Physics was my best HS subject. My junior year science fair entry was a Van de Graaff Generator. I had fanciful visions of making it into an 'atom smasher'. My project stopped just past HV Generator stage but short of 'Particle Accelerator', see Pict above. Dr. T. Marshall Hahn was either a judge or an official at the VA State level Science Fair that year. I won 2nd place ribbon.
Van de Graaff generators have a continuous wide flat belt running on two crowned pulleys. The belt is made of material optimized to carry static charge. The driving pulley is located below the Base Plate. The 2nd pulley is located in the HV dome. In the Base a low current HV dc power supply excites a 'charge spray brush' that spans across the belt width to place charge on the belt. A similar arrangement removes charge in the Dome. The Dome Pulley has an internal (rotating PM field) alternator to power the proton source inside the high voltage Dome.
There are 25 tubular dome size 'corona ring' electrodes (hula-hoop shape) mounted between the Dome and Base at a uniform pitch. Inside the rings there is a set of HV resistors (maybe 5000 Megohm each) connected in series ring-to-ring and terminated on the Dome at the top and the Base at the bottom. This provides a uniform linear voltage gradient from (Dome) top to (Base) bottom on both the 25 corona ring and the accelerator tube stations.
Also mounted between the Dome and Base is a ~5 inch outside diameter accelerator Tube.This Tube is a 25 pair laminated stack of annular Pyrex insulators and annular metal electrode rings (Blue) fabricated on the same pitch as the HV corona rings. The electrode rings are electrically insulated from each other by the Pyrex spacers. The ring spacer material selection criteria included matched thermal coefficient of expansion to prevent shear stress between Pyrex and electrode rings over full operating temperature conditions. The electrodes are a special cup shape (see illustration) such that charged particles accelerating down the tube can only see metal. This helps keep the beam focused by masking any uneven voltage gradients that may develop on the Pyrex insulators.
The 'Accelerator Tube' converts a Van de Graaff high voltage Generator into a Particle Accelerator or 'Atom Smasher'. When in operation the accelerator tube is highly evacuated to very low pressure by a mercury (or oil) diffusion pump backed up with a mechanical roughing pump.
At the top of the tube in the Dome, a small glass vial is mounted concentric to and opens into the accelerator Tube. Hydrogen gas is introduced into the vial at low pressure. Two circular electrodes circumscribe the vial at about 1 inch spacing and are energized with high voltage RF oscillator. Hydrogen gas is admitted to the vial which becomes ionized by the RF field (glows purple). A +dc HV electrode is mounted through the top of the vial. This electrode attracts and removes (-) electrons freed from the ionized hydrogen atoms (protons). The 1 (+MeV) Dome potential then accelerates the protons down the accelerator tube to a magnetic discriminator and target.
To reduce the energy dispersion in the proton beam before target encounter, the proton beam was passed through a 90 o collimated curved uniform magnetic field (shown BLUE)). High speed Protons entering the magnet trace a radius (circular) path in the magnetic field. The Radius of curvature R=m*v/(B*q) (see Lorentz force and MHD). Deuterium (found ~1 part in 6000 in nature) has one neutron (D has ~twice mass of H) giving it 2X the radius of curvature in the magnetic field effectively discriminating (eliminating) this particle. In addition, protons going too fast swing wide in the magnetic field and slow particles curve inward. A slit of adjustable width can tune the sharpness of the energy spectrum allowing only those protons of certain mass and energy to pass. Following the slit would be the target of interest for the experiment under way.
My [device] needed the same Type accelerator Tube as VPI, which would have been manageable. The 'proton particle' route, however, seemed very difficult to complete so I choose for my Van de Graaff to accelerate electrons (BTW, then need negative Dome polarity). One can get electrons by simply heating a tungsten filament to incandescence. This produces thermionic emission the same as in vacuum tubes. I was advised against this for my own safety as my Van de Graaff would have produced rather dangerous ~100 kV X-rays. It was a fine HV generator producing 8-14 inch sparks depending on the relative humidity during operation. At first it used fabric belts and later I bought a 4" wide (HyCo-Red electrostatic belt) which worked better. It would produce 10-20 uA on dry days.
I credit and owe Dr. T. Marshall Hahn for enabling my engineering career as a result of his advocating for my acceptance by VPI admissions (despite not so good a HS grades record). I will never forget that pivotal event. Dr. Hahn was President Of VPI when I graduated. For readers who got this far, thanks for indulging me.