Development of high radiation plasma scenarios with impurity seeding is necessary for the operation of burning plasma fusion devices with a reactor-relevant size such as DEMO, which should operate at high heating power and radiation fractions close to 95%. Dedicated H-mode impurity seeding experiments with both strike points on the lower vertical targets have been performed during the last JET campaigns with ITER-like wall. The main objective of these experiments was to extend impurity seeding to the highest radiation fraction independently of the confinement mode and with various impurity species (N2, Ne and Ar) and input heating powers. The impact of impurity seeding on the energy confinement has been studied as well as its dependence on the balance between main chamber and divertor radiation. Type I H-mode characteristics at BT = 2.7T, IP = 2.5MA (q95 = 3.3) and Greenwald density fraction up to 85% in low-triangularity magnetic equilibria (d = 0.22) have been examined. Respective scans over a deuterium fuelling range of 2¥1022 el/s ÷6.5¥1022 el/s and an impurity seeding range of 2¥1022 el/s ÷1.3¥1023 el/s with an additional NBI-power of 18MW have been performed. Stable highly radiative discharges are obtained with feed-forward impurity seeding. Over the full range the power loss (Ploss = Pin­dW/dt) remains above the H-mode threshold according to the empirical scaling [2]. However, this scaling has been derived from the plasma pulses with grad = Pradtot/Pheat 50% and it is aim of these experiments to investigate whether this empirical scaling is relevant for highly radiative discharges with impurity seedings.