Good reduction of elliptic curves defined over \({\mathbb Q}(^ 3\sqrt{2})\) (Q1820835)

A classical theorem of J. Tate states that there are no elliptic curves over \({\mathbb{Q}}\) with good reduction everywhere. Generalizations to complex quadratic fields are due to Stroeker and Setzer. Stroeker proved that if an elliptic curve on a imaginary quadratic field has a global minimal equation then it has bad reduction somewhere. Setzer proved that there are no elliptic curves with a good reduction everywhere on imaginary quadratic fields with class number prime to 6. In these papers the finiteness of the group of units is crucial. We prove that on \(K={\mathbb{Q}}(^ 3\sqrt{2})\) there are no elliptic curves with good reduction everywhere. In this field the group of units is infinite. We show first that any such curve must have a point of order two K-rational. We show then that the solutions of the diophantine equations we have to consider do not correspond to elliptic curves with good reduction everywhere using Tate's algorithm.