Zintl superalkalis as building blocks of supersalts

0 downloads 0 Views 3MB Size Report
Keywords Zintl ion . Ionization energy . Superalkali . Electron affinity . Superhalogens . DFT . Supersalts. Introduction. Large ensembles of bound atoms known ...
Journal of Molecular Modeling (2018) 24:306 https://doi.org/10.1007/s00894-018-3806-5

ORIGINAL PAPER

Zintl superalkalis as building blocks of supersalts G. Naaresh Reddy 1 & A. Vijay Kumar 1,2 & Rakesh Parida 1 & Arindam Chakraborty 3 & Santanab Giri 1 Received: 24 May 2018 / Accepted: 21 August 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract Alkali metal cations and halogen anions are common components of ionic salts. Recently, a new class of salts termed supersalts was reported, each of which contains a superalkali and a superhalogen that mimic an alkali metal cation and a halogen anion, respectively. Using three different functionals, namely B3LYP, wB97X, and M06-2X, we theoretically investigated a new subset of supersalts composed of Zintl-based superalkalis and inorganic superhalogens via computational modeling. The calculated dipole moment and first-order hyperpolarizability values for these supersalts indicate that they present nonlinear optical (NLO) behavior. The supersalts of Zintl superalkalis (Ca2P7, Sr2P7, Ba2P7) and superhalogens (BF4, BeF3, NO3) studied here were found to be stable. Keywords Zintl ion . Ionization energy . Superalkali . Electron affinity . Superhalogens . DFT . Supersalts

Introduction Large ensembles of bound atoms known as atomic clusters are currently of great interest to experimentalists as well as theoreticians. Such clusters represent a new phase of matter with properties that are governed by the size and composition of the cluster as well as the geometrical arrangement of the atoms within it [1]. The discovery of a new hypervalent atomic cluster with a lower ionization energy than that of an alkali metal was reported more than three decades ago by Gutsev and This paper belongs to Topical Collection International Conference on Systems and Processes in Physics, Chemistry and Biology (ICSPPCB2018) in honor of Professor Pratim K. Chattaraj on his sixtieth birthday Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00894-018-3806-5) contains supplementary material, which is available to authorized users. * Arindam Chakraborty [email protected] * Santanab Giri [email protected] 1

Department of Chemistry, National Institute of Technology Rourkela, Odisha 769008, India

2

Department of Chemistry, Utkal University, Vani Vihar Bhubaneswar, Odisha 751004, India

3

Faculty of Science, Jatragachi Pranabananda High School, New Town, Kolkata 700161, India

Boldyrev [2]. Boldyrev denoted atomic clusters of this type Bsuperalkalis,^ and proposed a general formula of XMn + 1 for them, where M represents an alkali metal atom and X is an electronegative atom of valence n. In another work, Gutsev and Boldyrev [3] showed that atomic clusters of general formula MXk + 1 (where M is a metal atom of valence k and X is a halogen atom) mimic halogen atoms and present much stronger electron affinities than halogens. These atomic clusters were subsequently termed Bsuperhalogens.^ The introduction of superalkalis and superhalogens opened up a new field of fundamental research in experimental and computational chemistry. Typical examples of superalkalis include FLi2, OLi3, and NLi4 [4]. Experimental studies advocating the existence of superalkalis [5–14] and superhalogens [15–18] have already been reported. While superalkalis exhibit strong reducing properties, superhalogens act as weakly coordinating strong oxidants. Superalkalis can be used to synthesize a variety of charge-transfer salts by coordinating them with species that have low electron affinities. Superalkalis and superhalogens have attracted considerable research interest over the past decade, and are now considered to be the basic building blocks for a new class of molecules, the supersalts [19–23]. Atomic and molecular clusters can be used to tailor materials with specific properties [24, 25], and superalkalis can be employed in a similar manner to produce tailored supersalts [26–29]. A recent work [30] reported the existence of a new class of superalkalis in which a Zintl ion [31–36] core is coordinated

306

Page 2 of 13

J Mol Model (2018) 24:306

section. The section after that reports and discusses relevant findings of our investigation. In the final section, we draw important conclusions from our work.

Computational details Fig. 1 Optimized geometry of the P73− Zintl ion

with organic ligands, yielding organo-Zintl superalkali clusters. In the present paper, we describe our use of computational modeling to design supersalts by ligating Zintl superalkali molecules with some common inorganic superhalogens such as BeF3, BF4, and NO3. The P73− core of the Zintl ion was coordinated with alkaline earth metals to build M2P7 (M = Ca, Sr, and Ba) superalkali clusters, which were subsequently bonded with superhalogens to produce stable supersalts. Our investigation is of particular relevance given a recent report [37] of the synthesis of Ba2P7X (X = Cl, Br, I) Zintl salts in which a P73− core is coordinated with two Ba atoms to yield the Zintl superalkali Ba2P7. The present paper is organized as follows. Details of the computational methodology we used are given in the next Fig. 2 Optimized geometries of the neutral and cationic clusters [P7M2] (M = Ca, Sr, Ba)

The studied molecules were optimized and calculations were performed at various levels of theory, including the use of the Becke three-parameter Lee–Yang–Parr (B3LYP) [38, 39], wB97X [40], and M06-2X [41] functionals. The basis set def2-TZVPP (default 2 triple-zeta valence plus polarization function) [42] was employed for all atoms at all levels of theory. No symmetry constraints were applied during geometry optimization. All calculations were performed using the Gaussian 09 program [43]. Projected density of states (PDOS) calculations were carried out using the GaussSum software package [44]. Electron affinity (EA) values were calculated in order to study the superhalogen behavior of the molecules, and were obtained as the difference between the energy of the anion in its ground state and that of the corresponding optimized neutral molecule. The ionization energy (IE) values of Zintl clusters were calculated as the difference

J Mol Model (2018) 24:306

Page 3 of 13 306

Table 1 Ionization energies (IEs) of the clusters [P7M2] (M = Ca, Sr, Ba) calculated at the B3LYP, M06-2X, and wB97X levels of theory using the Def2-TZVPP basis set

S. no.

Molecule

Ionization energy (IE, in eV) B3LYP/def2-TZVPP

M06-2X/def2-TZVPP

wB97X/def2-TZVPP

1

P7Ca2

4.58

4.60

3.86

2 3

P7Sr2 P7Ba2

4.60 4.35

4.42 4.20

4.50 4.20

S. no. represents Serial number (Sl. No.)

between the energy of the neutral molecule and the energy of the corresponding cation in the ground state. Natural bond orbital (NBO) [45] analysis was performed to determine the nature of the bonding in the complexes. The binding energy (BE) values of supersalts were evaluated using the following relation: Binding energy (BE) = E(superalkali cation (P7M2+(M = Ca, Sr, Ba)) + (E(superhalogen anion (Y−(Y = BeF3, BF4, NO3))) – E(supersalt)). Theoretical investigations of some molecular parameters were carried out in order to explore the origin of the nonlinear behavior of the studied systems at the microscopic level. The study involved calculating some common parameters such as the dipole moment (μ), polarizability (α), and first hyperpolarizability (β) tensors for the supersalts. The first hyperpolarizability is a third-rank tensor that can be described by a 3 × 3 × 3 matrix. Due to Kleinman symmetry, the 27 components of the 3D matrix can be reduced to ten

β¼

h

βxxx þ β xyy þ β xzz

2

components. The Gaussian 09 output provides these ten components [46, 47] as βxxx, βxxy, βxyy, βyyy, βxxz, βxyz, βyyz, βxzz, βyzz, and βzzz. However, theoretical chemists are mainly concerned with β||, which is the component parallel to the direction of charge transfer in the ground state, as well as the total hyperpolarizability βtot. Using the x, y, and z components, the magnitudes of the total static dipole moment (μ), the isotropic polarizability (α), and the first-order hyperpolarizability tensor β can be calculated via the following equations:   Dipole moment ¼ μ0 ¼ μx 2 þ μy 2 þ μz 2   : Static polarizability ¼ α0 ¼ αxx þ αyy þ αzz =3  1=2 First order hyperpolarizability ¼ β ¼ βx þ βy þ βz The complete equation for calculating the magnitude of the first hyperpolarizability (β) from the Gaussian 09 output is given below:

 2  2 i1=2 þ βyyy þ βyzz þ β yxx þ βzzz þ βzxx þ β zyy :

structure (C3v point group). The optimized geometry of the P73− Zintl ion is shown in Fig. 1.

Results and discussion [P7M2] (M = Ca, Sr, and Ba) Zintl superalkali clusters We first considered the well-known Zintl ion P73− [48]. The P7 core is a 35-electron system that requires three more electrons to attain a stable electronic shell configuration. P73− is a group 15 Zintl ion (E73−, E = P, As, Sb) that has a nortricyclane-like

[P7M2] (M = Ca, Sr, Ba) Zintl superalkalis clusters were obtained by ligating the P73− Zintl ion core with alkaline earth metals (Ca, Sr, and Ba). All of the [P7M2] complexes considered in this

Fig. 3 Optimized geometries of the neutral and anionic forms of BeF3, BF4, and NO3

306

J Mol Model (2018) 24:306

Page 4 of 13

Table 2 Calculated electron affinities (EAs) of BeF3, BF4, and NO3 obtained at the B3LYP, M06-2X, and wB97X levels of theory using the def2-TZVPP basis set

S. no.

Molecule

Electron affinity (EA, in eV) B3LYP/def2-TZVPP

M06-2X/def2-TZVPP

wB97X/def2-TZVPP

1

BeF3

6.30

6.72

6.43

2 3

BF4 NO3

6.90 3.80

7.30 4.24

7.00 3.97

work have 39 valance electrons, so they would be expected to readily eject one electron in order to be able to adopt a stable octet geometry. Thus, the neutral complexes behave like alkalis and become stable cations, in contrast to alkali and alkaline earth metals. The optimized geometries of the [P7M2] (M = Ca, Sr, Ba) superalkali clusters and their corresponding cations are shown in Fig. 2. From this figure, it is clear that the neutral and cationic forms of the P7Ca2 molecule adopt the same geometry, unlike the P7Sr2 and P7Ba2 clusters, where the metal atoms are positioned differently in the optimized cationic and Fig. 4 Optimized geometries and the relative energies of the [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) supersalts

neutral geometries. The calculated ionization energy (IE) values of the [P7M2] (M = Ca, Sr, Ba) clusters obtained at three different levels of theory are given in Table 1. Similar values were attained at all levels of theory. This clearly shows that all of the clusters have IE values that are lower than those of Li (5.39 eV) and Na (5.14 eV), the signature alkali metals. Thus, the low IE values of the [P7M2] (M = Ca, Sr, Ba) Zintl clusters fit well with their application as superalkalis. We also optimized both the neutral clusters and their cations with different spin multiplicities to determine their stabilities. Based on the calculated

−0.90 −0.82 −0.94 0.91 0.94 0.95 0.89 0.92 0.93

−0.91 −0.94 −0.95

B3LYP/def2TZVPP M06-2X/def2TZVPP B3LYP/def2TZVPP

wB97X/def2TZVPP Anion (BeF3, BF4, NO3)

M06-2X/def2TZVPP

Page 5 of 13 306

Cation (P7Ca2)

relative energies of different states, we found the doublet to be more stable for the neutral cluster and the singlet to be more stable for the cation. The optimized geometries of, the Cartesian coordinates of the atoms in, and the HOMOs and LUMOs of [P7M2] (M = Ca, Sr, Ba) are given in Figures S1–S2 of the BElectronic supplementary material^ (ESM). Different isomers of [P7Ca2] and [P7Ca2]+ are given in Figure S3 of the ESM. The optimized geometries of the neutral and cationic clusters [P7M2] (M = Ca, Sr, Ba) in different spin states along with their relative energies are also given in Figure S4 of the ESM.

−0.89 −0.92 −0.93 0.90 0.82 0.94 6.291 6.183 5.700 6.303 6.184 5.724

M062X/def2TZVPP B3LYP/def2TZVPP

The optimized geometries of the superhalogens BeF3, BF4, and NO3 in their neutral and monocharged anionic forms, as calculated at three different levels of theory, are shown in Fig. 3. The optimized geometries of and the Cartesian coordinates of the atoms in BeF3, BF4, and NO3 are given in Figure S5 of the ESM, while the calculated electron affinity (EA) values of BeF3, BF4, and NO3 are given in Table 2. BeF3 and BF4 are conventional superhalogens as they satisfy the formula MXk + 1 (where M represents the metal center, X represents the halogen, and k is the valence of the metal center). NO3, on the other hand, is an unconventional superhalogen as it contains nether a metal nor a halogen atom. Figure 3 clearly shows that the molecular geometries of the neutral and anionic forms are the same for BeF3, BF4, and NO3. The EA values of BeF3, BF4 and NO3 show only a minor dependence on the level of theory used and are higher than those of halogen atoms (in good agreement with previously reported data [49, 50]), thereby warranting their use as superhalogen molecules.

The supersalts [(P7M2)(NO3)], [(P7M2)(BeF3)], and [(P7M2)(BF4)] (M = Ca, Sr, Ba)

[(P7Ca2)(NO3)] 6.15 [(P7Ca2)(BeF3)] 6.02 [(P7Ca2)(BF4)] 5.50

wB97X/def2TZVPP

The superhalogens

wB97X/def2TZVPP

NBO charge separation Binding energy (in eV) Supersalt

Table 3 Calculated binding energies (BEs) and NBO charge separations of [(P7Ca2)Y] (Y = NO3, BeF3, BF4) supersalts obtained at the B3LYP, M06-2X, and wB97X levels of theory using the def2TZVPP basis set

J Mol Model (2018) 24:306

[(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) The optimized geometries of the [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4,) supersalts are represented in Fig. 4. The optimized geometries of and the Cartesian coordinates of the atoms in [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) are given in Figure S6 of the ESM. The binding energy (BE) values of the [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) supersalts and the associated natural bond orbital (NBO) charge distributions over the cationic superalkalis and anionic superhalogens are given in Table 3. The calculated PDOS diagrams are portrayed in Fig. 5 and the core and ligand contributions to the frontier molecular orbitals (FMOs) of the [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) complexes are given in supporting information Table S3. From Fig. 4, it is clear that BeF3 and BF4 bond with the P7Ca2 superalkali in two different orientations, leading to two isomeric forms of each supersalt. However, the [(P7Ca2)(NO3)] complex does not present multiple isomeric forms. It is also apparent that

306

Page 6 of 13

J Mol Model (2018) 24:306

Fig. 5 Calculated PDOS diagrams of [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) supersalts, as obtained at the B3LYP/def2-TZVPP level of theory. The cation (P7Ca2), the anion (NO3, BeF3, BF4), and the total DOS are shown in green, red, and blue, respectively

in both the [(P7Ca2)(BeF3)] and [(P7Ca2)(BF4)] supersalts, isomer 1 is more stable than isomer 2. This may be attributed to the fact that the incoming superhalogen (BeF3 or BF4) prefers to bind to just one of the two Ca centers of the superalkali in isomer 1 but to both Ca sites in isomer 2.

The interaction of the superhalogen with both Ca centers in isomer 2 probably leads to some strain or distortion in the superhalogen, which leads to a relatively unstable supersalt. Natural bond orbital (NBO) analysis of the [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) supersalts shows

J Mol Model (2018) 24:306

Page 7 of 13 306

Fig. 6 Optimized geometries and the relative energies of the [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4) supersalts

that the calculated NBO charge on the P7Ca2 system is close to +1, which is balanced by an equal and opposite negative charge on the NO3, BeF3, or BF4 in the supersalt. This charge distribution pattern clearly indicates the presence of strong ionic bonding between the P 7 Ca 2 superalkali and the NO3, BeF3, or BF4 superhalogen. The P 7 Ca 2 molecule loses an electron to form the P 7 Ca 2 +

superalkali cation, while the NO3, BeF3, or BF4 molecule gains an electron to yield the superhalogen anion NO3−, BeF3−, or BF4−. The calculated binding energy values of the [(P7M2)Y] (M = Ca, Y = NO3, BeF3, BF4) supersalts are positive, implying that these are stable systems. A comprehensive picture of the bonding in the supersalts can be obtained from density of states (DOS) and projected

Table 4 Calculated binding energies (BEs) and NBO charge separations of [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4), as calculated at the B3LYP, M06-2X, and wB97X levels of theory using the def2-TZVPP basis set System

Binding energy (in eV)

NBO charge separation Cation (P7Sr2)

B3LYP/ def2TZVPP [(P7Sr2)(NO3)] 5.75 [(P7Sr2)(BeF3)] 5.65 [(P7Sr2)(BF4)] 5.15

Anion (BeF3, BF4, NO3)

M062X/ wB97X/ B3LYP/ def2-TZVPP def2-TZVPP def2TZVPP

M06-2X/ wB97X/ B3LYP/ def2-TZVPP def2-TZVPP def2TZVPP

5.88 5.80 5.30

0.93 0.95 0.96

5.90 5.83 5.34

0.91 0.94 0.95

0.93 0.95 0.96

−0.91 −0.94 −0.95

M06-2X/ wB97X/ def2-TZVPP def2-TZVPP −0.93 −0.95 −0.96

−0.93 −0.95 −0.96

306

Page 8 of 13

J Mol Model (2018) 24:306

Fig. 7 Calculated PDOS diagrams of [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4) supersalts, as obtained at the B3LYP/def2-TZVPP level of theory. The cation (P7Sr2), the anion (NO3, BeF3, or BF4), and the total DOS are shown in green, red, and blue, respectively

density of states (PDOS) calculations, which highlight the contributions of [P7Ca2] (superalkali) and NO3, BeF3, and BF4 (superhalogens) to the frontier molecular orbitals of the supersalts. The plots for the [(P7Ca2)Y] (Y = NO3, BeF3, BF4) complexes and the tables of percentage contributions shown in Fig. 5 reveal that in almost all cases, the major contributor

to both the HOMO and the LUMO is the [P7Ca2] superalkali. However, there are notable contributions from the ligand to higher molecular orbitals such as LUMO+5, LUMO+4, etc. The HOMO–LUMO gap was found to gradually increase for the supersalts in the following order: [(P 7 Ca 2 )(NO 3 )] < [(P7Ca2)(BeF3)] < [(P7Ca2)(BF4)].

J Mol Model (2018) 24:306

Page 9 of 13 306

Fig. 8 Optimized geometries and the relative energies of [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) supersalts

[(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4)

molecular orbitals (FMOs) of the [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4) supersalts are given in supporting information Table S3. For each system, the molecular geometry of isomer 2 is observed to be slightly distorted than that of isomer 1. Thus, the less distorted isomer 1 is thought to be less strained and hence more stable. The calculated relative energies of the isomers mimic this trend and indicate that isomer 1 is more stable than isomer 2. The calculated NBO charge on the P7Sr2 superalkali system and those on the NO3, BeF3, and BF4 subunits in the [(P 7M 2)Y] (M = Sr, Y = NO3, BeF3, BF4) supersalts present a similar trend to that noted for their Ca-based counterparts. Thus, strong ionic

The optimized molecular geometries of the plausible isomers of the supersalts [(P 7 Sr 2 )(NO 3 )], [(P 7 Sr 2 )(BeF 3 )], and [(P7Sr2)(BF4)] are depicted in Fig. 6 along with their relative energies. The optimized geometries of and the Cartesian coordinates of the atoms in [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4) are given in Table S6 in the ESM. The calculated binding energy (BE) values of and the NBO charges on the cationic and anionic superalkalis and superhalogens are given in Table 4, while the calculated PDOS are elaborated in Fig. 7 diagrams and the core and the ligand contributions to the frontier

Table 5 Binding energies (BEs) and NBO charge separations of [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) as calculated at the B3LYP, M06-2X, and wB97X levels of theory using the def2-TZVPP basis set System

Binding energy (in eV)

NBO charge separation Cation (P7Ba2)

B3LYP/ M062X/ wB97X/ B3LYP/ M06-2X/ def2-TZVPP def2-TZVPP def2-TZVPP def2-TZVPP def2-TZVPP (P7Ba2)(NO3) 5.30 (P7Ba2)(BeF3) 5.19 (P7Ba2)(BF4) 4.81

5.42 5.35 5.03

5.45 5.40 5.06

0.93 0.95 0.96

0.94 0.96 0.97

Anion (BeF3, BF4, NO3) wB97X/ B3LYP/ M06-2X/ def2-TZVPP def2-TZVPP def2-TZVPP

wB97X/ def2-TZVPP

0.94 0.96 0.97

−0.94 −0.96 −0.97

−0.93 −0.95 −0.96

−0.94 −0.96 −0.97

306

Page 10 of 13

J Mol Model (2018) 24:306

Fig. 9 Calculated PDOS diagrams of [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) supersalts, as obtained at the B3LYP/def2-TZVPP level of theory. The cation (P7Ba2 ), the anions (NO3, BeF3, BF4), and the total DOS are shown in green, red, and blue, respectively

bonding appears to exist in the [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4) systems, with the P7Sr2+ superalkali cation binding with a NO3−, BeF3−, or BF4− anion to form a

stable ion pair. The stability of each ion pair can be further rationalized using their positive binding energy (ΔE) values.

J Mol Model (2018) 24:306 Table 6 Values of the dipole moment μt (in debyes), polarizability αt (in 0.1482 × 10−24 esu), and hyperpolarizability βt (in 0.0086393 × 10−30 esu) for the [(P7M2)Y] (M = Ca, Sr, Ba, and Y = BeF3, BF4, NO3) supersalts

Page 11 of 13 306

Salt

Dipole moment (in debyes)

Polarizability(αt) (in esu)

First-order hyperpolarizability(βt) (in esu)

P7Ca2BeF3 P7Ca2BF4 P7Ca2NO3 P7Sr2BeF3 P7Sr2BF4 P7Sr2NO3 P7Ba2BeF3 P7Ba2BF4 P7Ba2NO3

13.01 12.38 12.22 14.56 15.22 13.91 15.64 15.05 14.96

36.4560 × 10−24 36.6172 × 10−24 38.1570 × 10−24 39.0714 × 10−24 39.3056 × 10−24 40.9002 × 10−24 41.9035 × 10−24 41.9880 × 10−24 43.9860 × 10−24

1.0223 × 10−30 1.0341 × 10−30 1.0671 × 10−30 1.1804 × 10−30 1.1673 × 10−30 1.2319 × 10−30 1.2118 × 10−30 1.2041 × 10−30 1.2795 × 10−30

The calculated density of states (DOS) and projected density of states (PDOS) data shown in the plots for [(P7M2)Y] (M = Sr, Y = NO3, BeF3, BF4) in Fig. 7, which also tabulates the percentage contributions, depict a similar trend to that seen for their Ca counterparts. The main contributor to both the HOMO and the LUMO is the P7Sr2 superalkali. The ligand, however, makes a notable contribution to higher molecular orbitals such as LUMO+5 and LUMO+4, etc. The HOMO–

LUMO gap gradually increases along the sequence [(P7Sr2)(NO3)] < [(P7Sr2)(BeF3)] < [(P7Sr2)(BF4)]. [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) The optimized molecular geometries of the plausible isomers of the [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) supersalts along with their relative energies are shown in Fig. 8. The

Fig. 10 Static and average dipole moments, polarizabilities, and first-order hyperpolarizabilities of [(P7M2)Y] (M = Ca, Sr, Ba and Y = BeF3, BF4, NO3) supersalts

306

J Mol Model (2018) 24:306

Page 12 of 13

optimized geometries of and the Cartesian coordinates of the atoms in the [(P7M 2)Y] (M = Ba, Y = NO3, BeF3, BF4) supersalts are given in Table S6 of the ESM. The calculated binding energy (BE) values of and the NBO charges on the cationic and anionic superalkalis and superhalogens are given in Table 5, while the calculated PDOS are illustrated in Fig. 9 diagrams and the core and ligand contributions to the frontier molecular orbitals (FMOs) of the [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) supersalts are given in supporting information Table S3. The less distorted and hence less strained isomer 1 is more stable than isomer 2, which is clear according to the calculated relative energy values. Natural bond orbital (NBO) charge calculations for the superalkali (P7Ba2) and the superhalogen (NO3, BeF3, BF4) centers suggest that all of these supersalts are ionic in nature and are stabilized by strong electrostatic interactions between the superalkali cation and the superhalogen anion. Their positive binding energies further support the stability of these ionic supersalts. Plots of the density of states (DOS) and projected density of states (PDOS) for the [(P7M2)Y] (M = Ba, Y = NO3, BeF3, BF4) supersalts are shown in Fig. 9, along with the percentage contributions, and they clearly show the same trend observed for the Ca-based and Sr-based systems. The chief contributor to the HOMO and the LUMO is the superalkali (P7Ba2), while the ligand makes an important contribution to higher molecular orbitals such as LUMO+5 and LUMO+4. The HOMO– LUMO gap gradually increases in the order [(P7Ba2)(NO3)] < [(P7Ba2)(BeF3)] < [(P7Ba2)(BF4).

Nonlinear optical properties The hyperpolarizability of a molecule is a measure of its NLO activity. DFT was employed to study the NLO behavior of some materials in our work. The total molecular dipole moment and its components, the total molecular polarizability and its components, and the first-order hyperpolarizability and its components were computed at the B3LYP/def2-TZVPP level of theory. The results are reported in Table 6. The NLO behavior of a molecule is normally determined by comparing its total molecular dipole moment and its mean first-order hyperpolarizability with those of the potassium dihydrogen phosphate (KDP) molecule, which is used as a benchmark. The hyperpolarizability of the KDP molecule is 0.685 × 10−30 esu [51]. The dipole moments of our studied systems range between 12.22 and 15.64 debye, whereas the associated hyperpolarizabilities range between 1.0223 and 1.2795 × 10−30 esu. It is therefore clear that the total dipole moment and total first hyperpolarizability values of the molecules investigated here greatly exceed the corresponding values of the KDP molecule. This result implies that the molecular clusters being investigated will exhibit marked NLO properties. The variations in the dipole moment, polarizability, and first-order hyperpolarizability for the studied systems are depicted pictorially in Fig. 10. We also calculated the dipole moment, polarizability, and

hyperpolarizability values of [(P7M2)Y] (M = Ca, Sr, Ba and Y = BeF3, BF4, NO3) supersalts at the wb97X and M06-2X levels of theory using the def2-TZVPP basis set. We found that although the values obtained at the two levels of theory were different, both sets of data exhibited similar trends. The calculated values of the [(P7M2)Y] (M = Ca, Sr, Ba and Y = BeF3, BF4, NO3) supersalts at the wB97X and M06-2X levels of theory are shown in Tables S1 and S2 of the ESM.

Conclusions Zintl clusters can act as good superalkalis, so in this work we designed and theoretically investigated some novel supersalts obtained by interacting Zintl superalkalis with superhalogens. [P7M2] (M = Ca, Sr, Ba) superalkalis were attained by coupling a P 7 3− Zintl ion core with alkaline earth metals. The superhalogens NO3, BeF3, and BF4 were used in this study. The superalkalis [P7M2] (M = Ca, Sr, Ba) have ionization energies that are lower than those of alkali metals, while the superhalogens (NO3, BeF3, and BF4) have stronger electron affinities than halogens. Therefore, the mutual ionization of the superalkali and superhalogen to form the cation and anion, respectively, is a spontaneous process that eventually leads to the formation of a Zintl supersalt [(P7M2)Y] (M = Ca, Sr, Ba and Y = BeF3, BF4, NO3). The binding energies of the Zintl supersalts are positive but lower than those of conventional ionic salts. This may be due to the large sizes of the interacting superalkali and superhalogen molecules compared to the corresponding alkali metals and halogen atoms of conventional salts. This large size probably reduces the charge densities on the superalkali cation and superhalogen anion, thereby decreasing the cumulative electrostatic attractive forces, which leads to an increased distance between the cation and anion centers. The calculated DOS and PDOS were used to explore the structural features of the complexes. The findings of this work could impact on the design of cathodes in batteries where the release of cations at low energy is desirable. Our findings are a promising step towards the design and synthesis of novel structural edifices made from Zintl supersalt clusters that use Zintl superalkalis and superhalogens as building blocks. Acknowledgements This work is supported by a Department of Science and Technology INSPIRE award (no. IFA14-CH-151) from the Government of India. Utilization of the resources and computational facilities of the National Institute of Technology Rourkela are also acknowledged.

References 1. 2. 3.

Jena P (2013) J Phys Chem Lett 4:1432 Gutsev GL, Boldyrev AI (1982) Chem Phys Lett 92:262 Gutsev GL, Boldyrev AI (1981) Chem Phys 56:277

J Mol Model (2018) 24:306 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

Rehm E, Boldyrev AI, Schleyer PVR (1992) Inorg Chem 31:4834 Wu CH, Kudo H, Ihle HR (1979) J Chem Phys 70:1815 Veljkovic M, Neskovic O, Zmbov KF, Miletic M (1996) Rapid Commun Mass Spectrom 10:619 Neskovic O, Veljkovic M, Velickovic S, Petkovska LJ, Peric-Grujic A (2003) Rapid Commun Mass Spectrom 17:212 Yokoyama K, Haketa N, Hasimoto M, Furukawa K, Tanaka H, Kudo H (2000) Chem Phys Lett 320:645 Yokoyama K, Haketa N, Tanaka H, Furukawa K, Kudo H (2000) Chem Phys Lett 330:339 Hou N, Li Y, Wu D, Li ZR (2013) Chem Phys Lett 575:32 Tong J, Li Y, Wu D, Li ZR, Huang XR (2011) J Phys Chem A 115: 2041 Tong J, Li Y, Wu D, Wu ZJ (2012) Inorg Chem 51:6081 Tong J, Wu Z, Li Y, Wu D (2013) Dalton Trans 42:577 Sun WM, Li Y, Wu D, Li ZR (2013) J Phys Chem C 117:24618 Wang XB, Ding CF, Wang LS, Boldyrev AI (1999) J Chem Phys 110:4763 Wu MM, Wang H, Ko YJ, Kandalam AK, Kiran B, Wang Q, Sun Q, Bowen KH, Jena P (2011) Angew Chem Int Ed 50:2568 Pathak B, Samanta D, Ahuja R, Jena P (2011) ChemPhysChem 12: 2422 Gutsev GL, Rao BK, Jena P, Wang XB, Wang LS (1999) Chem Phys Lett 312:598 Giri S, Behera S, Jena P (2014) J Phys Chem A 118:638 Srivastava AK, Misra N (2014) New J Chem 38:2890 Ulime EA, Pogrebnoi AM, Pogrebnaya TP (2017) Curr J Appl Sci Technol 22:1 Zintl E, Morawietz W (1938) Z Anorg Allg Chem 236:372 Jansen M (1977) Z Anorg Allg Chem 435:13 Khanna SN, Jena P (1992) Phys Rev Lett 69:1664 Khanna SN, Jena P (1995) Phys Rev B Condens Matter Mater Phys 51:13705 Li Y, Wu D, Li ZR (2008) Inorg Chem 47:9773 Yang H, Li Y, Wu D, Li ZR (2012) Int J Quantum Chem 112:770

Page 13 of 13 306 28. 29.

Srivastava AK, Misra N (2014) Mol Phys 112:2621 Jing YQ, Li ZR, Wu D, Li Y, Wang BQ, Gu FL, Aoki Y (2006) ChemPhysChem 7:1759 30. Giri S, Reddy GN, Jena P (2016) J Phys Chem Lett 7:800 31. Chapman DJ, Sevov SC (2008) Inorg Chem 47:6009 32. Ugrinov A, Sevov SC (2003) J Am Chem Soc 125:14059 33. Knapp CM, Large JS, Rees NH, Goicoechea JM (2011) Dalton Trans 40:735 34. Knapp C, Zhou B, Denning MS, Rees NH, Goicoechea JM (2010) Dalton Trans 39:426 35. Feierabend M, Von-Hänisch C (2014) Chem Commun 50:4416 36. Zhou B, Denning MS, Jones C, Goicoechea JM (2009) Dalton Trans 9:1571 37. Dolyniuk JA, Kovnir K (2013) Crystals 3:431 38. Becke AD (1993) J Chem Phys 98:5648 39. Lee C, Yang W, Parr RG (1988) Phys Rev B Condens Matter Mater Phys 37:785 40. Chai JD, Head-Gordon M (2008) J Chem Phys 128:084106 41. Zhao Y, Truhlar DG (2006) Theor Chem Accounts 120:215 42. Weigand F, Ahlrichs R (2005) Phys Chem Chem Phys 7:3297 43. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani G, Barone V, Mennucci B, Petersson GA et al (2010) Gaussian 09, revision B.01. Gaussian, Inc., Wallingford 44. Boyle NMO, Tenderholt AL, Langner KM (2008) J Comput Chem 29:839 45. Read AE, Curtiss LA, Weinhold F (1988) Chem Rev 88:899 46. Kleinman DA (1962) Phys Rev 126:1977 47. Adant C, Dupuis M, Bredas JL (1995) Int J Quantum Chem 56:497 48. Turbervill RS, Goicoechea PJM (2012) Chem Commun 48:1470 49. Behera S, Samanta D, Jena P (2013) J Phys Chem A 117:5428 50. Paduani C, Jena P (2012) J Phys Chem A 116:1469 51. Saravana Kumar G, Murugakoothan P (2015) AIP Conf Proc 1665: 100006