SEM Fusor: easy nuclear fusion?

SEM Fusor

Confinement of deuterium ions and electrons with a static electromagnetic field

Sitemap    Possible to achieve nuclear fusion?                       

New address:

In a "normal" fusor an electric field  heations to conditions suitable for nuclear fusion.  Because most of the ions fall into the wires of the inner grid (and electrons collide with the wall), these fusors suffer from high conduction losses and do unfortunately not produce net energy. Would it be possible to design a new kind of fusor without this inner grid but that still confines the ions (and electrons) more or less in the centre?

Let's consider the following:


Fig. 1. Design of SEM fusor. (Shem)




It consists of:

  • A vacuum chamber with a low concentration of deuterium gas.

  • Two rings at a constant voltage of  about +  200 kV and one ring  and two spheres  at a constant voltage of about - 200 kV (these voltages can be reversed and the middle ring can also be deleted).

  • A constant magnetic field of about 1,5 tesla, produced by, for example, a superconducting magnet.

  • In a normal fusor the deuterium gas is ionized because of the voltage difference between the inner grid and the wall. Hopefully in our SEM fusor something similar occurs.


When do deuterium D+  ions  fuse?


If two  D+  ions at the same but opposite speed collide head on,  their speed must be at least 8,3E6 m/s in order to come at such a short distance from one each other that the strong nuclear force becomes larger than the repulse Coulomb force. (see temp.- speed calculations )


But because of  i.a. tunneling this speed can be lower:  In, in hyperphysics and other sources a temperature of about 450E6  degrees C is mentioned for D-D fusion reaction to occur. This corresponds with a mean speed of  about 2E6 m/s .



Confining  D+  ions with speeds up to 3E6 m/s


In a field of 1,5 tesla deuterium ions with a speed of 1,4 E7 m/s move in circles with a radius of 0,2 m (see   radius D+ ions in a magn.  field ).  So with this magnetic field it seems to be possible to avoid that the particles escape sidewards.


A positive charged sphere at 151 V and with a radius of 8 cm will stop a D+ particle with a speed of 3E6 m/s and an initial distance of 30 cm from the centre of the sphere. See calculationSo it seems to be possible to confine the fast D+  ions in the vertical directions with a static electric field.


The electrons, by the way, are a lot easier to confine, because of their smaller mass.



Computer simulation


We use our simulation program to simulate the SEM fusor design:


D+  ions and electrons are generated, more or less in the centre region with random speeds in all directions up to 3E6 m/s. They interact with the static electromagnetic field and with each other according Coulomb's and Biot-Savart law (non relativistic).  Leapfrog integration is used



Fig. 2. Screenshot of a simulation experiment.



This is a cross section in a vertical plane through the centre of the vacuum chamber; the three dimensional positions of the particles are projected in this plane. Old positions become black.



Result: both the deuterium ions and the electrons stay confined (up to speeds < 3,5E6 m/s)





  • If the rings and spheres of the SEM fusor are charged, will the low concentration of deuterium gas be ionized? Or could we perhaps inject electrons and/or deuterium ions inside the vacuum chamber?

  • In the simulation non relativistic formulas are used. But the particles achieve quite high speeds. What will be the error because of this?

  • Deuterium ions with very high speeds ( > 3,5E6 m/s) will still escape, according the simulation. How many?. Also the accelerating and decelerating of the ions causes losses.. (bremstrahlung). How big will be the losses?

  • In our simulation only a couple of hundreds of particles can be generated (because of lack of computer power). In reality there will be millions and millions more . Will their behaviour then be alike?

  • In the simulation a part of the electrons seem to concentrate in the centre. Will they form a kind of virtual electrode and attract the positive deuterium ions to the centre? (as in a polywell ) This would improve our fusor!

  • Without the charged spheres our design is quite similar to a Penning Trap .  The SEM fusor can be considered as a modified Penning trap, but instead of confining only one type of ion, both positive and negative particles are confined. Is this really a new idea and, perhaps..,  the egg of Columbus?  I searched the internet, but could not find something similar.

  • Charging the rings and spheres with +/- 200 kV, is it realizable? It seems to be possible to acquire dc power supplies up to 200 kV in internet.

  • Will (easy) nuclear fusion take place, similar as in the "normal" fusor? How much fusion will take place?


See also:

Video of a simulation experiment in youtube

More  information, calculations and experiments
Site map 


20 February 2014 - 2017      by  Rinze Joustra