In-situ Determination of the Thermo-physical Properties of Nano-particulate Layers Developed in Engine Exhaust Gas Heat Exchangers and Opportunities for Heat Exchanger Effectiveness Recovery.

The use of exhaust gas recirculation (EGR) in internal combustion engines has significant impacts on engine combustion and emissions. EGR can be used to reduce in-cylinder NOx production, reduce particulate matter, reduce fuel consumption, and enable advanced forms of combustion such as HCCI and PCI. To maximize the benefits of EGR, the exhaust gases are often cooled with liquid to gas heat exchangers. A common problem with this approach is the build-up of a fouling deposit layer inside the heat exchanger due to thermophoresis of exhaust stream particulates and condensation of volatiles. This deposit layer lowers the effectiveness of the heat exchanger at decreasing the exhaust gas temperature. The overall heat exchanger effectiveness is significantly influenced by the thermo-physical properties of the resulting deposit layer. Prior efforts have been made to quantify these properties, however measurements were performed ex-situ and in the absence of deposit volatiles. To generate more representative insights into the properties of these deposits, a novel optical measurement technique was developed to capture the native behavior of deposits in-situ. A visualization rig was designed and built to simulate an EGR cooler while providing optical and infrared access to the deposit. An in-situ methodology was developed based on 1-D conduction and measures heat flux, deposit wall temperature, deposit interface temperature, and the deposit thickness to calculate the deposit thermal conductivity at varying thicknesses and exhaust conditions. Results indicate that the novel methodology is capable of measuring and tracking deposit conductivity over a range of conditions. The measurement becomes more reliable with thicker deposit layers and at hotter interface temperatures. Deposit conductivity was shown to be independent of layer thickness, however varied with deposit surface temperature and volatile composition. Hypothesized removal mechanisms were also investigated with the visualization rig. Results show that a high pressure upstream flow transient into a quiescent chamber is capable of removing 30% of a deposit layer down to the bare substrate while significantly thinning the remaining deposit layer. Velocity based removal was more effective when combined with water condensation, producing almost 50% deposit removal.PhDMechanical EngineeringUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/102356/1/asalvi_1.pd