic chamber’s environmental parameters lay within the range of values permitted by the legislation; an ambient temperature of 24 °C and 45–50% relative humidity were used for all tests. The tests were conducted on a bi-fuel vehicle of European manufacture, powered by an SI engine, with fuel supplied to the engine either via the vehicle’s multipoint injection (MPI) gasoline fuel injection system or its MPI natural gas injection system, depending on the fuel type in use. Start-up of the engine is always performed on gasoline and is automatically switched over to CNG operation after a few seconds (when the vehicle is in CNG mode). The aftertreatment system of the car tested consisted of a three way catalytic converter (TWC) specially adapted for bi-fuel cars (gasoline and CNG). Certain changes to the engine construction of the test vehicle were also implemented by the manufacturer. The construction of the cylinder head was optimized for CNG combustion by making changes to the valves and valve seat materials. These changes in the engine, together with the additional (CNG) fueling system, together with the installation of a larger TWC, cause a slight increase in vehicle mass compared to the standard (gasoline only) model. Notwithstanding these important departures from the characteristics of the mono-fuel version of the test vehicle, no further modifications of any kind were made and the test vehicle is representative of its vehicle model, as used on the road. Two fuels were tested: standard European gasoline and a highquality CNG mixture available in the EU. Further details on the two fuels used can be found in Appendix A. 4. Results and discussion 4.1. Comparison of emissions results from the test vehicle running on CNG to Euro 6 limits The test vehicle was found to fully comply with Euro 6 limits, by a considerable margin, as shown in Table 2. While the vehicle is
certified to the Euro 5 standard, whose limits are identical to the Euro 6 standard, the observed order of magnitude of the margin is surprisingly large, being on the order of 70–80%. However, the test vehicle is a small passenger car with a relatively low mass and an engine of power output typical for such a vehicle. In short, the vehicle is likely to present a base-case scenario, although there was some evidence that emissions from vehicle operation on CNG could be optimized still further (discussed in Section 4.2).
4.2. Comparison of results obtained from the test vehicle running on CNG and gasoline in turn Results obtained using the two fuels are presented Figs. 5–9. Each of Figs. 5–9 shows both phases of the test cycle, e.g. the UDC and EUDC, as well as for the complete NEDC (UDC + EUDC) and the first sub-cycle (‘1st ECE’). It is important to note that only the value for the entire NEDC is of interest in terms of legislative limits; 1st ECE, UDC and EUDC values are presented for informa-
300
HC emission [mg/km]
20
250
Gasoline
200
CNG
150 100 50 0 1st ECE
UDC
EUDC
NEDC
Fig. 5. HC emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
200
NMHC emission [mg/km]
Vehicle speed [km/h]
140
Gasoline 150
CNG 100
50
0 1st ECE
UDC
EUDC
NEDC
Fig. 6. NMHC emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
138
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
CO emission [mg/km]
1500
Gasoline
1250
CNG
1000 750 500 250 0 1st ECE
UDC
EUDC
NEDC
Fig. 7. CO emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
NOX emission [mg/km]
100
Gasoline 75
CNG 50
25
0 1st ECE
UDC
EUDC
NEDC
Fig. 8. NOx emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
250
CO2 emission [g/km]
Gasoline 200 CNG 150 100 50 0 1st ECE
UDC
EUDC
NEDC
Fig. 9. CO2 emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
tion only. Strictly speaking, the value for the 1st ECE is not directly comparable to the other values, being obtained via integration of modal emissions data, rather than from bag readings. Regarding emissions of HC, during the UDC phase, and for entire NEDC, emissions of total hydrocarbons are 20% higher when running on CNG than when running on gasoline (Fig. 5). During the ECE, emissions are high from both fuel types, with only a 7% difference observed. While this difference is numerically small, it was judged to be statistically significant, given that five repeat
tests were performed. However, for NMHC, this situation is very different (Fig. 6): NMHC emission was noticeably lower when running on CNG, the difference being approximately 30% over the entire cycle, with a particularly significant difference over the ECE. For CO (Fig. 7), the 1st ECE shows a large emissions penalty when running on CNG, but emission is in fact slightly lower during the UDC, and then higher during the EUDC, resulting in increased overall emission of CO when running on CNG (a difference of 22% over the entire NEDC). Emission of HC, NMHC and CO is highest during the 1st ECE cycle, reflecting poor emissions performance during and immediately following cold start. In contrast, emissions of both HC and NMHC are sufficiently low during the EUDC that the observed differences between gasoline and CNG are unlikely to be significant. CO and HC are exhaust gas components which can be effectively removed by a Three Way Catalyst (TWC). The TWC’s effectiveness after reaching its light off temperature is high and emissions over the entire cycle are a function of the time taken to reach light off temperature. The output of HC and CO from the engine also determine emissions performance, particularly so when the TWC has not yet reach its light off temperature. Gaseous fuels such as CNG have been reported to permit better air–fuel mixing (e.g. [15]), yet on the other hand the injection of gas brings its own set of challenges and difficulties in forming a homogenous mixture. The cyclic variation which results from factors fluctuations in the equivalence ratio [13] will likely have a significant impact on CO emissions. The increase observed in HC emissions when running on CNG may be due similar complex stability and mixture homogeneity factors, or simply due to the greater proportion of CH4 in the exhaust (CH4 being a difficult molecule to oxidize catalytically in the TWC, even in a system optimized for that purpose). The relatively low carbon fraction present in the fuel reduces the mass of air required to achieve the complete oxidation of the carbon present in the fuel, thereby potentially reducing emission of CO, as more carbon atoms are fully oxidized to CO2. As with HC, mixing behavior for a gaseous fuel such as CNG can have a strong influence on CO emissions. The fact that CO emissions are higher during the second phase when running on CNG and the substantial increase in CO emission during the 1st ECE hints at problems with mixture formation when running on CNG. The poor performance during the 1st ECE may be attributable to the transition from gasoline to CNG operation. A gaseous fuel such as CNG has no lubricating effect whatsoever (unlike gasoline), which could perhaps explain the poor CO emissions performance during the 1st ECE, where friction is very high. It is possible that the enrichment required for high load sections of the EUDC is more pronounced when running on CNG, thereby causing greater emissions of CO. Furthermore, it is reasonable to expect that where engine speed and fuel flow rates are high, mixture formation problems are likely to have a noticeable impact on CO emissions. However, it should be noted that CO emissions were comparatively low for both fuels, both being well below the Euro 6 limit. Since the test vehicle was sold as a Euro 5 vehicle, and given that its emissions were observed to meet Euro 5 emissions limits by some margin, it appears that the manufacturer had little incentive to further improve emissions performance beyond the level observed. However, further reduction of regulated emissions is likely to be possible for the vehicle tested here: most prominently the emission of HC (in terms of the aftertreatment system, as discussed above), but also of CO when operating in CNG mode. The emission of NMHC when operating in CNG mode could perhaps be somewhat reduced by a more rapid switchover to CNG operation following start-up and by further measures to limit the contribution of the lubricating oil to HC emissions. Emissions of NOx from the vehicle when fueled with gasoline are almost twice as high as when the vehicle was running on
139
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
Further research on CNG should also consider the impact on unregulated exhaust pollutant species, as future legislation may yet set additional limits. Irrespective of the legal status of a compound, a wide range of chemicals present in exhaust gas can affect air quality and the climate, both directly and indirectly.
4.3. A comparison of instantaneous emissions of regulated pollutants and CO2 in the vehicle’s exhaust The instantaneous emissions of HC, CH4, CO, NOx and CO2 measured in the undiluted exhaust gas at a frequency of 1 Hz over the test cycle are shown in Figs. 10–14. Note that in certain cases the emissions associated with the cold-start surge exceed the vertical scale. While emissions of HC for the two fuels exhibit some similarities (particularly the cold start peak), later portions of the cycle show certain key differences (Fig. 10). Emissions of CH4 are selfevidently very different when methane is used as the fuel (Fig. 11). Methane emissions associated with transient events (e.g. accelerations, decelerations) have a noticeable impact on the HC concentration. The conspicuous absence of hydrocarbon and methane peaks at the beginning of decelerations when running on gasoline, compared to the noticeable peaks observed for CNG, again suggest emissions phenomena related to mixture formation and control over the quantity of fuel supplied to the cylinder under transient conditions. Another possible explanation for this observation that when running on gasoline the TWC is able to eliminate
200
20
HC emission [mg/s]
Gasoline
Vehicle speed
16
150
14 12
100
10 8 6
50
4
Vehicle speed [km/h]
CNG
18
2 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 10. Instantaneous emissions of HC in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
200
20 Gasoline
Vehicle speed
16
150
14 12
100
10 8 6
50
4
Vehicle speed [km/h]
CNG
18
CH4 emission [mg/s]
CNG (Fig. 8), although this trend was not observed over the 1st ECE. The combustion of natural gas occurs at a higher temperature than gasoline, due to the high adiabatic combustion temperature of methane. Higher temperatures lead to increased formation of oxides of nitrogen. However, these increased temperatures also tend to reduce somewhat the time required for the TWC to reach its light off temperature, partially offsetting this unfavorable characteristic. The link between in-cylinder peak temperature and NOx formation is well-known. In addition to peak temperature, the distribution of hottest parts of the charge, thermal homogeneity and the timing of heat release can have an impact on NOx formation. The lower NOx emission results presented above may be at least partially due to systematic changes in these parameters when running on CNG than on gasoline. A recent review [8] reported a reversed trend: a 5% increase in NOx emissions when running on CNG, compared to gasoline, yet another recent study [19] reported reductions in emissions of HC, CO and NOx from 46 somewhat older test vehicles when gasoline was substituted for CNG. Zhang et al. [19] also found that differences in emissions were negligible at low vehicle mileages, but became steadily more apparent as the vehicle accumulated mileage. The relatively pristine state of the test vehicle used in this study may have had an impact on the relative performance of CNG and gasoline in terms of emissions. Of particular note is the fact that the test vehicle could easily meet Euro 6 emissions limits when running on both fuel types, but on-road CNG emissions performance during later stages of the vehicle’s life (i.e. at much higher mileages) could be substantially different. CO2 emission results are shown for CNG and gasoline in Fig. 9. The relative magnitudes of emissions from the two fuel types remain remarkably constant (24–25%), mainly no doubt due to the lower carbon fraction found in CNG than in gasoline. It is important to note that the observed difference in CO2 emissions is almost two orders of magnitude greater than the calculated uncertainty of the CO2 emissions measurement, indicating an unequivocal reduction. As mentioned previously, the low carbon:hydrogen ratio of methane (the chief component of natural gas) is advantageous regarding CO2 emissions. Emission of CO2 is a function of fuel consumption, as well as fuel H:C ratio, and using a fuel with a higher H:C ratio does not always guarantee reduced CO2 emissions, but for the vehicle used in this study the resulting effect was significantly reduced CO2 emissions, observed over both phases of the cycle. While the Euro 6 regulations do not cover CO2, this finding is of note, as vehicles meeting the Euro 6 standard will also have to comply with other pieces of EU legislation governing fleet average CO2 emissions. The observed reduction in CO2 emissions is in line with the typical 25% reduction for CNG [7]. It would be worthwhile performing further testing over additional driving cycles to determine whether the comparable CO2 emissions advantages were also observed under other operating conditions. However, the ratio of CO2 emissions for the two fuels was practically identical, whether considered over the 1st ECE (length around 1 km), or the entire NEDC (length around 11 km), indicating that fuel chemistry was the controlling factor and that the CO2 performance of CNG and gasoline do not vary strongly with journey length or driving cycle. However, this point should be investigated further, since many journeys cover significantly less (or substantially more) than 11 km. The results presented in this section have shown CNG has been observed to perform well in terms of emissions. However, this study has focused only on regulated emissions (as defined in the Euro 6 legislation). Usage of alternative fuels (and particularly gaseous fuels) affects the combustion process in many ways, which includes impacts on exhaust emissions of compounds which are not currently regulated – for example ammonia (NH3) and nitrous oxide (N2O). Recent research implies that usage of CNG could have measurable impacts on emissions of NH3 [27,28] and N2O [29].
2 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 11. Instantaneous emissions of CH4 in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
50
CNG
CO emission [mg/s]
45
Gasoline
200
Vehicle speed
40
150
35 30
100
25 20 15
50
10
Vehicle speed [km/h]
140
5 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 12. Instantaneous emissions of CO in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
200
8 Gasoline
Vehicle speed
NOx emission [mg/s]
150
6 5
100
4 3
50
2
Vehicle speed [km/h]
CNG
7
1 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 13. Instantaneous emissions of NOx in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
200
8 CNG
Gasoline
Vehicle speed
150
6 5
100
4 3
50
2
Vehicle speed [km/h]
CO2 emission [g/s]
7
but rapidly decreases to low values in both cases, most probably due to the TWC reaching light-off. Later, high-load sections of the EUDC cause different CO emissions responses, but with a tendency for higher emission from CNG, in agreement with the results presented in Section 4.2. For NOx and CO2, where emissions are non-trivial over the majority of the cycle, emissions traces are observed to coincide very well, with the main difference being the height of the peaks and the concentration observed at idle or at constant speed (cruise) (Figs. 13 and 14). This is chiefly due to thermal effects in the case of NOx, and the fuel’s carbon:hydrogen ratio in the case of CO2. At a few points early in the UDC, the height of the NOx concentration peak for CNG is considerably higher than that of gasoline, which could have been due to reduced NOx conversion efficiency in the TWC [27].
5. Conclusions An analysis performed on exhaust emissions from a Euro 5 bi-fuel light duty vehicle running on CNG and gasoline revealed overall significant differences between emissions from the two fuel types. On the basis of these analyses, it was found that the test vehicle met the Euro 6 emissions limits when using both fuel types, but with key differences in the emissions trends: NMHC emissions were much lower when running on CNG, while HC emissions were somewhat higher. Emissions increases were observed for HC and CO when running on CNG, but these increases were far too small to cause problems with the Euro 6 limits. NOx emissions from the vehicle when fueled with CNG were about around half of those observed when running on gasoline. CO2 emissions were 24% lower when the vehicle was fueled with CNG. For the first phase (the UDC), which is critical for determining total emissions during the NEDC, emissions of NMHC, CO and NOx were all lower when the vehicle was running on CNG than when it was running on gasoline. The UDC is also broadly representative of a short urban journey (no extra-urban phase) and so these observed emissions reductions over the UDC are of considerably practical interest. The start-up event caused high emissions for both fuel types. Emissions trends over the 1st ECE cycle showed benefits from running on CNG (NMHC, CO2); relatively small differences (HC, NOx); and excess CO emissions. Measurement of instantaneous emissions of HC, CH4, CO, NOx and CO2 in the exhaust revealed noticeable differences in the profiles of these pollutants over the NEDC, specifically:
1 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 14. Instantaneous emissions of CO2 in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
the vast majority of the hydrocarbons emitted during such events, whereas when running on CNG some methane survives the TWC, as evidenced by the behavior of the CH4 and HC traces. Interestingly, a recent study employing a similar test methodology [12] did not report such behavior. Regarding CO, it is clear that emission for both fuel types is dominated by the cold start peak (Fig. 12). As the vehicle begins to move, the concentration of CO from both fuel types is very high,
Qualitatively similar traces for NOx and CO2, with the main variations being in the height of the peaks and the emissions observed at idle; HC and CH4 emissions were substantially different, appearing to be strongly affected by the high levels of methane in the exhaust gas when running on CNG; CO start-up surges were much larger for CNG, possibly reflecting combustion difficulties when switching over from gasoline to CNG; CO traces for the remainder of the cycle were broadly similar, but the high load transient events of the EUDC phase caused differing emissions responses. Emissions results presented here are more favorable than a comparable study [14], which used a Euro 4 vehicle for testing.
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
Assuming these two test vehicles are representative of their respective emissions standards, it would appear that a considerable improvement has been made. Further improvements to the fuel delivery strategy and TWC could likely at least partially overcome the observed excess emission of hydrocarbons. This paper has discussed emissions in the context of emissions limits and has therefore used the legislative test procedure. However, emissions testing of CNG under more realistic, ‘‘real world’’ conditions is also necessary in order to have a complete picture of the environmental performance of usage of CNG in Euro 5/6 vehicles.
[8]
[9]
[10] [11]
[12]
Appendix A [13]
See Tables A1 and A2. [14] Table A1 Key properties of the gasoline fuel used in the test program. Property
Unit
Value
Research octane number Density @ 15 °C Benzene content Aromatic content Olefin content Ethanol content Sulfur content
– kg/dm3 % vol/vol % vol/vol % vol/vol % vol/vol mg/kg
95.5 0.7535 <1.0 <35 <18 5.0 5.6
Table A2 Key properties of the CNG fuel used in the test program.
[15] [16]
[17]
[18]
[19]
Property
Unit
Value
Methane content Ethane content Propane C4+ Nitrogen content Carbon dioxide content Enthalpy of combustion (HCV) Net heating value (LCV) Wobbe index Relative density Density @ 0 °C, 101.325 kPa
% mol/mol % mol/mol % mol/mol % mol/mol % mol/mol % mol/mol MJ/m3 MJ/m3 MJ/m3 – kg/m3
97.45 0.79 0.24 0.12 1.32 0.09 39.77 35.86 52.71 0.5694 0.7362
[20] [21]
[22]
[23]
[24]
References [1] Directive 2009/28/EC of The European Parliament and of The Council of 23.04.2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. Official Journal of the European Union 2009; L140: 16–47. [2] Kadijk G, et al. Supporting analysis regarding test procedure flexibilities and technology deployment for review of the light duty vehicle CO2 regulations. Final report, framework contract on vehicle emissions, service request #6, 2012. [accessed 20.01.2013]. [3] Mock P, German J, Bandivadekar A, Riemersma I. Discrepancies between typeapproval and ‘‘real-world’’ fuel-consumption and CO2 values. International council on clean transportation working paper 2012-02, 2012. [accessed 08.01.2013]. [4] http://www.cngprices.com/station_map.php [accessed 20.09.2013]. [5] http://www.jikovcng.com/savings-calculator/ [accessed 20.09.2013]. [6] US department of energy. Clean cities alternative fuel price report. July 2013. [accessed 21.09.2013]. [7] International energy agency. The contribution of natural gas vehicles to sustainable transport. International energy agency working paper, 2010.
[25]
[26]
[27]
[28]
[29]
141
[accessed 24.01.2013]. Takeshita T. Assessing the co-benefits of CO2 mitigation on air pollutants emissions from road vehicles. Appl Energy 2012;97:225–37. http://dx.doi.org/ 10.1016/j.apenergy.2011.12.029. NGV Global. Natural gas vehicle knowledge base – current natural gas vehicle statistics, 2012. [accessed 08.01.2013]. Bielaczyc P, Szczotka A, Woodburn J. Development of vehicle exhaust emission testing methods – BOSMAL’s new emission testing laboratory. Combust Engines 2011;1(144):3–12. Bielaczyc P, Pajdowski P, Szczotka A, Woodburn J. Development of automotive emissions testing equipment and test methods in response to legislative, technical and commercial requirements. Combust Engines 2013;1(152):28–41. United Nations – economic commission for Europe. Regulation 83, revision 4 – uniform provisions concerning the approval of vehicles with regards to the emission of pollutants according to engine fuel requirements, 2011. Czerwinski J, Zimmerli Y, Hilficker T, Bach C, Forss AM, Heen N. Unregulated emissions with TWC, gasoline & CNG. SAE Int J Engines 2010;3(1):1099–112. http://dx.doi.org/10.4271/2010-01-1286. Bielaczyc P, Szczotka A, Swiatek A, Woodburn J. A comparison of ammonia emission factors from light-duty vehicles operating on gasoline, liquefied petroleum gas (LPG) and compressed natural gas (CNG). SAE Int J Fuels Lubricants 2012;5(2):751–9. http://dx.doi.org/10.4271/2012-01-1095. Borsari V, de Assuncao JV. Nitrous oxide emissions from gasohol, ethanol and CNG light duty vehicles. Climatic Change 2012;111(3–4):519–31. http:// dx.doi.org/10.1007/s10584-011-0203-9.
Glossary Light-off: The temperature at which a catalytic conversion system reaches 50% efficiency.
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
An assessment of regulated emissions and CO2 emissions from a European light-duty CNG-fueled vehicle in the context of Euro 6 emissions regulations Piotr Bielaczyc ⇑, Joseph Woodburn, Andrzej Szczotka BOSMAL Automotive Research and Development Institute Ltd, Sarni Stok 93, Bielsko-Biala 43-300, Poland
h i g h l i g h t s Compressed natural gas is a promising fuel type for light-duty vehicles. Euro 5 and 6 emissions standards will reduce emissions to very low levels. A bi-fuel Euro 5 vehicle (CNG/gasoline) was tested on a chassis dynamometer. When operating on CNG the vehicle easily met Euro 6 limits. Carbon dioxide emissions were 24–25% lower when running on CNG.
a r t i c l e
i n f o
Article history: Received 26 February 2013 Received in revised form 28 October 2013 Accepted 4 December 2013 Available online 27 December 2013 Keywords: Passenger car Alternative fuel Compressed natural gas Exhaust emission Euro 6
a b s t r a c t Natural gas is one of the most promising alternative fuels to meet the upcoming stringent Euro 6 emissions regulations in the European Union, as well as the planned reductions in CO2 emissions. For sparkignition engines, bi-fuel fuelling equipment is widely available and engine conversion technology for European automobiles is well established, thereby facilitating usage of natural gas in its compressed form (CNG). In light of the promising characteristics and increasing usage of natural gas as a vehicular fuel, this study investigates emissions from a passenger car featuring a spark-ignition engine capable of running on both CNG and standard gasoline. Results from emissions testing of the vehicle on a chassis dynamometer are presented and discussed in the context of the Euro 6 emissions requirements. The test vehicle featured a multipoint gas injection system and was an unmodified, commercially available European vehicle meeting the Euro 5 standard. The results indicated that when fueled with CNG, such a vehicle can comfortably meet Euro 6 emissions limits, with certain differences observed in the emissions according to the fuel type used. Furthermore, when running on CNG the vehicle was observed to emit considerably less carbon dioxide than when fueled with gasoline, with the reduction closely agreeing with the results of other studies. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Worldwide population growth and industrialization have resulted in increases in demand for energy in the transportation Abbreviations: CH4, methane; CARB, California Air Resources Board; CNG, compressed natural gas; CO, carbon monoxide; CO2, carbon dioxide; CVS, constant volume sampling; EC, European commission; ECE, Economic Commission for Europe; ECU, electronic control unit; EU, European Union; EUDC, extra urban driving cycle; GHG, greenhouse gases; HC, hydrocarbons; mpg, miles per gallon; MPI, multipoint [fuel] injection; N2O, nitrous oxide; NEDC, New European Driving Cycle; NMHC, non-methane hydrocarbons, HC–CH4; NOx, oxides of nitrogen; RON, research octane number; SI, spark ignition; TWC, three-way catalyst; UDC, urban driving cycle. ⇑ Corresponding author. Tel.: +48 338130598. E-mail address: [email protected] (P. Bielaczyc). 0306-2619/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2013.12.003
sector, among other sectors. As a result, air pollution and anthropogenic greenhouse gas emissions have become key global problems. Various options are available to militate against these effects, one of which is making use of fuel types with more favorable emissions characteristics. The use of alternative fuels, mainly biodiesel, gasoline–alcohol blends, natural gas and liquefied gasolinium gas in vehicular applications has grown in recent years in European Union countries, the United States, Japan, India, Brazil and many other markets. Legislation is also in place to encourage or effectively force further adoption of these fuel types (e.g. [1]). European Union requirements regarding vehicle emissions for passenger cars and light commercial vehicles were introduced in 2009 for type approval and in 2011 for all new types, specified as Euro 5, with further requirements (Euro 6) planned from 2014
135
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
2,5 2,3
* PM - vehicles with direct injection
CO THC NMHC
2,0
Emission [g/km]
onwards. These regulations set limits for emissions of HC, CO, and NOx; CO2 emissions are covered by separate legislation. While harmful emissions are generally of relatively little direct consequence to a vehicle’s owner/operator, CO2 emissions are inherently linked to fuel consumption, which is of great importance to the owner. Two recent reports have further underlined the discrepancy between declared (i.e. type approval) CO2 emissions and the real values actually observed during vehicle usage [2,3]. In light of ongoing discussion on these discrepancies, fuel types with well-established benefits regarding CO2 emissions and regulated emissions are understandably of great interest to legislators. Such factors are less important to the average vehicle owner, but CNG has the added advantage of potentially lowering refueling costs for light-duty vehicles (compared to gasoline), based on data on pricing submitted by consumers and retailers [4,5]. Pricing data from the USA also indicates the potential for substantial reductions in refueling costs in that country, when running on CNG [6]. It is in this context that usage of natural gas as an automotive fuel seems bound to increase over the next few years – indeed, demand for natural gas for use in European road vehicles is projected to double between 2015 and 2020 [7]. A recent techno–economic analysis of a broad range of fuel types revealed CNG to outperform gasoline in terms of fuel economy, with a vehicle purchase cost that was close to that of a gasoline vehicle [8]. As things stand, it appears likely that fossil fuels will remain the chief source of energy in the transportation sector for the foreseeable future, but the aforementioned factors indicate pressure to move towards increased usage of alternative fuels. Since energy density is an important factor for automotive fuels, gases are inherently at a disadvantage compared to liquid fuels (e.g. gasoline, bioethanol, and diesel), but this unfavorable characteristic can be partially offset by compression of natural gas, creating compressed natural gas (CNG). In light of these factors, CNG is currently the best alternative to conventional transport fuels. Various basic physical and technical parameters of CNG make it a very good fuel for turbocharged SI engines (discussed in Section 2). While the energy content of the fuel is somewhat lower, as an automotive fuel CNG generally has favorable drivability characteristics and has proven relatively popular with consumers in multiple markets. By the end of 2011 the global NGV fleet numbered more than 15 million vehicles, with around 2.5 million vehicles coming into use in 2011 alone [9]. Separate (but qualitatively similar) legislative moves taking place in the EU and the USA can also be considered to be drivers of increased interest in (and usage of) CNG as an automotive fuel. In the EU, the commission’s proposal to enforce a fleet average CO2 emissions limit of 95 g/km over the NEDC by 2020, together with ongoing discussions over aiming for a fleet average figure of 70 g/km by 2025. In the US, the target requiring carmakers to increase fuel economy in new vehicles sold between 2011 and 2025 to finally reach 54.5 mpg as CAFÉ (corporate average fuel economy) by 2025. The proposed CAFÉ rule grants incentives to plug-in electric and hybrid vehicles, with the final rule adding CNG-powered vehicles to the list. Natural gas is mainly obtained from gas wells or is driven off as a by-product during production of crude oil. The gas typically contains 80–99% methane, together with some higher hydrocarbons and impurities [10]. A product broadly equivalent to natural gas can also be produced biogenically (termed ‘biomethane’). Natural gas for automotive spark ignited engines continues to receive considerable attention in the literature (e.g. [11–13]). When compared to usage of gasoline as a vehicular fuel, CNG exhibits significant potential for the reduction of both gaseous emissions [10,14–19], and solid emissions [8,10,17,18,20,21]. Carbon dioxide emissions from a vehicle fueled with methane are typically some 25% lower than the CO2 emissions from a similar vehicle fuelled with gasoline [7], helped by the fact that the carbon
NOx PM
1,5 1,0
1,0
1 ,0
1,0
0,5 0,2
0,15
0,1 0,1
0,08
0,06 0,0050* 0,068 (0,0045)
0,1
0,06 0,0050* 0,068 (0,0045)
0,0 Euro 3
Euro 4
Euro 5
Euro 6
Fig. 1. Progress in European emission regulations for passenger cars fitted with spark ignition engines.
to hydrogen ratio of methane is low – approximately 52% lower than that of gasoline. Usage of biomethane can dramatically reduce fuel life-cycle greenhouse gas emissions, compared to gasoline, particularly when employed in tandem with other technologies such as hybridization [22]. EU legislation admits the possibility of lifecycle greenhouse gas reductions as high as 86% [1]. California Air Resources Board (CARB) legislation makes reference to a list of the standard CO2-equivalent emissions from the quantity of fuel which contains one megajoule of energy, with all types of CNG listed faring substantially better than gasoline [23]. It is also worth highlighting that EU legislation already in force [1] negates the CO2 emissions produced from combustion of biofuels (including biogas) in its fuel life-cycle assessments. Given the potential for synergistic effects between efforts to reduce emissions of harmful pollutants and CO2 [8], research on emissions resulting from usage of CO2 is a priority. The aim of this study was to assess and compare the emissions performance of a Euro 5 vehicle operating on gasoline and CNG, for comparison to the planned Euro 6 limits, shown in Fig. 1. CO2 emissions were also measured and analyzed. Furthermore, instantaneous concentrations of regulated compounds and CO2 were measured undiluted at the vehicle’s tailpipe, in order to gain further insight into emissions phenomena affected by the fuel type in use (i.e. CNG or gasoline). 2. CNG technology for light-duty vehicles Natural gas is already widely used in bi-fuel light-duty vehicles (LDV) and light-commercial-vehicles (LCV) as an alternative to gasoline. The physicochemical properties of CNG are important in all discussions in this area and Table 1 presents typical characteristic properties. At this stage of development, vehicle and engine configTable 1 Typical values of key properties of natural gas from an automotive viewpoint (Bielaczyc [14]). Property (units /conditions)
Value
Carbon to hydrogen ratio Relative density (kg/dm3 at 15 °C/1 bar) Boiling point (°C/1 bar) Flashpoint (°C) Octane number (RON/MON) Methane number Stoichiometric air/fuel ratio by mass Lower heating value (MJ/kg) Methane concentration (volumetric %) Ethane concentration (volumetric %) Nitrogen concentration (volumetric %) Carbon dioxide concentration (volumetric %) Sulfur concentration (ppm, mass) Wobbe index (MJ/m3)
0.25–0.33 0.72–0.81 162 540–650 120–130 80–99 17.2 38–50 80–99 2.7–4.6 0.1–15 1–5 <5 41–58
136
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
urations play a significant role in determining exhaust emissions performance and compliance with applicable emissions legislation. To meet low exhaust emission limits, dedicated CNG fuelling control systems must be employed. Because of the lower energy density of natural gas (compared to gasoline), certain changes to the engine control system are necessary to avoid power loss. Engine technologies such as variable valve timing, EGR or highly refined systems for the direct injection of natural gas can help to enhance the emissions performance and improve fuel efficiency of gas engines. The aforementioned technology for light-duty CNG vehicles is already well established, and nowadays CNG vehicles have comparable performance to their gasoline-powered counterparts. It is advisable to employ spark-ignited stoichiometric combustion for all operating conditions, so that a three-way catalyst (TWC), modified and improved for more effective elimination of CH4 can be used. The TWC must be specially designed for the conversion of CH4which is difficult to achieve. The substantial shift
Fig. 2. Internal view of the climatic chamber within the test laboratory used in this study.
in the chemistry of the exhaust gas which occurs (depending on the fuel type currently in use) presents a major challenge, as a bifuel vehicle must be capable of running on both its designated fuel types without incurring excess emissions or drivability problems. 3. Experimental program A series of laboratory tests were carried out in the climate-controlled exhaust emissions laboratory (Figs. 2 and 3) at BOSMAL Automotive R&D Institute in Bielsko-Biala, Poland as part of a broader research program investigating the influence of alternative fuels on exhaust emissions from automotive vehicles with spark ignition and compression ignition engines. The test laboratory and its analyzers have been described in detail elsewhere [24,25]; all equipment is fully compliant the applicable EU regulations for legislative testing [26]. Testing was performed in accordance with the EU’s legislative exhaust emissions test procedure for vehicles featuring spark ignition engines [26], which involves collecting a sample of diluted exhaust gas in sample bags for accurate chemical analysis, achieved by means of a temperature-controlled constant volume sampler. The analyzers directly measure total hydrocarbons (HC), carbon monoxide (CO), oxides of nitrogen (NOx), carbon dioxide (CO2) and methane (CH4); non-methane hydrocarbons (NMHC) are calculated from (HC–CH4). Concentrations of regulated compounds measured from the sampling bags are converted into mass and divided by the exact distance covered by the vehicle during the emissions test, to produce a result in (mg/km) (but (g/km) for CO2), as used by the legislation. Five tests were performed on each fuel type, with mean values taken for data analysis. Additionally, continuous concentrations of regulated pollutants were also measured during the execution of the emissions tests, to gain further insight into emissions phenomena. These concentrations are converted into instantaneous emissions (units (mg/s) or (g/s)). In line with current EU legislation, the New European Driving Cycle was employed. This test cycle is executed by driving the vehicle on a chassis dynamometer and the cycle consists of two phases: the urban driving cycle (UDC), followed by the extra-urban driving cycle (Fig. 4). Crucially, this driving cycle is performed with a cold
Fig. 3. Schematic of the exhaust emissions testing setup used in this study.
137
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
Table 2 A comparison of emissions from the test vehicle when running on CNG to the relevant Euro 6 limits. Emissions and emissions margins are given to the same number of decimal places as featured in the legislation.
UDC
120
EUDC
100 80 60
Compound
Mean emissions (mg/km)
Euro 6 limit (mg/km)
Margin (%)
HC NMHC CO NOx
28 12 313 13
100 68 1000 60
72 82 69 78
40
0
0
200
400
600
800
1000
1200
Time [s] Fig. 4. The New European Driving Cycle.
engine (cold start; temperature of the lubricating oil and all metallic components of the engine approximately equal to the ambient temperature). The UDC consists of 4 repetitions of the ECE elementary cycle, and as such each repetition of the cycle can be regarded as a sub-cycle (hereafter referred to as ‘1st ECE’. Modal emissions data were integrated over the first 195 s of the UDC (i.e. the 1st ECE) to provide emissions factors for this initial sub-cycle. The ECE is of particular interest as it contains the start-up event, the first time the vehicle begins to move and the first gearshifts – all transient events with significant impacts on emissions – as well as the shift from gasoline to CNG operation (when in CNG mode). All tests were performed under identical test conditions, on both CNG and standard European gasoline (in turn). Measures were taken to pre-condition and heat soak the vehicle before each test, in line with the legislation and good emissions testing practice. The climatic chamber’s environmental parameters lay within the range of values permitted by the legislation; an ambient temperature of 24 °C and 45–50% relative humidity were used for all tests. The tests were conducted on a bi-fuel vehicle of European manufacture, powered by an SI engine, with fuel supplied to the engine either via the vehicle’s multipoint injection (MPI) gasoline fuel injection system or its MPI natural gas injection system, depending on the fuel type in use. Start-up of the engine is always performed on gasoline and is automatically switched over to CNG operation after a few seconds (when the vehicle is in CNG mode). The aftertreatment system of the car tested consisted of a three way catalytic converter (TWC) specially adapted for bi-fuel cars (gasoline and CNG). Certain changes to the engine construction of the test vehicle were also implemented by the manufacturer. The construction of the cylinder head was optimized for CNG combustion by making changes to the valves and valve seat materials. These changes in the engine, together with the additional (CNG) fueling system, together with the installation of a larger TWC, cause a slight increase in vehicle mass compared to the standard (gasoline only) model. Notwithstanding these important departures from the characteristics of the mono-fuel version of the test vehicle, no further modifications of any kind were made and the test vehicle is representative of its vehicle model, as used on the road. Two fuels were tested: standard European gasoline and a highquality CNG mixture available in the EU. Further details on the two fuels used can be found in Appendix A. 4. Results and discussion 4.1. Comparison of emissions results from the test vehicle running on CNG to Euro 6 limits The test vehicle was found to fully comply with Euro 6 limits, by a considerable margin, as shown in Table 2. While the vehicle is
certified to the Euro 5 standard, whose limits are identical to the Euro 6 standard, the observed order of magnitude of the margin is surprisingly large, being on the order of 70–80%. However, the test vehicle is a small passenger car with a relatively low mass and an engine of power output typical for such a vehicle. In short, the vehicle is likely to present a base-case scenario, although there was some evidence that emissions from vehicle operation on CNG could be optimized still further (discussed in Section 4.2).
4.2. Comparison of results obtained from the test vehicle running on CNG and gasoline in turn Results obtained using the two fuels are presented Figs. 5–9. Each of Figs. 5–9 shows both phases of the test cycle, e.g. the UDC and EUDC, as well as for the complete NEDC (UDC + EUDC) and the first sub-cycle (‘1st ECE’). It is important to note that only the value for the entire NEDC is of interest in terms of legislative limits; 1st ECE, UDC and EUDC values are presented for informa-
300
HC emission [mg/km]
20
250
Gasoline
200
CNG
150 100 50 0 1st ECE
UDC
EUDC
NEDC
Fig. 5. HC emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
200
NMHC emission [mg/km]
Vehicle speed [km/h]
140
Gasoline 150
CNG 100
50
0 1st ECE
UDC
EUDC
NEDC
Fig. 6. NMHC emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
138
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
CO emission [mg/km]
1500
Gasoline
1250
CNG
1000 750 500 250 0 1st ECE
UDC
EUDC
NEDC
Fig. 7. CO emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
NOX emission [mg/km]
100
Gasoline 75
CNG 50
25
0 1st ECE
UDC
EUDC
NEDC
Fig. 8. NOx emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
250
CO2 emission [g/km]
Gasoline 200 CNG 150 100 50 0 1st ECE
UDC
EUDC
NEDC
Fig. 9. CO2 emissions over the NEDC cycle and its UDC and EUDC phases from the vehicle fueled with gasoline and CNG in turn.
tion only. Strictly speaking, the value for the 1st ECE is not directly comparable to the other values, being obtained via integration of modal emissions data, rather than from bag readings. Regarding emissions of HC, during the UDC phase, and for entire NEDC, emissions of total hydrocarbons are 20% higher when running on CNG than when running on gasoline (Fig. 5). During the ECE, emissions are high from both fuel types, with only a 7% difference observed. While this difference is numerically small, it was judged to be statistically significant, given that five repeat
tests were performed. However, for NMHC, this situation is very different (Fig. 6): NMHC emission was noticeably lower when running on CNG, the difference being approximately 30% over the entire cycle, with a particularly significant difference over the ECE. For CO (Fig. 7), the 1st ECE shows a large emissions penalty when running on CNG, but emission is in fact slightly lower during the UDC, and then higher during the EUDC, resulting in increased overall emission of CO when running on CNG (a difference of 22% over the entire NEDC). Emission of HC, NMHC and CO is highest during the 1st ECE cycle, reflecting poor emissions performance during and immediately following cold start. In contrast, emissions of both HC and NMHC are sufficiently low during the EUDC that the observed differences between gasoline and CNG are unlikely to be significant. CO and HC are exhaust gas components which can be effectively removed by a Three Way Catalyst (TWC). The TWC’s effectiveness after reaching its light off temperature is high and emissions over the entire cycle are a function of the time taken to reach light off temperature. The output of HC and CO from the engine also determine emissions performance, particularly so when the TWC has not yet reach its light off temperature. Gaseous fuels such as CNG have been reported to permit better air–fuel mixing (e.g. [15]), yet on the other hand the injection of gas brings its own set of challenges and difficulties in forming a homogenous mixture. The cyclic variation which results from factors fluctuations in the equivalence ratio [13] will likely have a significant impact on CO emissions. The increase observed in HC emissions when running on CNG may be due similar complex stability and mixture homogeneity factors, or simply due to the greater proportion of CH4 in the exhaust (CH4 being a difficult molecule to oxidize catalytically in the TWC, even in a system optimized for that purpose). The relatively low carbon fraction present in the fuel reduces the mass of air required to achieve the complete oxidation of the carbon present in the fuel, thereby potentially reducing emission of CO, as more carbon atoms are fully oxidized to CO2. As with HC, mixing behavior for a gaseous fuel such as CNG can have a strong influence on CO emissions. The fact that CO emissions are higher during the second phase when running on CNG and the substantial increase in CO emission during the 1st ECE hints at problems with mixture formation when running on CNG. The poor performance during the 1st ECE may be attributable to the transition from gasoline to CNG operation. A gaseous fuel such as CNG has no lubricating effect whatsoever (unlike gasoline), which could perhaps explain the poor CO emissions performance during the 1st ECE, where friction is very high. It is possible that the enrichment required for high load sections of the EUDC is more pronounced when running on CNG, thereby causing greater emissions of CO. Furthermore, it is reasonable to expect that where engine speed and fuel flow rates are high, mixture formation problems are likely to have a noticeable impact on CO emissions. However, it should be noted that CO emissions were comparatively low for both fuels, both being well below the Euro 6 limit. Since the test vehicle was sold as a Euro 5 vehicle, and given that its emissions were observed to meet Euro 5 emissions limits by some margin, it appears that the manufacturer had little incentive to further improve emissions performance beyond the level observed. However, further reduction of regulated emissions is likely to be possible for the vehicle tested here: most prominently the emission of HC (in terms of the aftertreatment system, as discussed above), but also of CO when operating in CNG mode. The emission of NMHC when operating in CNG mode could perhaps be somewhat reduced by a more rapid switchover to CNG operation following start-up and by further measures to limit the contribution of the lubricating oil to HC emissions. Emissions of NOx from the vehicle when fueled with gasoline are almost twice as high as when the vehicle was running on
139
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
Further research on CNG should also consider the impact on unregulated exhaust pollutant species, as future legislation may yet set additional limits. Irrespective of the legal status of a compound, a wide range of chemicals present in exhaust gas can affect air quality and the climate, both directly and indirectly.
4.3. A comparison of instantaneous emissions of regulated pollutants and CO2 in the vehicle’s exhaust The instantaneous emissions of HC, CH4, CO, NOx and CO2 measured in the undiluted exhaust gas at a frequency of 1 Hz over the test cycle are shown in Figs. 10–14. Note that in certain cases the emissions associated with the cold-start surge exceed the vertical scale. While emissions of HC for the two fuels exhibit some similarities (particularly the cold start peak), later portions of the cycle show certain key differences (Fig. 10). Emissions of CH4 are selfevidently very different when methane is used as the fuel (Fig. 11). Methane emissions associated with transient events (e.g. accelerations, decelerations) have a noticeable impact on the HC concentration. The conspicuous absence of hydrocarbon and methane peaks at the beginning of decelerations when running on gasoline, compared to the noticeable peaks observed for CNG, again suggest emissions phenomena related to mixture formation and control over the quantity of fuel supplied to the cylinder under transient conditions. Another possible explanation for this observation that when running on gasoline the TWC is able to eliminate
200
20
HC emission [mg/s]
Gasoline
Vehicle speed
16
150
14 12
100
10 8 6
50
4
Vehicle speed [km/h]
CNG
18
2 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 10. Instantaneous emissions of HC in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
200
20 Gasoline
Vehicle speed
16
150
14 12
100
10 8 6
50
4
Vehicle speed [km/h]
CNG
18
CH4 emission [mg/s]
CNG (Fig. 8), although this trend was not observed over the 1st ECE. The combustion of natural gas occurs at a higher temperature than gasoline, due to the high adiabatic combustion temperature of methane. Higher temperatures lead to increased formation of oxides of nitrogen. However, these increased temperatures also tend to reduce somewhat the time required for the TWC to reach its light off temperature, partially offsetting this unfavorable characteristic. The link between in-cylinder peak temperature and NOx formation is well-known. In addition to peak temperature, the distribution of hottest parts of the charge, thermal homogeneity and the timing of heat release can have an impact on NOx formation. The lower NOx emission results presented above may be at least partially due to systematic changes in these parameters when running on CNG than on gasoline. A recent review [8] reported a reversed trend: a 5% increase in NOx emissions when running on CNG, compared to gasoline, yet another recent study [19] reported reductions in emissions of HC, CO and NOx from 46 somewhat older test vehicles when gasoline was substituted for CNG. Zhang et al. [19] also found that differences in emissions were negligible at low vehicle mileages, but became steadily more apparent as the vehicle accumulated mileage. The relatively pristine state of the test vehicle used in this study may have had an impact on the relative performance of CNG and gasoline in terms of emissions. Of particular note is the fact that the test vehicle could easily meet Euro 6 emissions limits when running on both fuel types, but on-road CNG emissions performance during later stages of the vehicle’s life (i.e. at much higher mileages) could be substantially different. CO2 emission results are shown for CNG and gasoline in Fig. 9. The relative magnitudes of emissions from the two fuel types remain remarkably constant (24–25%), mainly no doubt due to the lower carbon fraction found in CNG than in gasoline. It is important to note that the observed difference in CO2 emissions is almost two orders of magnitude greater than the calculated uncertainty of the CO2 emissions measurement, indicating an unequivocal reduction. As mentioned previously, the low carbon:hydrogen ratio of methane (the chief component of natural gas) is advantageous regarding CO2 emissions. Emission of CO2 is a function of fuel consumption, as well as fuel H:C ratio, and using a fuel with a higher H:C ratio does not always guarantee reduced CO2 emissions, but for the vehicle used in this study the resulting effect was significantly reduced CO2 emissions, observed over both phases of the cycle. While the Euro 6 regulations do not cover CO2, this finding is of note, as vehicles meeting the Euro 6 standard will also have to comply with other pieces of EU legislation governing fleet average CO2 emissions. The observed reduction in CO2 emissions is in line with the typical 25% reduction for CNG [7]. It would be worthwhile performing further testing over additional driving cycles to determine whether the comparable CO2 emissions advantages were also observed under other operating conditions. However, the ratio of CO2 emissions for the two fuels was practically identical, whether considered over the 1st ECE (length around 1 km), or the entire NEDC (length around 11 km), indicating that fuel chemistry was the controlling factor and that the CO2 performance of CNG and gasoline do not vary strongly with journey length or driving cycle. However, this point should be investigated further, since many journeys cover significantly less (or substantially more) than 11 km. The results presented in this section have shown CNG has been observed to perform well in terms of emissions. However, this study has focused only on regulated emissions (as defined in the Euro 6 legislation). Usage of alternative fuels (and particularly gaseous fuels) affects the combustion process in many ways, which includes impacts on exhaust emissions of compounds which are not currently regulated – for example ammonia (NH3) and nitrous oxide (N2O). Recent research implies that usage of CNG could have measurable impacts on emissions of NH3 [27,28] and N2O [29].
2 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 11. Instantaneous emissions of CH4 in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
50
CNG
CO emission [mg/s]
45
Gasoline
200
Vehicle speed
40
150
35 30
100
25 20 15
50
10
Vehicle speed [km/h]
140
5 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 12. Instantaneous emissions of CO in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
200
8 Gasoline
Vehicle speed
NOx emission [mg/s]
150
6 5
100
4 3
50
2
Vehicle speed [km/h]
CNG
7
1 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 13. Instantaneous emissions of NOx in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
200
8 CNG
Gasoline
Vehicle speed
150
6 5
100
4 3
50
2
Vehicle speed [km/h]
CO2 emission [g/s]
7
but rapidly decreases to low values in both cases, most probably due to the TWC reaching light-off. Later, high-load sections of the EUDC cause different CO emissions responses, but with a tendency for higher emission from CNG, in agreement with the results presented in Section 4.2. For NOx and CO2, where emissions are non-trivial over the majority of the cycle, emissions traces are observed to coincide very well, with the main difference being the height of the peaks and the concentration observed at idle or at constant speed (cruise) (Figs. 13 and 14). This is chiefly due to thermal effects in the case of NOx, and the fuel’s carbon:hydrogen ratio in the case of CO2. At a few points early in the UDC, the height of the NOx concentration peak for CNG is considerably higher than that of gasoline, which could have been due to reduced NOx conversion efficiency in the TWC [27].
5. Conclusions An analysis performed on exhaust emissions from a Euro 5 bi-fuel light duty vehicle running on CNG and gasoline revealed overall significant differences between emissions from the two fuel types. On the basis of these analyses, it was found that the test vehicle met the Euro 6 emissions limits when using both fuel types, but with key differences in the emissions trends: NMHC emissions were much lower when running on CNG, while HC emissions were somewhat higher. Emissions increases were observed for HC and CO when running on CNG, but these increases were far too small to cause problems with the Euro 6 limits. NOx emissions from the vehicle when fueled with CNG were about around half of those observed when running on gasoline. CO2 emissions were 24% lower when the vehicle was fueled with CNG. For the first phase (the UDC), which is critical for determining total emissions during the NEDC, emissions of NMHC, CO and NOx were all lower when the vehicle was running on CNG than when it was running on gasoline. The UDC is also broadly representative of a short urban journey (no extra-urban phase) and so these observed emissions reductions over the UDC are of considerably practical interest. The start-up event caused high emissions for both fuel types. Emissions trends over the 1st ECE cycle showed benefits from running on CNG (NMHC, CO2); relatively small differences (HC, NOx); and excess CO emissions. Measurement of instantaneous emissions of HC, CH4, CO, NOx and CO2 in the exhaust revealed noticeable differences in the profiles of these pollutants over the NEDC, specifically:
1 0
0
200
400
600
800
1000
0 1200
Time [s] Fig. 14. Instantaneous emissions of CO2 in the vehicle’s exhaust over the NEDC cycle from the vehicle fueled with gasoline and CNG in turn.
the vast majority of the hydrocarbons emitted during such events, whereas when running on CNG some methane survives the TWC, as evidenced by the behavior of the CH4 and HC traces. Interestingly, a recent study employing a similar test methodology [12] did not report such behavior. Regarding CO, it is clear that emission for both fuel types is dominated by the cold start peak (Fig. 12). As the vehicle begins to move, the concentration of CO from both fuel types is very high,
Qualitatively similar traces for NOx and CO2, with the main variations being in the height of the peaks and the emissions observed at idle; HC and CH4 emissions were substantially different, appearing to be strongly affected by the high levels of methane in the exhaust gas when running on CNG; CO start-up surges were much larger for CNG, possibly reflecting combustion difficulties when switching over from gasoline to CNG; CO traces for the remainder of the cycle were broadly similar, but the high load transient events of the EUDC phase caused differing emissions responses. Emissions results presented here are more favorable than a comparable study [14], which used a Euro 4 vehicle for testing.
P. Bielaczyc et al. / Applied Energy 117 (2014) 134–141
Assuming these two test vehicles are representative of their respective emissions standards, it would appear that a considerable improvement has been made. Further improvements to the fuel delivery strategy and TWC could likely at least partially overcome the observed excess emission of hydrocarbons. This paper has discussed emissions in the context of emissions limits and has therefore used the legislative test procedure. However, emissions testing of CNG under more realistic, ‘‘real world’’ conditions is also necessary in order to have a complete picture of the environmental performance of usage of CNG in Euro 5/6 vehicles.
[8]
[9]
[10] [11]
[12]
Appendix A [13]
See Tables A1 and A2. [14] Table A1 Key properties of the gasoline fuel used in the test program. Property
Unit
Value
Research octane number Density @ 15 °C Benzene content Aromatic content Olefin content Ethanol content Sulfur content
– kg/dm3 % vol/vol % vol/vol % vol/vol % vol/vol mg/kg
95.5 0.7535 <1.0 <35 <18 5.0 5.6
Table A2 Key properties of the CNG fuel used in the test program.
[15] [16]
[17]
[18]
[19]
Property
Unit
Value
Methane content Ethane content Propane C4+ Nitrogen content Carbon dioxide content Enthalpy of combustion (HCV) Net heating value (LCV) Wobbe index Relative density Density @ 0 °C, 101.325 kPa
% mol/mol % mol/mol % mol/mol % mol/mol % mol/mol % mol/mol MJ/m3 MJ/m3 MJ/m3 – kg/m3
97.45 0.79 0.24 0.12 1.32 0.09 39.77 35.86 52.71 0.5694 0.7362
[20] [21]
[22]
[23]
[24]
References [1] Directive 2009/28/EC of The European Parliament and of The Council of 23.04.2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. Official Journal of the European Union 2009; L140: 16–47. [2] Kadijk G, et al. Supporting analysis regarding test procedure flexibilities and technology deployment for review of the light duty vehicle CO2 regulations. Final report, framework contract on vehicle emissions, service request #6, 2012.
[25]
[26]
[27]
[28]
[29]
141
Glossary Light-off: The temperature at which a catalytic conversion system reaches 50% efficiency.