Masters Theses

Date of Award

12-2004

Degree Type

Thesis

Degree Name

Master of Science

Major

Mechanical Engineering

Major Professor

David Irick

Committee Members

Ke Nguyen, Jeffrey Hodgson

Abstract

As the nation’s demand for energy grows along with concern for the environment, there is a pressing need for cleaner, more efficient forms of energy. The internal combustion engine is well established as one of the most reliable forms of power production. They are commercially available in power ranges from 0.5 kW to 6.5 MW, which make them suitable for a wide range of distributed power applications from small scale residential to large scale industrial. In addition, alternative fuels with domestic abundance, such as natural gas, can play a key role in weaning our nations dependence on foreign oil. Lean burn natural gas engines can achieve high efficiencies and can be conveniently placed anywhere natural gas supplies are available. However, the aftertreatment of NOx emissions presents a challenge in lean exhaust conditions. Unlike carbon monoxide and hydrocarbons, which can be catalytically reduced in lean exhaust, NOx emissions require a net reducing atmosphere for catalytic reduction. Unless this challenge of NOx reduction can be met, emissions regulations may restrict the implementation of highly efficient lean burn natural gas engines for stationary power applications.

While the typical three-way catalyst is ineffective for NOx reduction under lean exhaust conditions, several emerging catalyst technologies have demonstrated potential. The three leading contenders for lean burn engine de-NOx are the Lean NOx Catalyst (LNC), Selective Catalytic Reduction (SCR) and the Lean NOx Trap (LNT). Similar to the principles of SCR, an LNT catalyst has the ability to store NOx under lean engine operation. Then, an intermittent rich condition is created causing the stored NOx to be released and subsequently reduced. However, unlike SCR, which uses urea injection to create the reducing atmosphere, the LNT can use the same fuel supplied to the engine as the reductant. LNT technology has demonstrated high reduction efficiencies in diesel applications where diesel fuel is the reducing agent.

The premise of this research is to explore the application of Lean NOx Trap technology to a lean burn natural gas engine where natural gas is the reducing agent. Natural gas is primarily composed of methane, a highly stable hydrocarbon. The two primary challenges addressed by this research are the performance of the LNT in the temperature ranges experienced from lean natural gas combustion and the utilization of the highly stable methane as the reducing agent.

The project used an 8.3 liter lean burn natural gas engine on a dynamometer to generate the lean exhaust conditions. The catalysts were packaged in a dual path aftertreatment system, and a set of valves were used to control the flow of exhaust to either leg during adsorption and regeneration. The rich conditions for regeneration were created by injecting natural gas directly into the exhaust stream. An oxidation and reforming catalyst were placed upstream of the LNT to enhance the utilization of the methane.

The duration of time for catalyst adsorption (sorption period) and the amount of fuel for regeneration (injection rate) were the two primary variables used in developing the regeneration strategy. The goal of this study was to optimize the regeneration strategy for 5 modes of engine operation (10%, 25%, 50%, 75% and 100% load) at 1800 rpm. In optimizing this strategy, NOx reduction efficiencies greater than 90% were demonstrated for 25% and 50% engine load. Testing at 10%, 75% and 100% load revealed the temperature dependence of both the LNT and oxidation catalyst. Low temperatures at 10% load hindered the oxidation catalyst’s ability to break down the methane, while the storage capacity of the LNT falls off at the higher temperatures of 75% and 100% load. This created a narrow temperature window in which the performance could be optimized.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS