Bound systems of interacting electrons. A step beyond density functional theory.
Miguel Lagos · 2025
Read Bound systems of interacting electrons. A step beyond density functional theory. by Miguel Lagos (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-18789.txt.
Introduction
Miguel Lagos presented in 2025 a method for bound electron systems that aims past density functional theory. The setup treats an interacting electron gas among heavy positive cores, using second quantization. Lagos derives ladder operators that step between stationary states of neighboring energies. Applying the lowering operator to the ground state must give zero, since no lower state exists. That fact yields equations for the ground state itself.
He explains the method on the two-electron helium atom to keep the notation light. Solving the equations gives closed forms for the ground state, its energy, and its electron density. The lowest-order result misses the true energy by well under one percent. Lagos argues the path to larger systems and finer approximations is straightforward. The work also clarifies what density functional theory rests on.
Catalog record
| CREATOR | Miguel Lagos |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-18789
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bound systems of interacting electrons. A step beyond density functional theory. Miguel Lagos Abstract The non–relativistic interacting electron gas in an external field of positively charged massive cores is dealt with in the scheme of second quantization. Ladder operators that change between stationary states of contiguous energy eigenvalues are derived. The method is particularized to the two–electron Helium atom in order to explain it avoiding too much notation. Applying the lowering operator on the ground state must give zero because no state with lower energy does exist. Equations for the ground state and ground state energy are obtained this way and solved, giving closed–form expressions for the ground state, its energy and electronic density of Helium. The theory in its lowest order gives 0.63% error. The application to more complex systems and higher degrees of approximation seems straightforward. The foundations of the density functional theory and how to go beyond it are seen quite clearly. The study of the electronic structure of molecules and solids has been greatly facilitated by the breakthrough of the density functional theory (DFT). By demonstrating that the ground state energy of the N electrons system in a external field can be regarded as a unique functional of the electron density, Hohenberg and Kohn [ 1 ] found a bypass to the problem of solving the many–body Schrödinger equation that reduces it, in principle, to the search for the density functional that minimizes the energy. Instead of seeking for a wave function of 3N real variables, the problem is substituted by determining a function of just 3 variables. Afterwards, Kohn and Sham [ 2 ] applied these ideas to a conveniently defined non–interacting system yielding the same density as the original problem of interacting electrons [ 3 ] . The procedure has proven to give valuable information on the stationary structure of atoms, molecules and solids in ground state, but a complete knowledge of the dynamical state of the system demands knowing the state vector. This communication introduces a method that may compete with the DFT in simplicity and accuracy, but gives also the state vector in the second quantized formalism. The derivations are restricted here to systems of two energy levels or bands, as are the n=0 and n=1 states of He, bonding and anti–bonding states of H 2 , or the \sigma and \pi bonding orbitals of conjugate polymers. The application to polymers and more complex systems is left for a forthcoming communication because the study of the two–electron systems is enough to explain the method. Just to show the power of the method the ground state vector is used to calculate the density and excitation energy. As an interesting byproduct of the present approach, it gives an alternate demonstration of Hohenberg–Kohn theorem and Kohn–Sham procedure of regarding the system as an ensemble of free fermions in effective potentials. In general, the N –electron system whose Hamiltonian operator in coordinate space is H=\sum_{i=1}^{N}\left(-\frac{\hbar^{2}}{2m}\nabla_{i}^{2}+V(\vec{r_{i}})\right)+\sum_{i<j}U(\vec{r_{i}},\vec{r_{j}}) (1) has a representation H=\sum_{k}\epsilon_{k}c_{k}^{\dagger}c_{k}+\sum_{klmn}U_{klmn}c_{k}^{\dagger}c_{l}^{\dagger}c_{m}c_{n}, (2) in the Fock space of states [ 4 ] . The fermion operators c_{k} are the coefficients of the field operator \Psi(\vec{r},t)=\sum_{k}c_{k}(t)u_{k}(\vec{r}) (3) expanded in the complete set of orthonormal functions {u_{k}(\vec{r})} , defined by the solutions of the eigenvalue equation \left(-\frac{\hbar^{2}}{2m}\nabla^{2}+V(\vec{r})\right)\,u_{k}(\vec{r})=\epsilon_{k}u_{k}(\vec{r}). (4) The off–diagonal coefficients are U_{klmn}=\int\,d^{3}\vec{r}\,u_{k}^{*}(\vec{r})u_{l}^{*}(\vec{r^{\prime}})U(\vec{r},\vec{r^{\prime}})u_{m}(\vec{r^{\prime}})u_{n}(\vec{r}) (5) and the indices k , l , m and n incorporate the spin indices \uparrow and \downarrow . The occupancy of the one–electron state k is n_{k}=c_{k}^{\dagger}c_{k} , which has the only eigenvalues 0 and 1 as the operators c_{k} obey Fermi–Dirac anti–commutation rules, \{c_{k},c_{k^{\prime}}\}=0 and \{c_{k},c_{k^{\prime}}^{\dagger}\}=\delta_{k{k^{\prime}}} . Notice that the restriction i<j in Eq. ( 1 ) for H in coordinate space, precluding double interactions between particles, is absent in the Fock space formulation. The condition i<j follows directly from the demonstration of the equivalence of the two representations. All interaction terms are allowed in the second quantized formalism, and the commutation properties of the creation and annihilation operators will determine what of them are relevant in each situation. This is important for the correct counting of the interaction terms. Retaining in the one–electron basis set \{u_{k}\} only the states of at most the two lowest energy levels, u_{1\uparrow} , u_{1\downarrow} , u_{2\uparrow} and u_{2\downarrow} , and assuming no spin interactions, the Hamiltonian operating in a Fock space of states with zero total spin becomes \displaystyle H= \displaystyle\epsilon_{1}(n_{1\uparrow}+n_{1\downarrow})+\epsilon_{2}(n_{2\uparrow}+n_{2\downarrow}) (6) \displaystyle+V_{1}n_{1\uparrow}n_{1\downarrow}+V_{2}n_{2\uparrow}n_{2\downarrow}+U(n_{1\uparrow}n_{2\downarrow}+n_{1\downarrow}n_{2\uparrow}) \displaystyle+\overline{U}(c_{1\uparrow}^{\dagger}c_{1\downarrow}^{\dagger}c_{2\downarrow}c_{2\uparrow}+c_{2\uparrow}^{\dagger}c_{2\downarrow}^{\dagger}c_{1\downarrow}c_{1\uparrow}+c_{1\uparrow}^{\dagger}c_{2\downarrow}^{\dagger}c_{1\downarrow}c_{2\uparrow} \displaystyle+c_{2\uparrow}^{\dagger}c_{1\downarrow}^{\dagger}c_{2\downarrow}c_{1\uparrow}), where, in Gauss units, V_{1}=\int d^{3}\vec{r}d^{3}\vec{r^{\prime}}\,|u_{1}(\vec{r})|^{2}|u_{1}(\vec{r^{\prime}})|^{2}\frac{e^{2}}{|\vec{r}-\vec{r^{\prime}}|}, (7) V_{2}=\int d^{3}\vec{r}d^{3}\vec{r^{\prime}}\,|u_{2}(\vec{r})|^{2}|u_{2}(\vec{r^{\prime}})|^{2}\frac{e^{2}}{|\vec{r}-\vec{r^{\prime}}|}, (8) U and \overline{U} are the correlation and exchange integrals U=\int d^{3}\vec{r}\,d^{3}\vec{r^{\prime}}\,u_{1}^{*}(\vec{r})u_{2}^{*}(\vec{r^{\prime}})\frac{e^{2}}{|\vec{r}-\vec{r^{\prime}}|}u_{2}(\vec{r^{\prime}})u_{1}(\vec{r}), (9) and \overline{U}=\int d^{3}\vec{r}\,d^{3}\vec{r^{\prime}}\,u_{1}^{*}(\vec{r})u_{2}^{*}(\vec{r^{\prime}})\frac{e^{2}}{|\vec{r}-\vec{r^{\prime}}|}u_{1}(\vec{r^{\prime}})u_{2}(\vec{r}). (10) The method will be introduced by calculating a well known system of two interacting electrons, the helium atom. In the scheme settled before the one–electron basis set of functions is constrained to just the s and p one–electron hydrogenic states u_{1s}(\vec{r})=\frac{1}{\sqrt{\pi a^{3}}}\exp{\left(-\frac{r}{a}\right)}\chi_{s}\quad\epsilon_{1}=-\frac{2me^{4}}{\hbar^{2}}\quad a=\frac{\hbar^{2}}{2me^{2}} (11) and u_{2s}(\vec{r})=\frac{1}{2\sqrt{2\pi a^{3}}}\left(1-\frac{r}{2a}\right)\exp{\left(-\frac{r}{2a}\right)}\chi_{s}\quad\epsilon_{2}=-\frac{me^{4}}{2\hbar^{2}}, (12) where \chi_{s} , s=\uparrow,\,\downarrow , denotes spin eigenstates and m , e are the electron mass and fundamental charge. The Hamiltonian of the helium atom in Born–Oppenheimer approximation and standard notation H=-\frac{\hbar^{2}}{2m}\nabla_{1}^{2}-\frac{\hbar^{2}}{2m}\nabla_{2}^{2}-\frac{2e^{2}}{r_{1}}-\frac{2e^{2}}{r_{2}}+\frac{e^{2}}{|\vec{r}_{2}-\vec{r}_{1}|} (13) in second quantization formalism reads \displaystyle H= \displaystyle\epsilon_{1}(n_{1\uparrow}+n_{1\downarrow})+\epsilon_{2}(n_{2\uparrow}+n_{2\downarrow})+V_{1}n_{1\uparrow}n_{1\downarrow} (14) \displaystyle+V_{2}n_{2\uparrow}n_{2\downarrow}+U(n_{1\uparrow}n_{2\downarrow}+n_{1\downarrow}n_{2\uparrow}) \displaystyle+\overline{U}(\tilde{\psi}+\tilde{\psi}^{\dagger}-\tilde{\phi}-\tilde{\phi}^{\dagger}), where the operators \tilde{\psi} and \tilde{\phi} are defined by \tilde{\phi}=c_{1\uparrow}^{\dagger}c_{2\downarrow}^{\dagger}c_{2\uparrow}c_{1\downarrow}\qquad\tilde{\psi}=c_{1\uparrow}^{\dagger}c_{1\downarrow}^{\dagger}c_{2\downarrow}c_{2\uparrow}. (15) Making repeated use of the elementary identities \displaystyle[AB,C]=A[B.C]+[A,C]B, (16) \displaystyle[A,BC]=[A,B]C+B[A,C], \displaystyle[AB,C]=A\{B.C\}-\{A,C\}B, \displaystyle[A,BC]=\{A,B\}C-B\{A,C\}, it can be shown that the operators ( 15 ) have the commutation properties \displaystyle[\tilde{\phi},\tilde{\phi}^{\dagger}]=-n_{1\uparrow}n_{2\downarrow}(1-n_{1\downarrow}-n_{2\uparrow})+n_{1\downarrow}n_{2\uparrow}(1-n_{1\uparrow}-n_{2\downarrow}) (17) \displaystyle[\tilde{\psi},\tilde{\psi}^{\dagger}]=n_{1\uparrow}n_{1\downarrow}(1-n_{2\uparrow}-n_{2\downarrow})-n_{2\uparrow}n_{2\downarrow}(1-n_{1\uparrow}-n_{1\downarrow}). One can also demonstrate that \displaystyle[n_{1\uparrow},\tilde{\psi}^{\dagger}]=[n_{1\downarrow},\tilde{\psi}^{\dagger}]=-\tilde{\psi}^{\dagger}, (18) \displaystyle[n_{2\uparrow},\tilde{\psi}^{\dagger}]=[n_{2\downarrow},\tilde{\psi}^{\dagger}]=\tilde{\psi}^{\dagger}, and \displaystyle[n_{1\uparrow}n_{1\downarrow},\tilde{\psi}^{\dagger}]=-\tilde{\psi}^{\dagger}(n_{1\uparrow}+n_{1\downarrow}-1) (19) \displaystyle=-(n_{1\uparrow}+n_{1\downarrow}+1)\tilde{\psi}^{\dagger} \displaystyle[n_{2\uparrow}n_{2\downarrow},\tilde{\psi}^{\dagger}]=\tilde{\psi}^{\dagger}(n_{2\uparrow}+n_{2\downarrow}+1) \displaystyle=(n_{2\uparrow}+n_{2\downarrow}-1)\tilde{\psi}^{\dagger}. Moreover, \displaystyle[n_{1\uparrow}n_{2\uparrow},\tilde{\psi}^{\dagger}]=\tilde{\psi}^{\dagger}(n_{1\uparrow}-n_{2\uparrow}-1)=(n_{1\uparrow}-n_{2\uparrow}+1)\tilde{\psi}^{\dagger}, (20) \displaystyle[n_{1\downarrow}n_{2\downarrow},\tilde{\psi}^{\dagger}]=\tilde{\psi}^{\dagger}(n_{1\downarrow}-n_{2\downarrow}-1)=(n_{1\downarrow}-n_{2\downarrow}+1)\tilde{\psi}^{\dagger}, \displaystyle[n_{1\uparrow}n_{2\downarrow},\tilde{\psi}^{\dagger}]=\tilde{\psi}^{\dagger}(n_{1\uparrow}-n_{2\downarrow}-1)=(n_{1\uparrow}-n_{2\downarrow}+1)\tilde{\psi}^{\dagger}, \displaystyle[n_{1\downarrow}n_{2\uparrow},\tilde{\psi}^{\dagger}]=\tilde{\psi}^{\dagger}(n_{1\downarrow}-n_{2\uparrow}-1)=(n_{1\downarrow}-n_{2\uparrow}+1)\tilde{\psi}^{\dagger}, and [\tilde{\phi},\tilde{\psi}^{\dagger}]=[\tilde{\phi}^{\dagger},\tilde{\psi}^{\dagger}]=0, (21) together with a series of similar equations for the \phi –operators, e. g. [n_{1\uparrow},\tilde{\phi}]=-[n_{2\uparrow},\tilde{\phi}]=-[n_{1\downarrow},\tilde{\phi}]=[n_{2\downarrow},\tilde{\phi}]=\tilde{\phi} . From these commutation properties it readily follows that \displaystyle[H,\tilde{\psi}^{\dagger}]=2\big[-\epsilon_{1}+\epsilon_{2}-V_{1}(n_{1\uparrow}+n_{1\downarrow}+1) (22) \displaystyle+V_{2}(n_{2\uparrow}+n_{2\downarrow}-1) \displaystyle+U(n_{1\uparrow}+n_{1\downarrow}-n_{2\uparrow}-n_{2\downarrow}+2)\big]\tilde{\psi}^{\dagger} \displaystyle+2\overline{U}\big[n_{1\uparrow}n_{1\downarrow}(1-n_{2\uparrow}-n_{2\downarrow})-n_{2\uparrow}n_{2\downarrow}(1-n_{1\uparrow}-n_{1\downarrow})\big], \displaystyle[H,\tilde{\psi}]=2\big[\epsilon_{1}-\epsilon_{2}+V_{1}(n_{1\uparrow}+n_{1\downarrow}-1) (23) \displaystyle-V_{2}(n_{2\uparrow}+n_{2\downarrow}+1) \displaystyle-U(n_{1\uparrow}+n_{1\downarrow}-n_{2\uparrow}-n_{2\downarrow}-2)\big]\tilde{\psi} \displaystyle-2\overline{U}\big[n_{1\uparrow}n_{1\downarrow}(1-n_{2\uparrow}-n_{2\downarrow})-n_{2\uparrow}n_{2\downarrow}(1-n_{1\uparrow}-n_{1\downarrow})\big], where care was taken in Eqs. ( 22 ) and ( 23 ) of placing the \psi –operators at the right side. Notice that all the commutators in the preceding sub–section have the remarkable property of being expressed in terms of just the occupation numbers n_{1s} and n_{2s} of the truncated one–electron basis. Although the occupation numbers are in general not conserved magnitudes, their mean values will be taken here as approximately good quantum numbers. More precisely, it will be assumed that when the number operators n_{k} are applied directly to a stationary state of the system they can be replaced by their classical expectation value. These expectation values are such that they make extremal the mean energy of the system. The scheme proposed here for the He atom is particularly simple because we have only four basis states and the occupation numbers are not expected to depend on the spin. Hence one can define \langle n_{1\uparrow}\rangle=\langle n_{1\downarrow}\rangle=1-\eta,\qquad\langle n_{2\uparrow}\rangle=\langle n_{2\downarrow}\rangle=\eta, (24) where 0\leq\eta\leq 1 is a unique variational parameter. The commutators ( 22 ) and ( 23 ) then become \displaystyle[H,\tilde{\psi}^{\dagger}]=2\big[-\epsilon_{1}+\epsilon_{2}-V_{1}(3-2\eta)-V_{2}(1-2\eta) (25) \displaystyle+4U(1-\eta)\,\big]\tilde{\psi}^{\dagger}+2\overline{U}(1-2\eta)(1-2\eta+2\eta^{2}) and \displaystyle[H,\tilde{\psi}]=2\big[\epsilon_{1}-\epsilon_{2}+V_{1}(1-2\eta)-V_{2}(1+2\eta) (26) \displaystyle+4U\eta\big]\tilde{\psi}-2\overline{U}(1-2\eta)(1-2\eta+2\eta^{2}). Defining now the two operators \psi_{+}^{\dagger}=\tilde{\psi}^{\dagger}+\Lambda_{+}\quad\text{and}\quad\psi_{-}=\tilde{\psi}+\Lambda_{-} (27) where \Lambda_{+}=\frac{-(1-2\eta)(1-2\eta+2\eta^{2})\overline{U}}{-\epsilon_{1}+\epsilon_{2}-V_{1}(3-2\eta)-V_{2}(1-2\eta)+4U(1-\eta)}\,, (28) \Lambda_{-}=\frac{(1-2\eta)(1-2\eta+2\eta^{2})\overline{U}}{\epsilon_{1}-\epsilon_{2}+V_{1}(1-2\eta)-V_{2}(1+2\eta)+4U\eta}\,, (29) the commutators ( 25 ) and ( 26 ) become \displaystyle[H,\psi_{+}^{\dagger}]=2[-\epsilon_{1}+\epsilon_{2}-V_{1}(3-2\eta)-V_{2}(1-2\eta) (30) \displaystyle+4U(1-\eta)]\psi_{+}^{\dagger} and \displaystyle[H,\psi_{-}]=2\big[\epsilon_{1}-\epsilon_{2}+V_{1}(1-2\eta)-V_{2}(1+2\eta) (31) \displaystyle+4U\eta\big]\psi_{-}. This way the operators \psi_{+}^{\dagger} and \psi_{-} are raising and lowering operators. They are not just conjugate operators because the energy levels of the system are not equally spaced, as is for the harmonic oscillator. If H|n\rangle=E_{n}|n\rangle then vectors \psi_{+}^{\dagger}|n\rangle and \psi_{-}|n\rangle are