Lean-burn spark ignition direct injection (SIDI) engines offer significant potential for improving engine efficiency and reducing greenhouse gas emissions. However, NOx reduction in lean-burn SIDI engines presents significant challenges. One of the exhaust architectures that is currently being investigated and developed for lean-gasoline applications is the passive ammonia-SCR (three-way catalyst selective catalyst reduction) system. It involves the use of the closed-coupled three-way catalyst (TWC) to generate NH3 (during fuel-rich conditions) for use by the downstream NH3-SCR for NOx reduction. However, the NH3 formed in TWC has to be controlled so that there is enough NH3 for NOx reduction in SCR, and at the same time, there is no NH3 slip in the tailpipe. A mathematical model for the TWC, which can predict the net NH3 coming out of the TWC, will be very useful for control algorithm development and for understanding and optimizing the exhaust architecture. The focus of this work is on the kinetic modeling of NH3 formation and oxidation reactions in a three way catalyst (TWC). Controlled steady-state and transient test cell experiments were performed at different air-to-fuel ratios and different engine conditions. This data was used to expand the existing TWC global reaction to include three additional global reactions that can account for NH3 formation and oxidation. A part of the experimental data was used to estimate the kinetic parameters of the above reactions (using a one-dimensional mathematical model for the TWC) through optimization techniques. The estimated kinetic parameters are able to predict the rest of the steady-state and controlled transient experimental data reasonably well.