Abstract
Site-occupation embedding theory (SOET) is an in-principle-exact multi-determinantal extension of density-functional theory for model Hamiltonians. Various extensions of recent developments in SOET (Senjean et al. in Phys Rev B 97:235105, 2018) are explored in this work. An important step forward is the generalization of the theory to multiple-impurity sites. We also propose a new single-impurity density-functional approximation (DFA) where the density-functional impurity correlation energy of the two-level (2L) Hubbard system is combined with the Bethe ansatz local density approximation (BALDA) to the full correlation energy of the (infinite) Hubbard model. In order to test the new DFAs, the impurity-interacting wavefunction has been computed self-consistently with the density-matrix renormalization group method (DMRG). Double occupation and per-site energy expressions have been derived and implemented in the one-dimensional case. A detailed analysis of the results is presented, with a particular focus on the errors induced either by the energy functionals solely or by the self-consistently converged densities. Among all the DFAs (including those previously proposed), the combined 2L-BALDA is the one that performs the best in all correlation and density regimes. Finally, extensions in new directions, like a partition-DFT-type reformulation of SOET, a projection-based SOET approach, or the combination of SOET with Green functions, are briefly discussed as a perspective.
Similar content being viewed by others
References
Pulay P (1983) Chem Phys Lett 100:151
Saebo S, Pulay P (1993) Annu Rev Phys Chem 44:213
Hampel C, Werner H-J (1996) J Chem Phys 104:6286
Sun Q, Chan GK-L (2016) Acc Chem Res 49:2705
Georges A, Kotliar G (1992) Phys Rev B 45:6479
Georges A, Kotliar G, Krauth W, Rozenberg MJ (1996) Rev Mod Phys 68:13
Kotliar G, Vollhardt D (2004) Phys Today 57:53
Held K (2007) Adv Phys 56:829
Zgid D, Chan GK-L (2011) J Chem Phys 134:094115
Hettler M, Tahvildar-Zadeh A, Jarrell M, Pruschke T, Krishnamurthy H (1998) Phys Rev B 58:R7475
Hettler M, Mukherjee M, Jarrell M, Krishnamurthy H (2000) Phys Rev B 61:12739
Lichtenstein A, Katsnelson M (2000) Phys Rev B 62:R9283
Kotliar G, Savrasov SY, Pálsson G, Biroli G (2001) Phys Rev Lett 87:186401
Maier T, Jarrell M, Pruschke T, Hettler MH (2005) Rev Mod Phys 77:1027
Rohringer G, Hafermann H, Toschi A, Katanin A, Antipov A, Katsnelson M, Lichtenstein A, Rubtsov A, Held K (2018) Rev Mod Phys 90:025003
Kotliar G, Savrasov SY, Haule K, Oudovenko VS, Parcollet O, Marianetti CA (2006) Rev Mod Phys 78:865
Sun P, Kotliar G (2002) Phys Rev B 66:085120
Biermann S, Aryasetiawan F, Georges A (2003) Phys Rev Lett 90:086402
Karlsson K (2005) J Phys Condens Matter 17:7573
Boehnke L, Nilsson F, Aryasetiawan F, Werner P (2016) Phys Rev B 94:201106
Werner P, Casula M (2016) J Phys Condens Matter 28:383001
Nilsson F, Boehnke L, Werner P, Aryasetiawan F (2017) Phys Rev Mater 1:043803
Kananenka AA, Gull E, Zgid D (2015) Phys Rev B 91:121111
Lan TN, Kananenka AA, Zgid D (2015) J Chem Phys 143:241102
Zgid D, Gull E (2017) New J Phys 19:023047
Lan TN, Shee A, Li J, Gull E, Zgid D (2017) Phys Rev B 96:155106
Knizia G, Chan GK-L (2012) Phys Rev Lett 109:186404
Knizia G, Chan GK-L (2013) J Chem Theory Comput 9:1428
Bulik IW, Scuseria GE, Dukelsky J (2014) Phys Rev B 89:035140
Zheng B-X, Chan GK-L (2016) Phys Rev B 93:035126
Wouters S, Jiménez-Hoyos CA, Sun Q, Chan GK-L (2016) J Chem Theory Comput 12:2706
Wouters S, Jiménez-Hoyos CA, Chan GK-L (2017) Five years of density matrix embedding theory. In: Fragmentation. Wiley-Blackwell, Chap 8, p 227
Rubin NC (2016) ArXiv preprint arXiv:1610.06910
Tsuchimochi T, Welborn M, Van Voorhis T (2015) J Chem Phys 143:024107
Welborn M, Tsuchimochi T, Van Voorhis T (2016) J Chem Phys 145:074102
Ayral T, Lee T-H, Kotliar G (2017) Phys Rev B 96:235139
Frésard R, Wölfle P (1992) Int J Mod Phys B 6:685
Lechermann F, Georges A, Kotliar G, Parcollet O (2007) Phys Rev B 76:155102
Fromager E (2015) Mol Phys 113:419
Senjean B, Tsuchiizu M, Robert V, Fromager E (2017) Mol Phys 115:48
Senjean B, Nakatani N, Tsuchiizu M, Fromager E (2018) Phys Rev B 97:235105
Chayes J, Chayes L, Ruskai MB (1985) J Stat Phys 38:497
Gunnarsson O, Schönhammer K (1986) Phys Rev Lett 56:1968
Capelle K, Campo VL Jr (2013) Phys Rep 528:91
Hohenberg P, Kohn W (1964) Phys Rev 136:B864
Levy M (1979) Proc Natl Acad Sci 76:6062
Lima N, Oliveira L, Capelle K (2002) Europhys Lett 60:601
Lima NA, Silva MF, Oliveira LN, Capelle K (2003) Phys Rev Lett 90:146402
Capelle K, Lima N, Silva M, Oliveira L (2003) The fundamentals of electron density, density matrix and density functional theory in atoms molecules and the solid state. Springer, Berlin
Yamada K (1975) Prog Theor Phys 53:970
Carrascal DJ, Ferrer J, Smith JC, Burke K (2015) J Phys Condens Matter 27:393001
Carrascal DJ, Ferrer J, Smith JC, Burke K (2016) J Phys Condens Matter 29:019501
White SR (1992) Phys Rev Lett 69:2863
White SR (1993) Phys Rev B 48:10345
Verstraete F, Murg V, Cirac JI (2008) Adv Phys 57:143
Schollwöck U (2011) Ann Phys 326:96
Nakatani N. https://github.com/naokin/mpsxx
Kim M-C, Sim E, Burke K (2013) Phys Rev Lett 111:073003
Akande A, Sanvito S (2010) Phys Rev B 82:245114
Lieb EH, Wu FY (1968) Phys Rev Lett 20:1445
Dimitrov T, Appel H, Fuks JI, Rubio A (2016) New J Phys 18:083004
Xianlong G, Chen A-H, Tokatly I, Kurth S (2012) Phys Rev B 86:235139
Kurth S, Stefanucci G, Khosravi E, Verdozzi C, Gross E (2010) Phys Rev Lett 104:236801
Karlsson D, Privitera A, Verdozzi C (2011) Phys Rev Lett 106:116401
Ying Z-J, Brosco V, Lorenzana J (2014) Phys Rev B 89:205130
Elliott P, Burke K, Cohen MH, Wasserman A (2010) Phys Rev A 82:024501
Richardson R (1963) Phys Lett 3:277
Richardson R, Sherman N (1964) Nucl Phys 52:221
Limacher PA (2016) J Chem Phys 145:194102
Lieb EH (1983) Int J Quantum Chem 24:243
Dagotto E (1994) Rev Mod Phys 66:763
Acknowledgements
E.F. would like to dedicate this work to the memory of János G. Ángyán. He would also like to thank Andreas Savin for a fruitful discussion on the train from Middelfart to Copenhagen. B.S. thanks D. Carrascal for taking the time to check the parameterization of his Hubbard dimer functional, and M. Saubanère, L. Mazouin, and K. Deur for fruitful discussions. This work was funded by the Ecole Doctorale des Sciences Chimiques 222 (Strasbourg), the ANR (MCFUNEX project, Grant No. ANR-14-CE06- 0014-01), the “Japon-Unistra” network as well as the Building of Consortia for the Development of Human Resources in Science and Technology, MEXT, Japan for travel funding.
Author information
Authors and Affiliations
Corresponding author
Additional information
Published as part of the special collection of articles In Memoriam of János Ángyán.
Appendices
Appendix 1: Exact embedding potential at half-filling for multiple impurities
Let us consider any density \({\mathbf {n}}\equiv \lbrace n_i\rbrace _i\) summing up to a number \(N=\sum _in_i\) of electrons. Under hole-particle symmetry, this density becomes \((\underline{2} - {\mathbf {n}})\equiv \lbrace 2-n_i\rbrace _i\) and the number of electrons equals \(2L-N\) where L is the number of sites. We will prove that these two densities give the same correlation energy for the M-impurity-interacting system. Since, for any local potential \({\mathbf {v}}\), the variational principle in Eq. (5) reads as follows for an impurity-interacting system,
which gives, for any density \({\mathbf {n}}\),
thus leading to the Legendre–Fenchel transform expression,
By applying a hole-particle symmetry transformation to Eq. (67) [we will now indicate the number of particles in the impurity-interacting energies for clarity], we obtain
where \({\mathcal {E}}^{{\mathrm{imp}},2L-N}_M({\mathbf {v}})\) is the (\(2L-N\))-particle ground-state of the following M-impurity-interacting Hamiltonian:
Applying the hole-particle transformation to the creation and annihilation operators,
to the M-impurity-interacting Hamiltonian in Eq. (69) leads to
or, equivalently,
Then, by substituting and shifting the potential as follows,
we finally obtain
As readily seen from Eqs. (69) and (74), the \((2L-N)\)-electron ground-state energy \({\mathcal {E}}^{{\mathrm{imp}},2L-N}_M({\mathbf {v}})\) of \({\hat{H}}^{\mathrm{imp}}_M({\mathbf {v}})\) is connected to the N-electron ground-state energy \({\mathcal {E}}^{\mathrm{imp},N}_M(\mathbf {\tilde{v}})\) of \({\hat{H}}^\mathrm{imp}_M(\mathbf {\tilde{v}})\) by
Introducing Eq. (75) into Eq. (68) leads to
Note that the maximizing potential in Eq. (76), denoted by \(\tilde{v}^{\mathrm{emb}}_M({\mathbf {n}})\), is nothing but the exact embedding potential \(v^{\mathrm{emb}}_M({\mathbf {n}})\) which restores the exact density profile \({\mathbf {n}}\), by definition:
According to the shift in Eq. (73), this maximizing potential is related to the maximizing one in Eq. (68), denoted by \({\mathbf {v}}^\mathrm{emb}_M(\underline{2} - {\mathbf {n}})\), by
From equality (77), it comes
thus leading to, at half-filling,
Appendix 2: Fundamental relation between derivatives in t and U of the complementary bath per-site correlation energy for multiple impurities
If we denote \({\mathbf {v}}_M^{\mathrm{emb}}({\mathbf {n}})\) the maximizing potential in the Legendre–Fenchel transform of Eq. (67), we deduce from the linearity in t and U of the impurity-interacting Hamiltonian that [the dependence in t and U is now introduced for clarity]
thus leading to the fundamental relation
as a consequence of the stationarity condition fulfilled by \({\mathbf {v}}_M^{\mathrm{emb}}({\mathbf {n}})\). Since both the non-interacting kinetic energy [which is obtained when \(U=0\)] and the impurity Hx functional [first term in the right-hand side of Eq. (19)] fulfill the same relation, we conclude from the decomposition in Eq. (15) that
We finally obtain, by combining Eqs. (23), (83) and (101), the fundamental relation in Eq. (40).
Appendix 3: Lieb maximization and correlation energy derivatives for a single impurity
The impurity-interacting LL functional in Eq. (7) [we consider the particular case of a single impurity (\(M=1\)) in the following] can be rewritten as a Legendre–Fenchel transform [39, 40],
where \({\mathcal {E}}^{\mathrm{imp}}(t,U,{\mathbf {v}})\) is the ground-state energy of \({\hat{T}}+U{\hat{n}}_{0\uparrow }{\hat{n}}_{0\downarrow }+\sum _i v_i {\hat{n}}_i\). Note that the dependence in both t and U of \(F^{\mathrm{imp}}({\mathbf {n}})\) and \({\mathcal {E}}^\mathrm{imp}({\mathbf {v}})\) has been introduced for clarity. The so-called Lieb maximization [70] procedure described in Eq. (84) has been used in this work in order to compute accurate values of \(F^{\mathrm{imp}}(t,U,{\mathbf {n}})\) and \(T_\mathrm{s}(t,{\mathbf {n}})=F^{\mathrm{imp}}(t,U=0,{\mathbf {n}})\) for a 8-site ring. The impurity-interacting energy \({\mathcal {E}}^\mathrm{imp}(t,U,{\mathbf {v}})\) has been obtained by performing an exact diagonalization calculation based on the Lanczos algorithm [71]. The impurity correlation energy is then obtained as follows,
Since \(\partial F^{\mathrm{imp}}(t,U,{\mathbf {n}})/\partial U=d^\mathrm{imp}(t,U,{\mathbf {n}})\) is the impurity site double occupation obtained for the maximizing potential in Eq. (84) (see Eq. (30) and Eq. (A5) in Ref. [41]), it comes from Eq. (85),
Moreover, since
is the impurity-interacting kinetic energy obtained for the maximizing potential in Eq. (84) (see Eq. (35) and Eq. (B6) in Ref. [41]), which gives in the non-interacting case \(t\,\partial T_\mathrm{s}(t,{\mathbf {n}})/\partial t=T_\mathrm{s}(t,{\mathbf {n}})\), we recover from Eq. (85) the expression in Eq. (B8) of Ref. [41],
which can be further simplified as follows,
Interestingly, the derivatives in t and U are connected as follows, according to Eq. (86),
Thus we recover Eq. (83) in the particular single-impurity case.
Similarly, in the fully interacting case, the LL functional can be rewritten as follows, as a consequence of Eq. (5),
where the t- and U-dependence in both \(F({\mathbf {n}})\) and \(E({\mathbf {v}})\) is now made explicit. From the correlation energy expression,
and the expressions for the LL functional derivatives in t and U [those and their above-mentioned impurity-interacting analogs are deduced from the Hellmann–Feynman theorem],
and
it comes
and
Note that \(d_i(t,U,{\mathbf {n}})\) and \(T (t,U,{\mathbf {n}})\), which have been introduced in Eqs. (93) and (94), denote the site i double occupation and the total (fully interacting) kinetic energy, respectively. Both are calculated for the maximizing potential in Eq. (91). For a uniform density profile \(\underline{n}\), the per-site correlation energy reads
Since, in this case, \(d_i(t,U,\underline{n})=d(t,U,n)\) is site-independent, we finally obtain from Eqs. (94), (95), and (96),
and
By analogy with Eq. (89), the latter expression can be simplified as follows,
or, equivalently (see Eq. (98)),
Appendix 4: Derivatives of BALDA
1.1 Derivative with respect to U and t
As readily seen in Eq. (31), the derivative of the complementary bath per-site correlation energy functional with respect to U is necessary to compute double occupation in SOET. According to Eq. (24), it implies the derivative of the conventional per-site correlation energy, modeled with BALDA, which reads
and then for \(n > 1\):
where \(\partial \beta (U/t) / \partial U = ( \partial \beta (U/t) / \partial (U/t) )/t\), is computed with finite differences by solving Eq. (47) for \(\beta (U/t)\).
The derivative with respect to t is calculated according to Eq. (101).
1.2 Derivative with respect to n
To get the correlation embedding potential, the derivatives of the correlation functionals with respect to n is necessary. The derivative of the convention per-site density-functional correlation energy reads
and
Appendix 5: Derivatives of SIAM-BALDA
The derivatives of the SIAM-BALDA impurity correlation functional [Eq. (51)] are given with respect to U for \(n \leqslant 1\) as follows,
The derivative with respect to t is given according to Eq. (90). Then, the impurity correlation potential is determined by the derivative of the functional with respect to the occupation number n:
where
and
If \(n > 1\), the particle-hole formalism imposes to use \(\Gamma (t,2-n)\) instead of \(\Gamma (t,n)\). The derivatives with respect to n should be changed accordingly.
Appendix 6: Derivatives of 2L-BALDA
1.1 Parametrization of the correlation energy of the dimer
In this section, we summarize the parametrization of the Hubbard dimer correlation energy by Carrascal and co-workers [51, 52], necessary to understand the following derivations. The equations coming from their paper are referred to as (& N), where N is the number of the equation. We start from the definition of the correlation energy, where n is the occupation of the site 0 and \(u = U/2t\) is a dimensionless parameter,
where 2L refers to “two-level”, and
To account for particle-hole symmetry of the functional, the variable \(\rho = | n-1 |\) is used rather than n directly. We now simply follow the guidelines from Eq.(& 102) to (& 107), leading to
and
Then, they proposed a first approximation to \(g(\rho ,u)\), denoted by the label 0:
where
and
Plugging \(g=g_0(\rho ,u)\) into \(f(g,\rho )\) leads to the first parametrization of \(E_\mathrm{c}^{2L}(n)\) in Eq. (110). In this work, we implemented the more accurate parametrization, given in Eq.(& 114) [52]:
and where \(q(\rho ,u)\) is given in Eq. (& 115) by [52]:
The accurate pametrization of \(E_\mathrm{c}^{2L}(n)\) is obtained by plugging this \(g_1(\rho ,u)\) into \(f(g,\rho )\), instead of \(g_0(\rho , u)\).
In order to obtain the impurity correlation energy, a simple scaling of the interaction parameter U has to be applied on the conventional correlation energy, as demonstrated in Ref. [40] and given in Eq. (55), leading to
1.2 Derivative with respect to U and t
We compute the derivative with respect to the dimensionless parameter \(u = U/2t\). The \(\rho\)- and u- dependence of \(g(\rho , u)\) will be omitted for readability. Besides, many functions will be introduced, aiming to make the implementation and its numerical verification easier. Starting with
the impurity correlation functional reads, according to Eq. (119),
with
The derivative of \(h(g,\rho )\) is quite easy, as its only u-dependence is contained in g, so that:
with
where the function \(Y(g,\rho ) = \sqrt{1 - g^2 - \rho ^2}\) has been introduced. For the first approximation, \(g = g_0\) and
where \(G(\rho ,u) = N(\rho ,u)/D(\rho ,u)\) and
and
Their respective derivative with respect to u reads
and
with
Turning to the second approximation \(g = g_1\) implemented in this work, one get from the derivative of Eq. (117),
For convenience, we introduce two functions w(g, u) and v(g, u) so that
with
and
Finally, the last term in Eq. (131) reads, for \(q(\rho ,u) = j(\rho ,u)k(\rho ,u)/l(\rho ,u)\):
with
and
and their derivative with respect to u:
and
The derivative with respect to t is given according to Eq. (90).
1.3 Derivative with respect to n
Regarding the derivative with respect to n which is necessary to get the embedded correlation potential, it comes
where \(\partial \rho / \partial n = {\text {sign}}(n-1)\) and
We start with
where, for the first parametrization using \(g = g_0(\rho ,u)\),
with
and
Then, the right term in the right-hand side of Eq. (146) is derived as:
Turning to the second parametrization \(g = g_1\), the derivative with respect to \(\rho\) leads to
with
Finally,
with
and
Rights and permissions
About this article
Cite this article
Senjean, B., Nakatani, N., Tsuchiizu, M. et al. Multiple impurities and combined local density approximations in site-occupation embedding theory. Theor Chem Acc 137, 169 (2018). https://doi.org/10.1007/s00214-018-2368-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00214-018-2368-z