Potential of oxymethylenether-diesel blends for ultra-low emission engines

Fuel 209 (2017) 232–237

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Potential of oxymethylenether-diesel blends for ultra-low emission engines ⁎

MARK

Ahmad Omari , Benedikt Heuser, Stefan Pischinger Institute for Combustion Engines, RWTH Aachen University, Germany

A R T I C L E I N F O

A B S T R A C T

Keywords: Oxymethylenether OME Dimethoxymethane DMM Methylal E-fuels

In the race to counteract global warming, fossil fuel dependency and urban air pollution, it is clear that vehicle emissions have to be reduced considerably. A pathway out of today’s limitations in engine technology is synthetic fuel. In particular, oxymethylenether (OME) is an attractive candidate due to its soot free combustion. In this work, OME1 – diesel blends are considered aiming to outline their potential as a near future alternative fuel. Various blends ranging from 0% to 80% vol OME1 in diesel fuel were investigated in a single cylinder research engine. Results show that tradeoffs of NOx-soot and NOx-HC/CO are in general retarded with increasing OME1 content in the fuel. With 50% (vol) OME1 in diesel, the NOx-soot tradeoff is completely eliminated up to the higher end of moderate loads ∼10 bar. Furthermore, with increasing content of OME1 in the blend, exhaust gas temperatures reduce down to −7% (°K) and indicated thermodynamic efficiency increases up to +2% with respect to pure diesel. Due to the lower cetane number (CN) of blends containing OME1, the combustion noise level increase at lower loads compared to diesel combustion. At higher loads, the lower heating value of the OME1-containing blends results in a reduction of combustion noise. To identify an optimal blending range for OME1 in diesel, the soot reduction potential is systematically analyzed by a Gaussian process regression analysis at the higher load points. From the analysis an optimal blend ratio of 35% OME1 in diesel is suggested, being the best compromise between soot reduction (∼90%) and the deterioration of fuel properties like heating value (−15%) and cetane number. With a CN of 51, the 35% OME1-diesel blend still complies with the CN lower limit in the current EN590 Norm. Finally, we conclude that from the combustion and emission point of view, OME1 is a superior diesel fuel substitutive, in particular when higher blending ratios (20–40%) are considered.

1. Introduction Aiming to reduce greenhouse gas emission, fossil fuel dependency and mitigation of urban air pollution, upper limits for vehicle emissions are continuously reduced by the legislation authorities [1]. Upcoming Real Driving Emission (RDE) legislations are expected to confront engine developers with new challenges mostly leading to complex and costly exhaust aftertreatment systems [2,3]. Consequently, alternative fuels tend to become a vital pathway for the future. For diesel engines, such fuels must not only reduce the carbon footprint, e.g. biofuels, but also have the potential for drastic soot reduction. Thereby, new flexibilities are created to comply with strict limits for Nitrogen oxide (NOx) emissions. In this regard, oxygenated fuels are considered very promising [4–8]. In particular, the synthetic fuel oxymethylenether (OME1), which was already recognized within the last two decades [9,10], has recently drawn new attention [4,11–19]. Pure OME1 combustion shows no soot formation at all [20], not only because of the high amount of oxygen bounded in the molecule (42%), but also due to the lack of C-C bonds (CH3-O-CH2-O-CH3) [16]. In addition, a high

⁎

Corresponding author. E-mail address: [email protected] (A. Omari).

http://dx.doi.org/10.1016/j.fuel.2017.07.107 Received 23 May 2017; Received in revised form 24 July 2017; Accepted 26 July 2017 0016-2361/ © 2017 Elsevier Ltd. All rights reserved.

volatility and a low reactivity contribute to soot suppression as well. In a recent work of Härtl et al. [4] OME1 was identified to have the highest soot reduction potential among a broad range of oxygenated fuels. From the availability point of view, OME1 can be easily synthesized from methanol [14,21] thus having the potential for near future implementation as a pollution mitigating blending component in diesel fuel. In a long term view, OME1 might be produced via direct synthesis routes with recaptured carbon dioxide (CO2) and renewable hydrogen [22,23], making it a sustainable alternative fuel candidate for the future. In a recent work of Schmitz et al. [14] the production costs of OME1 out of methanol were predicted for a large scale production scenario. A conservative synthesis route was chosen with methanol and water as feedstock and formaldehyde as an intermediate step towards OME1. With this route, 1.33 kg of methanol is needed to synthesize 1 kg of OME1. The OME1 cost was varied over a wide range of methanol cost leading to the finding that the feedstock price is the major factor dominating the OME1 price. In an exemplary case it is shown that a methanol cost of 300US$/t leads to an OME1 price of 559US$/t, corresponding to an increase of 86% [14]. However, when accounting for

Fuel 209 (2017) 232–237

A. Omari et al.

the lower heating value of methanol compared to OME1, the cost of 1 MJ OME1 exceeds the cost of 1 MJ methanol by 60%. Considering an exemplary diesel fuel price of 0.5US$/liter at the refinery in comparison with the above mentioned OME1 price, the OME1 cost exceeds diesel by 70% per energy unit. The latter cost penalty can be further reduced when synthesis routes are optimized for OME1 production. Deutz et al. [24] have compared the environmental impact and the exergy balance of two OME1 synthesis routes (formaldehyde route vs. novel direct route according to [22]) and found that the exergy efficiency from H2 to OME1 could be improved from 74% to 86%. An additional concerning aspect of OME1 is the high vapor pressure. Vertin et al. [9] reported a Reid vapor pressure of ∼1 bar (at 37.8 °C) for pure OME1 in a closed tank. At 71 °C, ∼2.75 bar was measured. The consequence would be high fugitive emissions when opening the tank in addition to a flammability hazard. Hence, a sealed fuel tank to contain the fuel vapors and a quick-disconnect dispensing nozzle was proposed in their work. To evaluate the flammability hazard of the trapped fuel vapors in the tank, Vertin et al. [9] measured the upper and lower temperatures at which a self-propagating flame could be maintained in the tank. For blending rations in the range of 20% to 50% OME1 in diesel, the flammability region was found to be within −30 °C to +6 °C. Above 6 °C, the vapor-air mixture was too rich for flame propagation. Nevertheless, OME1 properties are not optimal to be used as a pure fuel, not only due to its low boiling point which imposes handling and storage issues similar to those of gaseous fuels. The low reactivity and lubricity are further shortcomings although they might be overcome with proper additives [4]. Finally, the low heating value (20.1 MJ/l) may put dedicated OME1 vehicles in an unattractive perspective compared to electric vehicles. Considering the above, the vision followed in this work is the partial substitution of OME1 in diesel fuel. Such an approach provides several benefits over the use of pure OME1 as a fuel. For example, small amounts (∼5%) of OME1 could be considered already nowadays thereby enhancing it’s the large-scale production and market introduction. In a future scenario, where the production capacities of OME1 by re-captured CO2 and renewable hydrogen have increased, higher blend ratios shall be considered. In such a scenario, OME1 availability is still limited, and the question arises whether it would be preferable to use the given amount of OME1 as a pure fuel or as a blend with diesel. It is anticipated by the authors that by utilizing the OME1 as a blend, a higher pollutant emission reduction can be achieved. In other words, running one vehicle on pure OME1 and one on diesel is anticipated to give less pollutant reduction potential than running two vehicles with a 50:50 OME1-diesel blend. Considering the latter, the fundamental question arises: What blend ratio would be the optimal one? This question is investigated in this work. In addition, a combustion and emission characterization was performed for several OME1-diesel blends on a single cylinder engine.

Table 1 Fuel properties. Property

unit

diesel (B7)

OME1-diesel blends

OME1 content Oxygen content Hydrogen content Carbon content Density Lower heating value Cetane number Lubricity (HFRR) Kinematic viscosity Flashpoint

vol% w.%

0 0.8

20 9.3

35 15.6

50 21.8

80 34.1

100 42.1

w.%

13.5

12.9

12.4

12.0

11.1

10.5

w.%

85.8

77.9

72.0

66.2

54.8

47.4

kg/L MJ/kg

0.835 42.8

0.836 38.8

0.840 35.9

0.845 33

0.853 27.2

0.859 23.4**

– μm

55 < 460

53 –

51 –

48 –

37 –

24 759 [4]

mm2/s

2–4

–

–

–

–

0.36 [11]

°C

> 55

–

−25*

–

–

−32 [25]

OME1

*DIN EN ISO 3679, **DIN 51900–2 mod. Table 2 Specification of single cylinder engine. Feature

Value

Bore/Stroke Displacement Number of valves Compression ratio Maximum boost press. Peak firing pressure Piston bowl geometry

75 mm/88.3 mm 0.39liter 4 15 4 bar (abs) 220 bar ω-shaped reentrant

Table 3 Emission measurement devices. Emission

Device

Accuracy

Soot NOx HC, as C3H8 equiv. CO, CO2 O2

AVL 415s (Filter paper method) Chemiluminescence detector* Flame ionization detector** Non dispersive infrared** Paramagnetic detector**

2% 1% 1% 1% 1%

*EcoPysics CLD 700 EL **Rosemount – NGA 2000 Series.

recirculation (EGR) rate is determined by the CO2 concentration in the intake plenum. The in-cylinder pressure is measured by a water cooled piezo electric pressure transducer (Kistler 6041A) and logged at a 0.1° crank angle resolution. Subsequently, cylinder pressure traces were processed with FEV’s “Combustion Analysis System”. Combustion and emission characteristics of the OME1-diesel blends were obtained at predefined screening load points (LP) given in Table 4. The boundary conditions were determined based on data from the equivalent four cylinder engine. The center of combustion was set constant for a given load point. Thus, changes in combustion and emission characteristics could be related to the fuel only. Finally, the screening data were evaluated by a Gaussian process regression analysis method. Here the

2. Methodology Four OME1-diesel blends in addition to EN590 diesel (B7) as a reference were investigated on a single cylinder engine. Furthermore, the cetane number (CN) was measured for various blend ratios with the advanced fuel ignition delay analyzer AFIDA 2805. The properties of the considered fuels are given in Table 1. Except for CN, blend properties are calculated according to the mass fractions of OME1 and diesel in the blend. Being a solvent, no miscibility issues of OME1 in diesel were observed in the complete blending range. Due to the low boiling temperature of OME1, a pressurized fuel tank was used to avoid possible cavitation in the supply fuel pump. Also, the fuel tank was hermetically sealed to prevent fumigation of OME1 out of the blend. Specifications of the single cylinder engine and emission measurement system are given in Table 2 and Table 3 respectively. In order to measure the fuel consumption a Coriolis flow meter is used. For the intake air flow an ultrasonic gas meter is used. The exhaust gas

Table 4 Engine load points and boundary conditions. Name

IMEP [bar]

Speed [min−1]

MFB50* [°C AaTDC]

Boost [bar]

Rail press. [bar]

LP1 LP2 LP3 LP4

4.3 6.8 9.4 14.8

1500 1500 2280 2400

6.6 5.8 9.2 10.8

1.07 1.5 2.29 2.6

720 900 1400 1800

*MFB50: degree crank angle where 50% of the fuel mass has burned.

233

Fuel 209 (2017) 232–237

A. Omari et al.

Fig. 1. Cetane numbers for OME1-diesel blends as dependent on the blending ratio.

soot reduction potential with respect to diesel was modeled as function of the OME1 content in the blend. In this way, tradeoff plots of soot reduction versus fuel properties could be generated allowing the identification of an optimal blending ratio. 3. Results 3.1. Derived cetane numbers With the advanced fuel ignition delay analyzer (AFIDA 2805) the derived cetane number for various OME1-diesel blends were measured. The CN is derived based on the measured ignition delay in a constant volume combustion chamber and a subsequent correlation to an equivalent mixture of cetane and 1-methylnaphthalin having a CN of 100 and 0 respectively. As can be seen in Fig. 1, pure OME1 shows a very low CN of 24. Thus, the addition of OME1 to diesel lowers the CN of the blend. However, it is remarkable to see that up to a blending ratio of 35% (vol) OME1, the CN remains above 51, which is the lower limit of the EN 590 norm for diesel fuel. 3.2. Thermodynamic combustion characteristics Indicated thermodynamic efficiency and exhaust gas temperatures are presented in Fig. 2 for the investigated load points at predefined NOx levels representative for EU6d. A clear reduction in exhaust gas temperature and increase in efficiency can be seen with increasing OME1 content in the blend. In particular at lower loads, the increase in thermodynamic efficiency shows a clear monotonic trend with the OME1 fraction in the

Fig. 3. Cylinder pressure traces for OME1-diesel blends at LP 1, 2, 3 and 4.

blend, reaching a +3% efficiency increase for the 80:20 OME1-diesel blend. Consequently, exhaust gas temperature reduces by ∼7% (in °K). For further understanding of this phenomenon, a combustion analysis was performed. In Fig. 3 and Fig. 4 the pressure traces and the burned mass fraction traces are shown for LP 1, 2, 3 and 4 at NOx levels of 0.2, 0.2, 0.4 and 0.6 g/kWh respectively. For the readers convenience, we remind that the injection timing was adjusted for each fuel to have the 50% mass fraction burned (MFB50) at the same degree crank angle - see Table 4. As can be seen in Fig. 3, LP2 clearly shows that with increasing OME1 content in the blend, peak pressures and pressure in the beginning of the expansion stroke (> 15 °C AaTDC) are increased. With increasing load, the peak pressure is reduce for the blends with higher OME1 content but the pressure trace at start of expansion remains higher. This behavior can be understood when considering the burned mass fraction traces presented in Fig. 4. Here it can be seen that with an increased OME1 content in the blend, a higher fraction of the fuel’s energy is released close to top dead center (TDC), i.e. the combustion process is more of an isochoric nature. Subsequently, less energy is released within the burn out phase (15 to 40 °C AaTDC) and thus expansion occurs effectively over a larger expansion ratio. The latter is in agreement with the observed trend for the indicated efficiency and exhaust gas temperatures in Fig. 2. The observed reduction of peak pressure with increasing OME1 content at LP4 can be referred to the almost overlapping location of

Fig. 2. Indicated thermodynamic efficiency and exhaust gas temperature for LP 1,2,3 and 4 at NOx levels of 0.2, 0.2, 0.4 and 0.6 g/kWh respectively.

234

Fuel 209 (2017) 232–237

A. Omari et al.

Fig. 5. Average in-cylinder gas temperature trace at LP 1, 2, 3 and 4 with NOx levels of 0.2, 0.2, 0.4 and 0.6 g/kWh, resp.

Fig. 4. Burned mass fraction curves for OME1-diesel blends at LP 1, 2, 3 and 4.

peak pressure and MFB50 (for LP4: MFB50 = 10.8 °C AaTDC). Bearing in mind that with increasing OME1 content the efficiency increases, less heat is released in total. Hence, at the time of MFB50, less heat is released and consequently a lower peak pressure occurs. The shortcoming of a completer combustion close to TDC is a higher average in-cylinder temperature as can be seen in Fig. 5. This contributes to higher NOx formation. Consequently, slightly increased EGR rates are necessary to maintain a certain NOx level. The latter behavior was clearly captured by the measured EGR rates at LP2 3 and 4 as shown in Fig. 6. In contrast, for LP1 the EGR rates show a strong decrease towards the 80:20 OME1-diesel blend. This may be referred to fact that at low loads the CN dominates the mixture formation. Hence, the low CN numbers of the 50:50 and 80:20 OME1-diesel blends contribute to a significantly higher premixed combustion share. Thereby, lower local temperatures occur and thus less EGR is needed to maintain a given NOx level. The high share in premixed combustion at LP1 with the 50:50 and 80:20 OME1–diesel blends is further reflected in the high carbon monoxide emission as will be shown below.

Fig. 6. Exhaust gas recirculation rates at LP 1, 2, 3 and 4 with NOx levels of 0.2, 0.2, 0.4 and 0.6 g/kWh, respectively.

3.3. Emission characteristics For the investigated load points, EGR variations are shown in terms of tradeoff plots of NOx versus particulate matter (PM), carbon monoxide (CO), unburned hydrocarbons (HC), and combustion sound level (CSL) [26]. All load points are considered in the discussion. However, the figures present only emissions at the “emission-relevant load points”, i.e. moderate to high load for PM (LP3 & LP4), moderate loads for CSL (LP2 & LP3) and low loads for HC and CO emissions (LP1 & LP2). From Fig. 7, a clear PM reduction can be seen for LP3 and LP4. In both load points, the 80:20 OME1–diesel blend shows filter smoke 235

Fuel 209 (2017) 232–237

A. Omari et al.

Fig. 9. Combustion sound level (CSL) at moderate loads.

decreasing CN with higher OME1 content in the blend and that the OME1 molecule is in fact a CO-chain. For LP1, HC emissions are continuously reduced with increasing blending ratio similar to LP2. In contrast, CO emissions are slightly reduced only for the 20:80 OME1diesel blend, but then increase with higher OME1 blending ratios. This might be referred to the fact that at very low loads the low cetane numbers becomes more dominant in increasing the share of premixed combustion, thereby increasing CO emissions. The behavior of LP3 and LP4 at the EU6d relevant NOx level is similar to LP2, i.e. the HC and CO emissions reduce with increasing content of OME1 in the blend. That is in agreement with the observed completer combustion close to TDC shown in Fig. 4. Härtel et al. have found similar results for pure OME1 [4]. Combustion sound levels are presented in Fig. 9 for LP2 and LP3. At lower loads, CSL is majorly controlled by the cetane number of the fuel. Thus, it can be seen in Fig. 9 that for LP2, CSL continually increases with increasing OME1 content in the blend. In contrast to low loads, at higher loads the heating value of the fuel becomes more dominant in determining the combustion sound level. In fact, with a lower volumetric heating value, longer injection durations occur leading to a reduced energy supply rate to the cylinder and thus to a lower heat release rate during combustion. That is captured at LP3, where the CSL values decrease as the OME1 content in the blend increases, with the exception of the 80:20 OME1-diesel blend. For the latter, the cetane number is at such a low level (CN = 37) that CSL remains higher than diesel. LP4 (not shown in the figure above) shows a similar behavior to LP3.

Fig. 7. Particulate matter emissions at higher loads for OME1-diesel blends.

number (FSN) values within the detection limit of the AVL415s smoke meter (0 ≤ FSN ≤ 0.01), even at very low NOx levels. At LP3 and very low NOx levels, the 50:50 OME1-diesel blend show PM amounts slightly above the detection limit. For LP4, a typical NOx-soot tradeoff then becomes visible for the 50:50 OME1-diesel blend as well. The strong PM reduction is attributed majorly by the high amount of oxygen bounded in the fuel and the lack of C-C bonds. Furthermore, the high volatility and low reactivity of OME1 contributes to improved mixture formation and air utilization, thereby reducing the local equivalence ratios. Also, the relative reduction of aromatics with increasing OME1 content in the blend contributes to a further suppression of PM formation. Fig. 8 shows the HC and CO emissions at low to moderate loads. Here, different trends are observed. For LP2, HC and CO emissions are on a comparable level to diesel for the 20:80 and 35:65 OME1-diesel blends. Then, for higher blending ratios, HC and CO emissions are continually reduced. This behavior is very surprising considering the

3.4. Optimal blending ratio for OME1-diesel blends For the identification of an optimal blending ratio, the soot reduction potential is considered with respect to the deterioration of important fuel properties. Because LP1 and LP2 produce no considerable amounts of soot, only LP3 and LP4 were used here for the Gaussian process regression analysis. Fig. 10 shows the particulate matter reduction for LP3 and LP4 as well as the cetane number as function of heating value, oxygen content and OME1 content. As can be seen, the major PM reduction occurs up to blending ratios of 40% OME1 in diesel. Beyond 40%, the heating value and CN are deteriorated without a worth improvement in PM reduction. Also, it is interesting to see that the major part of PM reduction is achieved for blends having CN > 50. Considering the results above and the fact that an acceptable lower heating value should be > 30 MJ/l, a 35% (vol) OME1 in diesel was identified as an optimal blending ratio. For that blending ratio, a soot reduction of ∼90% is achieved with only 10% and 15% deterioration in CN and heating value respectively. Fig. 8. HC and CO emissions at lower loads for OME1-diesel blends.

236

Fuel 209 (2017) 232–237

A. Omari et al.

was identified to give the best compromise between PM reduction and deterioration of fuel properties. Also, it was found that the percentage in PM reduction exceeds by far the percentage of OME1 in the blend. Hence, the author’s anticipation regarding whether by blending more emission reduction can be achieved compared to the use of OME1 as a pure fuel, was justified. Acknowledgements The research leading to this contribution was funded by the German Federal Ministry of Education and Research (BMBF) under Grant number 03SFK2A (Kopernikus “Power-to-X). The responsibility for the content lies with the authors. References [1] European Commission. Commission Regulation (EU) No 459/2012 of 29 May 2012. Official Journal of the European Union, 2012. [2] Blanco-Rodriguez D, Vagnoni G, Holderbaum B. EU6C-Segment diesel vehicles, a challenging segment to meet RDE and WLTP requirements. In: IFAC symposium “Advances in Automotive Control” Norköping Schweden, 2016. [3] Timothy VJ. Review of vehicular emission trands. SAE Int J Eng 2015. 2015-01-0993. [4] Härtl M, Seidenspinner P, Jacob E, Wachtmeister G. Oxygenate screening on a heavy-duty diesel engine and emission characteristics of highly oxygenated oxymethylene ether fuel. Fuel 2015;153:328–35. [5] Kozak M, Merkisz J, Bielaczyc P, Szczotka A. The Influence of Oxygenated Diesel Fuels on a Diesel Vehicle PM/NOx Emission Trade-Off. SAE Technical Paper 2009(2009–01-2696). [6] Cheng AS, Dibble RW, Buchholz BA. The Effect of oxygenates on diesel engine particulate matter//The effect of oxygenates on diesel engine particulate matter. SAE Technical Paper (2002–01-1705). [7] Janssen A, Muether M, Pischinger S, Kolbeck A, Lamping M. Tailor-Made Fuels: The potential of oxygen content in fuels for advanced diesel combustion systems//Tailormade fuels: the potential of oxygen content in fuels for advanced diesel combustion systems. SAE Technical Paper (2009–01-2765). [8] Ren Y, Huang Z, Miao H, Di Y, Jiang D, Zeng K, et al. Combustion and emissions of a DI diesel engine fuelled with diesel-oxygenate blends. Fuel 2008;87(12):2691–7. [9] Vertin KD, Ohi JM, Naegeli DW, Childress KH, Hagen GP, McCarthy CI et al. Methylal and methylal-diesel blended fuels for use in compression-ignition engines. SAE Technical Paper 1999(1999–01-1508). [10] Kocis D, Song K, Lee H, Litzinger T. Effects of dimethoxymethane and dimethylcarbonate on soot production in an optically-accessible DI diesel engine. SAE Technical Paper 2000;2000(2000–01-2795). [11] Lautenschütz L, Oestreich D, Seidenspinner P, Arnold U, Dinjus E, Sauer J. Physicochemical properties and fuel characteristics of oxymethylene dialkyl ethers. Fuel 2016;173:129–37. [12] Zhu R, Wang X, Miao H, Yang X, Huang Z. Effect of dimethoxy-methane and exhaust gas recirculation on combustion and emission characteristics of a direct injection diesel engine. Fuel 2011;90(5):1731–7. [13] Burger J, Siegert M, Ströfer E, Hasse H. Poly(oxymethylene) dimethyl ethers as components of tailored diesel fuel: properties, synthesis and purification concepts. Fuel 2010;89(11):3315–9. [14] Schmitz N, Burger J, Ströfer E, Hasse H. From methanol to the oxygenated diesel fuel poly (oxymethylene) dimethyl ether: an assessment of the production costs. Fuel 2016;185:67–72. [15] Kopernikus-Projekte. https://www.kopernikus-projekte.de/projekte/power-to-x [accessed 05.05.17]. [16] Feiling A, Münz M, Beidl C. Potential of the synthetic fuel OME1b for the soot-free diesel engine. MTZ 2016;02.2016. [17] Jacob E, Maus W. Oxymethylenether als potenziell CO2-neutraler Kraftstoff für saubere Dieselmotoren Teil 2: Erfüllung des Nachhaltigkeitsanspruchs. MTZ 2017;78. 03. [18] Härtl M, Gaukel K, Pélerin D, Wachtmeister G. Oxymethylenether als potenziell CO2neutraler Kraftstoff für saubere Dieselmotoren Teil 1: Motorenuntersuchungen. MTZ 2017;78. 02. [19] Härtl M, Seidenspringer P, Wachtmeister G, Jakob E. Synthetic diesel fuel ome1 a pathway out of the soot-nox trade-off. MTZ 2014;2014(75):07–8. [20] Svensson KI, Richards MJ, Mackrory AJ, Tree DR. Fuel composition and molecular structure effects on soot formation in direct-injection flames under diesel engine conditions. SAE Technical Paper 2005; 2005(2005–01-0381). [21] Weidert J-O, Burger J, Renner M, Blagov S, Hasse H. Development of an integrated reaction-distillation process for the production of methylal. Ind Eng Chem Res 2017;56(2):575–82. [22] Thenert K, Beydoun K, Wiesenthal J, Leitner W, Klankermayer J. Ruthenium-catalyzed synthesis of dialkoxymethane ethers utilizing carbon dioxide and molecular hydrogen. Angew Chem Int Ed Engl 2016;55(40):12266–9. [23] Zhang Q, Zhao H, Lu B, Zhao J, Cai Q. A novel strategy for conversion of methanol and CO2 into dimethoxymethane in a basic ionic liquid. J Mol Catal A: Chem 2016;421:117–21. [24] Deutz S, Bongartz D, Heuser B, Kätelhön A, Omari A, Schulze Langenhorst L et al. Cleaner production of cleaner fuels: wind-to-wheel – environmental assessment of CO2-based oxymethylene ether as drop-in fuel: (submitted - in review process). Energy Environ Sci. [25] Lide DR. CRC handbook of chemistry and physics. Taylor & Francis; 93rd ed. [26] Alt N, Nehl A, Heuer S, Combustion sound prediction within combustion system development. SIA “Vehicle Comfort”, 2004.

Fig. 10. Modeled PM reduction as function of OME1 content in the blend at LP3 and LP4.

4. Summary and conclusions Four OME1-diesel blends and diesel as a reference were investigated in a single cylinder research engine. Thermodynamic combustion analysis showed that with increasing OME1 content in the blend, a completer combustion was achieved closer to TDC. The latter resulted in a higher thermodynamic efficiency, lower exhaust gas temperatures and in general lower HC and CO emissions. Combustion sound level (CSL) did not show a consistent trend for all load points. CSL increased with higher OME1 content in the blend at the lower load points (LP1 & LP2). Here, the lower cetane number of the OME1 containing blends dominates the mixture formation resulting in a higher share of premixed combustion. Hence higher CSL are observed with respect to diesel. For higher loads, the lower heating value of the OME1-diesel blends becomes more dominant in reducing combustion sound level. Here the longer injection duration contributes to a reduction in heat release rate. The latter is with exception of the 80:20 OME1-diesle blend due to the very low cetane number of 37 which resulted in a higher CSL even at higher loads. A strong reduction of PM emissions was observed with increasing OME1 content in the blend. The major reason for the PM reduction is the increasing oxygen content in the fuel with increasing OME1 content in the blend. Furthermore, the lack of C-C bonds in OME1, the reduced aromatic content in the blend, the higher volatility and the lower reactivity with increasing OME1 content are contributing to the PM reduction as well. However, despite of the very low PM levels achievable with OME1 diesel blends, it is clear that vehicles fueled with such blends will still be equipped with a diesel particulate filter (DPF) system. Therefore, the main advantage of the low PM emissions is the possibility for a significant NOx reduction by increasing EGR rates, while still having low PM levels ensuring extended active regeneration intervals. Finally, the soot reduction potential was modeled by a Gaussian process regression analysis as function of OME1 content in the blend. From the model an optimal blending ratio of 35% (vol) OME1 in diesel

237