Beginning in 1929, R. J. Van de Graaff pioneered the Van de Graaff accelerator, in which a high potential difference is built up and maintained on a smooth conducting surface by the continuous transfer of positive static charges from a moving belt to the surface. When used as a particle accelerator, an ion source is located inside the high-voltage terminal. Ions are accelerated from the source to the target by the electric voltage between the high-voltage supply and ground.
The maximum energy obtainable from an electrostatic accelerator such as the Van de Graaff can be greatly increased by the application of the "tandem" principle. In a tandem Van de Graaff accelerator, first built in the 1950s, negative ions are first accelerated towards a positive high-voltage terminal in the center of a pressure tank. Inside the terminal the negative ions, which now have an energy in MeV equal to the terminal potential difference in megavolts (106 V) times the charge of the ion, pass through either a foil or gas "stripper" and are stripped of electrons, producing a positive-ion beam. This beam is then accelerated a second time away from the high-voltage terminal. Many tandem Van de Graaff accelerators are in operation throughout the world, including the 25-MV Holifield facility at Oak Ridge National Laboratory in Tennessee.
The limitation of these types of accelerators arises from the maximum practical potential difference that can be held by the charged surfaces. An additional problem in the tandem accelerator is the need to start with negative ions, which can be hard or impossible to obtain for some elements. Positive ion sources are available for a wider variety of elemental species. Positive ion sources can also produce ions of charge higher than one, which is all that is obtainable in negative ion sources.