Biogenous Ethers – Production and Operation in a Diesel Engine

Biofuels from lignocellulose can significantly contribute to the greenhouse gas emissions reduction of the transport sector. Oxygenated fuels like Di-n-butyl ether and oxymethylene ethers are among the most promising alternatives for diesel engines. The suitability of these compounds is investigated in a research project conducted at the Institute for Powertrains and Automotive Technology at Vienna University of Technology and the Institute of Chemical Engineering and Environmental Technology at Graz University of Technology. The fuels show CO 2 saving potential by up to 70 % and remarkable soot reduction properties.


DI-N-BUTYL ETHER AND OXYMETHYLENE ETHERS AS DIESEL SUBSTITUTE FUELS
Fuels from lignocellulosic biomass have the potential to contribute to sustainable future mobility targets by reducing the fossil CO 2 emissions of the transport sector. Of special interest for the diesel engine are oxygenated fuels, since they can help to solve traditional conflicts of objectives like the soot-NO x trade-off. Di-n-butyl Ether (DBE) and Oxymethylene Ethers (OME) are among the most promising fuel candidates.
Due to its high reactivity, DBE (C 8 H 18 O) shows a very short ignition delay and is used by some researchers as an ignition booster in blended fuels [1,2]. The properties of the other investigated fuel, OME, depend on the molecule size and therefore on the number of oxymethylene units (-CH 2 O-). The resulting formula is H 3 C-O(-CH 2 O-) n -CH 3 . Generally, OME with molecule sizes n = 3 to 6 are found to have the best suitability for diesel engine applications, since their properties are closest to diesel [3][4][5].
In this study, a certified diesel fuel is used as reference for comparison and is briefly referred to as CEC. DBE is additivized with 1000 ppm of monocarboxylic acid to ensure sufficient lubricity. OMEmix, the second investigated fuel, consists mainly of OME3 (approximately 78.7 wt%) and OME4 (approximately 19.8 wt%). The investigated compounds are injected into the combustion chamber both as a neat fuel and as a diesel-biofuel-blend with 20 vol% biogenic share. The most important fuel properties are summarized in TABLE 1.

PRODUCTION PATHWAYS OF DBE AND OMEMIX AS SECOND-GENERATION BIOFUELS
For DBE production, the pulping process of an ethanol organosolv process is simulated. The main components of lignocellulose (cellulose, hemicellulose and lignin) are separated. Afterwards, hemicellulose and cellulose are converted into fuels by fermentation and/or reactive processes and synthesis routes. A concept scheme of the biomass conversion is shown in FIGURE 1 (left). In this study  two different cases of side stream treatments are compared -case one is an integrated conversion of the hemicellulose fraction into 2-Methyltetrahydrofuran and case two is the thermal conversion of hemicellulose for steam and power generation. Plant size is set on an industrial scale with a biomass consumption of 101 t/h. Biomass source is the energy plant Miscanthus Sinensis and process operation is based on the work of [6]. During ethanol organosolv pulping, 40.7 t/h cellulose are gained. Cellulose is then converted into glucose by enzymatic hydrolyzation. Subsequent fermentation of glucose yields in acetone, butanol and ethanol. Product separation is performed by distillation. Formed ethanol is used for the pulping process to compensate ethanol losses. Additional need of energy is provided by coal combustion with a calculated CO 2 output of 111.3 gCO 2 /MJ. DBE can be produced by dehydration of butanol. Butene and water are the main byproducts from the reaction, which in this case is carried out with an Al 2 O 3 /SO 4 catalyst. Not converted butanol is recirculated and butene is used for energy generation. Eventually DBE is being purified through methods of extraction and distillation.
Basis for the OME production simulation is the manufacturing of 250.000 t/a of an OMEmix. Biomass gasification and subsequent methanol-synthesis is determined as reaction pathway, FIGURE 1 (right). Biomass feed in the calculations is 95 t/h spruce wood. Selected gasification process is the stepwise biomass gasification Carbo-V process, operated by Linde. Gas cleaning is modelled with a selexol process and methanol synthesis is simulated based on a slurry reactor of Air Products in a low pressure methanol synthesis process. Formaldehyde production is simulated with the BASF process [7] to provide formaldehyde for methylal and trioxane formation. Trioxane production is modelled based on the method of Grützner [8]. Additionally to trioxane, methylal is the second building block for OME production. After a reactive distillation with water as entrainer, the production of OME mixture is calculated following the design of Burger [9].
Evaluation criteria and the results of the simulations are illustrated in TABLE 2. To show a comparison with a first generation biofuel, a sodium methanol based biodiesel plant for rape seed oil conversion into biodiesel is simulated. However, rape seed does not belong to the second-generation biofuels and its availability is limited, whereas feed stock for DBE and OMEmix is highly available.

COMBUSTION INVESTIGATIONS
A modified common-rail diesel engine is used as a test engine. The investigations are carried out at two operating points: OP-1500/5 and OP-1500/15 at 5 and 15 bar indicated mean effective pressure and 1500 rpm. In order to create boundary conditions for the fuel comparison, an upper limit for NO x emissions of 2 g/kWh is set. The fuel is injected through a single injection event.

POLISH YOUR KNOWLEDGE POLISH YOUR KNOWLEDGE POLISH YOUR KNOWLEDGE WITH WITH WITH IST IST IST International Surface Technology International Surface Technology International Surface Technology
Success requires going beneath the surface. Enhance your industry knowledge with Germany's leading magazine for surface technology. International Surface Technology (IST), the "Best of JOT" in English takes you layer-by-layer, with articles that are practical and user-oriented. Read 3 issues per year and also receive the digital edition, plus free access to the online archive featuring expert articles.
Discover even more at: www.my-specialized-knowledge.com/ist the soot mass emissions of neat DBE operation remain lower in comparison to diesel and at 50 % EGR the reduction is more than 40 %. However, the exhaust gas in blend operation shows a remarkably higher soot mass content than diesel reference operation. The short ignition delay of neat DBE reduces the lift-off length of the injected fuel jet and leaves less time for spray evaporation, thus more locally fuel rich zones and more diffusive combustion occur. Due to the advantageous properties of the DBE molecule (oxygen content, no C=C double bonds), the negative effect of the high cetane number on soot production is compensated, FIGURE 2 (left). On the other hand, adding only 20 vol% DBE to the diesel fuel has a very negative impact on soot production, because the DBE fraction of the fuel autoignites earlier and consequently ignites the diesel portion [5].
In order to fulfill the requirement of maximal 2 g/kWh NO x at OP-1500/15, a high EGR rate and/or a delayed center of combustion (MFB50%) is needed. FIGURE 2 (right) shows the development of the soot mass emissions for diesel, 20 % DBE and neat DBE over a variation of the EGR rate at OP-1500/15. At higher load the fuel injection duration is longer and ignition and combustion begin before the end of injection. Due to high fuel and burnt gas amount, the locally rich combustion zones are more than at the lower load point. Because of the reduced lower heating value of DBE in regard to diesel, the injection of the highly ignitable ether takes longer, which means that a larger amount of the fuel is injected into a burning flame zone. All these aspects account for increased soot emission levels [5]. In terms of efficiency no significant differences are measured between the three fuels. The exhaust gas of pure OMEmix combustion shows extremely low particle number -even at 50 % EGR the value is in the range of 2 × 10 11 particles/kWh. For comparison, at this point diesel is measured to produce 1.4 × 10 15 particles/kWh [4]. At the higher load point OP-1500/15, no soot mass reduction is measured when operating with 20 % OMEmix, FIGURE 3 (right). Nevertheless, the superior qualities of neat OMEmix are present at higher loads as well. At neat OMEmix operation with 20 % EGR, soot mass is below 1 mg/kWh and particle number is measured to be 1 × 10 12 particles/kWh [4].

RESULTS FOR SOOT EMISSIONS AND EFFICIENCY AT OMEMIX OPERATION
The thermodynamic analysis of the low load point operation OP-1500/5 revealed no difference in combustion and efficiency between diesel, 20 % OMEmix and neat OMEmix. The way the indicated efficiency changes with increasing the EGR rate and changing the MFB50% at OP-1500/15 is shown in FIGURE 4. The OMEmix blend shows higher tolerance against high amounts of recirculated exhaust gas than diesel; the efficiencies are comparable.
Due to the much lower lower-heating value of OMEmix compared to diesel, the fuel injection takes more time and results in an extended duration of the main combustion, which decreases the thermodynamic efficiency. In order to reduce the injection duration, the injection pressure is chosen to be 1200 bar for the investigations with neat OMEmix (for comparison diesel is injected with 1000 bar). By further increasing the injection pressure, a part of the efficiency drawback can be compensated, as exemplarily shown for OMEmix at 1600 bar injection pressure.

SUMMARY
The production of DBE and OME as second-generation bio-fuels is feasible, but requires high amounts of biomass and high investment costs. The process costs, however, are acceptable. Both fuels prove good performance in the test engine. At higher load, OME combustion shows an efficiency disadvantage due to its low lower-heating value in combination with an unadapted injection system. Soot and particle emissions can be reduced to a very low level by using neat OMEs and in this case the classical soot-NO x trade-off is being avoided. DBE usage on the other hand cannot effectively decrease soot emissions in this study. Nevertheless, both fuels offer significant well-to-wheel CO 2 reduction potentials by up to 70 %.