I am following this paper to calculate the formation enthalpy of MgCu$_2$, in which the formation enthalpy is defined as follows:
where $E$(Mg$_x$X$_y$) is the total energy of the compound, and $E$(Mg) and $E$(X) are the total energies of pure elements in their stable structures. Since the influence of pressure on the condensed phases is ignored and the energies are calculated at $0$ K without any entropic contributions, the energy of formation is taken to be the enthalpy of formation.
Based on this, I calculated the formation enthalpy of MgCu$_2$. The calculation flow is shown below:
The ground state energy for the conventional cell of bulk MgCu$_2$ with Fd$\bar{3}$m space group:
$E$(MgCu2)=-.75231981E+02 (eV)
The ground state energy for the conventional cell of bulk Mg with P6$_3$/mmc space group:
$E$(Mg)=-.30124735E+01 (eV)
The ground state energy for the conventional cell of bulk Cu with Fm$_3$m space group:
$E$(Cu)=-.14921761E+02 (eV)
Then the enthalpy of formation is estimated as:
$\Delta E_f$=$E$(MgCu2) - $\dfrac{1}{3}$ $E$(Mg) - $\dfrac{2}{3}$ $E$ (Cu) = -64.2799825 (eV)
In the reference paper, the unit kJ/mol is utilized. Due to 1eV=96 kJ/mol, then the enthalpy of formation in kJ/mol is:
$\Delta E_f$ = -64.2799825 (eV) = -6170.87832 (kJ/mol)
which is very different from the result of the reference paper:
What I am missing?
PS: the lattice constants for all structures are kept the same and all results are also calculated by VASP package with consistent parameters.