You can accept H0 because the values are significantly different
When we use a z-test, we know the population mean and standard deviation. When we use a t-test, we do not know the population standard deviation and thus must estimate this using the sample data that we have collected. If you look at your z-table and t-table, tcrit for df(infinity) = zcrit because at df(infinity) we would have an entire population and no longer need an estimate.