Does Helium He Form Molecule He2?

This is an update of previous post on the same theme.

The Hydrogen atom H with one electron forms a molecule H2 with substantial binding energy of 0.17 Hartree.  

What then about the Helium atom He with two electrons? We know from school that He is viewed to be a noble gas and as such would not be expected to form a He2 molecule with any binding energy. 

Experiments gives clear evidence of existence of H2 but not so of He2. 

Theory in the form of Standard Quantum Mechanics StdQM gave no clear answer for a long time, but in 1997 computations were published by Komasa and Rychlewski showing very weak binding energy (0.00004) at a kernel distance of 5.6 Bohr (compared to 1.4 Bohr for H2), which must be the same as no-binding.

Testing RealQM on a coarse $50^3$ mesh gives (run this code and vary distance D) results, which are qualitatively in accordance with the above results by StdQM, in the sense that a no-binding is indicated by the following numbers with D kernel distance, $E$ total energy and $\Delta E$ energy difference in Hartree with positive value indicating very weak no-binding 

• D=12      $E=-5.806$
• D=9.6     $\Delta E = 0.013$
• D=8        $\Delta E = 0.014$
• D=6.4     $\Delta E = 0.015$
• D=4.8    $\Delta E = 0.021$
• D=3.2    $\Delta E = 0.043$

These values are to be compared with $\Delta E = -0.17$ with strong indication of bonding for H2 at distance 1.4 Bohr. Compare with this code for He atom on $100^3$ mesh.

Both StdQM and RealQM thus indicate no-binding of two He atoms to He2 molecule at distance smaller than 12 Bohr. 

On the other hand He can form weak He2 Dimer binding by van der Waals forces at a much bigger distance of 100 Bohr. 

RealQM does not include effects of Pauli repulsion, since there is no use of a Pauli Exclusion Principle for non-overlapping one-electron densities as the building blocks of RealQM. The above results by StdQM contradict strong presence of Pauli repulsion for He2.

The reason RealQM gives substantial binding for H2 but not He2, is that the two electrons of He occupy different half spaces separated by a plane through the kernel, and with these planes perpendicular to the axis between He kernels, the two outer electrons are prevented from entering the region between the kernels to form a bond.  

Another aspect is that the kernel repulsion range for He is about 4 times that of H, because of scaling with charge squared, while the electron range is smaller for He than for H, which means that decrease of energy by electron-kernel attraction with decreasing kernel distance is countered by increase of kernel-kernel repulsion with no net decrease of total energy and so no-binding for He. 

RealQM thus appears to capture the no-binding of He2 in a qualitative sense on a coarse mesh. If this is really the case, it is remarkable. 

PS The standard explanation that noble gasses like He do not want to form molecules is that such atoms have an outer "full shell" which does not invite to either covalent or ionic bond, which may have some truth but also is vague.