Utilisation of VOC in Diesel Engines. Ignition and Combustion of VOC Released by Crude Oil Tankers
MetadataShow full item record
The emission of VOC (Volatile Organic Compound) has been identified as a significant source of hydrocarbon pollution. It is many sources of VOC emission. A major one in Norway is caused by the offshore oil industry. In the Norwegian part of the North Sea, the contribution from shuttle tankers loading crude oil at the oil fields is about 215000 tons a year (1997). This emission represents both an energy loss and an environmental problem. Gas tankers have used boil-off gas from the cargo tanks as fuel for some time. However, for the current VOC project a new fuel injection concept is designed for tankers to take advantage of the energy present in the VOC evaporated from crude oil. The VOC is mixed with inert gas in these tankers, and thus the utilisation of this gas represents new challenges. The VOC project uses the concept of «Condensate Diesel Process» with pilot ignition. An experimental study of ignition and combustion of VOC Fuels reported here was initiated by the time it was decided to start a pilot project converting propulsion engines in shuttle tankers to use VOC Fuel. It is an experimental study carried out at the Marine Technology Centre (MTS). The objective was to study ignition and combustion of the chosen process in comparison with an ordinary diesel process. The experimental results have been discussed and are compared with theoretical considerations of injection, ignition and combustion. For experiments on combustion, a rapid compression machine "DyFo", which was developed during the years from 1992 to 1997, was redesigned to use VOC Fuel. The DyFo test rig was initially designed to study ignition and early combustion of spark ignited homogeneous gas/air charges. To study the ignition and early combustion of VOC Fuel injected at high pressure and ignited by pilot diesel fuel, a redesign was necessary. An important feature of the DyFo, is the visualisation of the combustion. The advantage of the DyFo test rig, compared to an engine, is its simplicity and controllability. All test parameters and variables are better controlled and this makes comparable tests easier to perform. In an engine the visualisation would suffer from combustion deposits disturbing the view through the quartz glasses, making the images more difficult to interpret. The simplicity is on the other side a drawback. Correct thermal conditions inside the cylinder is hard to obtain as the piston only is moved one half stroke (from BDC to TDC) for each test. External heating of the test rig is necessary, and this makes it difficult to obtain correct thermal conditions for other than low load conditions. An injector of a new, common rail type design is developed for injection of the VOC Fuel. The design, based on a conventional, redesigned diesel fuel nozzle involved a lot of practical problems. Leakages of gas into the combustion chamber have led to some accidents resulting in too high cylinder pressures. However, broken quartz glasses have been the most serious result of the malfunctions. The main purpose of this investigation has been to study the combustion of alkanes present in the VOC evaporated from crude oil. The primary objective was to verify whether or not the concept of the «Condensate Diesel Process» with pilot ignition can be used to utilise these hydrocarbons as fuel in Diesel engines. VOC Fuels of different composition have been tested. The experiments carried out have not revealed any problems concerning combustion when using pilot ignited VOC Fuel. Both ignition and combustion seem to follow the traditional diesel scheme. The composition of the VOC Fuel seems not to affect the ignition delay. A study of varying the value of different parameters has not unveiled any surprises regarding the ignition or early combustion. Separate tests with pure VOC Fuel components, however, show great differences in the ignition delay (both regarding mean value and variance) for different components. The component being most unwillingly and most unstable to ignite is propane. If, for some reason, the pilot ignition should fail, a VOC Fuel with mainly n-butane and higher alkanes will probably give no great operating problems, at least not at medium to high engine load. These conclusions are mainly based on the analysis of the dynamic signals from the pressure sensor giving the cylinder pressure curves. Where found appropriate, Schlieren images have been included in the discussion. However, the use of Schlieren images has been found to be less valuable for the Diesel process than for the Otto process. The reason for this is mainly that the radiation of visible light from the diffusion combustion of diesel oil and VOC Fuel (i.e. propane, iso-butane and n-butane) are quite different. First, this radiation disturbs the Schlieren image and second, the radiation from the combustion of diesel oil is far more intense than that of the VOC Fuel. The light VOC fraction of the vent gas – methane and ethane – is not utilised in the concept of «Condensate Diesel Process». This fraction represents about 15 % of the total energy in the VOC release when loading crude oil at the Statfjord field. At other fields as Gullfaks, this fraction can represent up to 50% or more of the total energy. After the VOC Fuel is produced, a residual VOC consisting of methane, ethane, some propane and inert gas is lost. A useful and simple way of utilising even this fraction is to mix it with the charge air at low pressure and feed the mixture into the cylinder where a pilot fuel spray ignites the charge. The method is found to have potential of being a suitable way, at least theoretically, to utilise the light VOC fraction. Some practical difficulties, however, may restrict the use of this fraction to medium and high engine loads. At lower loads the ignition delay increases due to the dilution with great quantities of inert gas. Another option to utilise the light VOC fraction is by capturing the gas in hydrates. No real study of this concept has been carried out, but an initial survey of possible solutions is described. A final conclusion of the potential of this concept cannot be drawn until more detailed work has been carried out. However, simply using the light VOC fraction extracted by melting the hydrate will be the most likely way. As a main conclusion it can be stated that the use of VOC Fuel in a «Condensate Diesel Process» is a feasible way of utilising energy otherwise lost. Venting the VOC to the atmosphere when loading crude oil into shuttle tankers represents both an energy loss and an environmental problem. By reducing both, the idea of using VOC as an engine fuel seems to be a good one.