Quantum Chemical Study of Redox-Switchable Second-Order

ARTICLE pubs.acs.org/JPCA

Quantum Chemical Study of Redox-Switchable Second-Order Nonlinear Optical Responses of D
π
A System BNbpy and Metal Pt(II) Chelate Complex Na-Na Ma,† Shi-Ling Sun,† Chun-Guang Liu,‡ Xiu-Xin Sun,† and Yong-Qing Qiu*,† †

Institute of Functional Material Chemistry, Faculty of Chemistry, Northeast Normal University, Changchun 130024, People’s Republic of China ‡ College of Chemical Engineering, Northeast Dianli University, Jilin 132012, People’s Republic of China

bS Supporting Information ABSTRACT: A second-order nonlinear optical (NLO) molecular switching with redox has been investigated in the present paper. The static first hyperpolarizabilities of 5-(BMes2)50 -(NPh2)-2,20 -bipyridine (BNbpy) containing three-coordinate organoboron, Pt(II) chelate complex Pt(BNbpy)Ph2, and their reduced forms have been calculated by density functional theory (DFT) combined with the analytic derivatives method. There is an enhancement of static first hyperpolarizabilities in the reduced form according to the calculations. That is, the βvec value of oneelectron-reduced form is ∼7 times as large as that of neutral form BNbpy; the βvec values of one- and two-electron-reduced forms are ∼3 and ∼4 times as large as that of neutral form Pt(BNbpy)Ph2, respectively. In particular, the βvec value of two-electronreduced form 3Pt(BNbpy)Ph22
is 1349  10
30 esu, ∼286 times larger than its neutral form. Moreover, the component βz value of the metal chelate complex Pt(BNbpy)Ph2 is 25  10
30 esu, which is ∼14 times as large as that of ligand BNbpy; the corresponding F
/CN
compounds show a decrease in βx values compared with the case of the ligand and Pt(II) complex. Analyses of geometries, density of states (DOS), and time-dependent DFT (TDDFT) calculations reveal that the one-electron reduction promotes the molecular conjugation in the x-axis and intensifies the interaction between the metal Pt(II) and ligand and then results in an enhancement of the static first hyperpolarizability, whereas the binding of F
/CN
to the B atom turns off the pπ
π* conjugation and has no effect on the conjugation of bipyridine, which leads to a decreasing β value in the x-axis.

1. INTRODUCTION Nonlinear optical (NLO) processes are being theoretically and experimentally exploited in a variety of optoelectronic modulators for optical telecommunication and photonic applications such as optical data manipulations, storage, and transmission.1 It is well-known that the molecular second-order NLO property depends not only on the nature of the πconjugated bridge but also on the strength of the donor and acceptor groups. In particular, three-coordinate organoboron compounds are excellent electron acceptors due to the empty pπ orbital on the boron center, which have been widely applied in materials chemistry as nonlinear optical materials, chemical sensors, and emitters for organic light-emitting diodes (OLEDs). When accompanied by electron donors such as amines, these molecules possess large electronic dipoles, which promote donor
acceptor charge transfer upon excitation with light.2 The work of Williams et al. and Kaim et al. suggested that the dimesitylboryl (Mes2) unit might be usefully incorporated into three-coordinate organoboron electronic materials.3 For example, the change from a small dipole moment in the ground state to a large dipole moment in the first excited state indicated r 2011 American Chemical Society

that donor
acceptor compounds using the Mes2 group as a πelectron acceptor might exhibit nonlinear optical behaviors. The NLO materials with Mes2 group have been investigated theoretically and experimentally.4 Recently, however, the new classes of triarylboranes based on 2,20 -bipyridine (bpy) BNbpy (5(BMes2)-50 -(NPh2)-2,20 -bipyridine) have been synthesized (Figure 1),5 which are strongly conjugated donor
acceptor systems, and the boron center in triarylbornes is attached directly to an electronegative N-heterocycle. As such, the BNbpy may be a potential excellent NLO material. In recent years, the organometallic and coordination compounds have been investigated as new second-order NLO materials, mainly because they can offer additional flexibility by introducing new charge transfer transitions between the metal and the ligand. Then, the effects of metal chelation on various push
pull ligands, such as substituted pyridines,6 bipyidines,7 and porphyrin,8 have been extensively studied. Exactly, BNbpy Received: June 26, 2011 Revised: September 30, 2011 Published: October 03, 2011 13564

dx.doi.org/10.1021/jp206003n | J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A

ARTICLE

Figure 1. Structural formulas of the systems at the focus of the present study.

has the excellent chelating ability due to 2,2-bipyidines. That is, strong electronic communication between the boron and the metal center is promoted with the chelation of metal ion. Nowadays, it is still a challenging subject to design the switchable NLO materials. To achieve a pronounced switching effect, the molecule must be stable in two or more states that exhibit notably different NLO responses. Complete reversibility and high switching speed are highly desirable in practical applications. Specific procedures including protonation/deprotonation,9 oxidation/ reduction,10 and photoisomerization11 have been employed to switch NLO responses. As a consequence of the enhanced π conjugation, BNbpy and the metal Pt(II) complex are stable toward reduction and display reversible reduction peaks.5 Therefore, it may provide a model for redox switching of NLO responses. The development of materials exhibiting switchable NLO properties can be expected to lead to novel applications for optoelectronic technologies.9a,12 In the present work, we have investigated the difference between push
pull systems BNbpy (1) and metal chelation complexes Pt(BNbpy)Ph2 (2) (Figure 1) in relation to second-order NLO properties, and their redox switching character of the NLO responses in detail. Moreover, because the three-coordinate boron center is capable of selective binding to anions such as fluorides and cyanides,13 we have also considered the NLO switching by F
/CN
.

of all systems have been fully optimized with no symmetry constraint using the B3LYP functional. The standard 6-31G(d) polarized double-ζ basis set, starting from a standard one, is for the nonmetal elements, and the Pt atom adopted the effective core potential (ECP) double-ζ (DZ) basis set of LanL2DZ to take into account the relativistic effect. No X-ray data are available; thus the geometry optimizations were followed by the frequency calculation of the normal modes of vibration to check the ground-state nature of the optimized structures. TDDFT is one of the most popular methods for the calculation of excitation energies in quantum chemistry due to its efficiency and accuracy. Therefore, we employed TDDFT to calculate the electron spectra of the studied systems. Also, to consider the solvent effects in the energy and excitation properties, the polarizable continuum model (PCM)15 was employed in the calculations of frequency and TDDFT for some systems. The response of a molecule to a homogeneous static electric field can be represented by the following two Taylor expansions:16 EðFÞ ¼ E0


∑i μ0i Fi
1=2 ∑ij αij FiFj
1=3 ∑ijk βijkFi FjFk


1=4

μi ðFÞ ¼ μ0i þ

2. COMPUTATIONAL DETAILS In the present paper, all the calculations were performed by using the GAUSSIAN 09W program package.14 The geometries

∑ijkl γijklFi FjFkFl þ :::

ð1Þ

∑j αij Fj þ ∑jk βijk Fj Fk þ ∑jkl γijklFjFk Fl þ ::: ð2Þ

13565

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A

ARTICLE

Table 1. Selected Bond Lengths (Å) and BLA Values Obtained at the B3LYP/6-31G(d)/LanL2DZ Level bond length

1

1


1F


1CN


2

2


1 2


3 2


2

2F


2CN


C1
C11

1.405

1.418

1.399

1.399

1.402

1.421

1.441

1.417

1.397

1.395

C2
C12

1.403

1.422

1.409

1.408

1.401

1.421

1.443

1.407

1.404

1.403

C1
C2

1.482

1.451

1.482

1.483

1.472

1.434

1.395

1.444

1.473

1.474

C1
N3

1.349

1.370

1.349

1.347

1.360

1.387

1.411

1.374

1.360

1.360

C2
N4

1.347

1.366

1.349

1.349

1.360

1.381

1.412

1.405

1.359

1.358

C5
N3

1.333

1.320

1.337

1.337

1.341

1.328

1.320

1.352

1.346

1.345

C6
N4

1.328

1.328

1.332

1.332

1.334

1.338

1.334

1.346

1.339

1.339

C9
C11 C10
C12

1.387 1.387

1.377 1.380

1.394 1.386

1.393 1.386

1.388 1.386

1.373 1.377

1.360 1.366

1.377 1.385

1.391 1.386

1.391 1.386

C7
C9

1.411

1.433

1.405

1.406

1.408

1.433

1.455

1.425

1.403

1.404

C8
C10

1.405

1.411

1.403

1.402

1.407

1.414

1.431

1.432

1.403

1.403

C5
C7

1.412

1.433

1.413

1.413

1.406

1.422

1.437

1.410

1.411

1.412

C6
C8

1.411

1.402

1.401

1.402

1.411

1.396

1.394

1.386

1.400

1.401

C7
B13

1.569

1.528

1.660

1.660

1.579

1.531

1.503

1.573

1.665

1.665

C8
N14

1.409

1.434

1.428

1.427

1.397

1.428

1.436

1.435

1.417

1.416

1.478

1.632

1.470

1.627

B13-X Pt
N3

2

2.193

2.180

2.171

2.180

Pt
N4

2.193

2.172

2.156

2.158

BLA(py1)

0.026

0.039

0.063

0.012

BLA(py2)

0.042

0.035

0.064

0.015

where E0 is the energy of the molecule in the absence of an electric field, μ0i is its permanent dipole moment, αij is the dipole polarizability, and βijk and γijkl are the first and second hyperpolarizablities, respectively. As implied by this, the polarizabilities can be computed by employing derivatives of the energy or dipole moment with respect to the incident electric field. To compute the polarizability and hyperpolarizability, one option is to take the derivates either analytically or numerically. In this work, the static first hyperpolarizability βvec was calculated by analytical third energy derivatives, which is more efficient and less expensive than numerical derivatives,17 where βvec refers to the hyperpolarizability along the molecular dipole moment defined as follows: βvec ¼

μβ

i i ∑ jμj i ¼ x, y, z

βi ¼ βiii þ

1 3

∑ ðβijj þ βjij þ βjji Þ

j6¼ i

ð3Þ

ð4Þ

where μi denotes the dipole moment along i direction of ground state. As shown in Figure 1, in 1, 1
and the corresponding F
/ CN
compounds, the dipolar axis is normally taken as x and the βx value is then dominant. For a series of 2, the dipolar axis lies in x and z and the βx and βz values are dominant. To utilize the merits of range-separated and global hybrids, the CAM (Coulomb-attenuating model) was proposed as an applicable approach for the first hyperpolarizability.18 Moreover, mPWPW91* is an excellent hybrid DFT functional for the first hyperpolarizability of transition metal complexes.19 Therefore, the βvec values of all the systems were calculated by using the CAM-B3LYP20 and mPWPW91* functionals. The LanL2DZ basis set was used for transition metal Pt, whereas the nonmetal elements took 6-31+G(d) all-electron basis set.

3. RESULTS AND DISCUSSION 3.1. Geometries. The geometry structures and electronic properties of the 1, 2 and their reduced forms 1
, 2
, 22
with all real frequency have been obtained using B3LYP/UB3LYP/631G(d) level (LanL2DZ basis set on metal ion). For the ligand 1, there are two conformations, as shown in Figure 1. The calculation result shows that the BNbpy (trans) structure is 6.90 kcal/ mol more stable than BNbpy0 (cis). For the Pt(II) chelate complex, the Pt(II) ion electronic configuration is d8, and the geometry is square-planar.5 Therefore, two possible spin states (closed-shell singlet, open-shell triplet) have been considered for 2, and it is found that the energy of the closed-shell singlet is 43.14 kcal/mol more stable than that of the open-shell triplet, which illustrates that the ground state of 2 is closed-shell singlet. In addition, for the second one-electron-reduced process of Pt(II) complex 2, two possible spin states (the singlet state 1 2
2 and triplet state 322
) have been also optimized. It can be found that the singlet 122
is 13.87 kcal/mol more stable than the triplet state 322
. The selected bond distances for a series of optimized structures are shown in Table 1. Most of the bond distances of bipyridines increase through one-electron reduction of 1 except for C9
C11, C10
C12, and C6
C8, which decrease in 0.010
0.003 Å. Particularly, the length of C7
B13 shortens and the distance of C8
N14 elongates (number of atoms, see Figure 2). This suggests that the reduction process enhances the conjugation between the B atom and the pyridine ring. Furthermore, the decreasing bond length (1.482 f 1.451 Å) of C1
C2 demonstrates that the conjugation between two pyridine rings is also enhanced. These changes on structure must affect the molecular NLO properties. For the first one-electron reduction process of 2, two pyridines expand with the increasing bond lengths C1
C11, C2
C12, C1
N3, C2
N4, C7
C9, C8
C10, and C5
C7; the distances C1
C2 and C7
B13 decrease, and the C8
N14 bond increases, whereas others change slightly; and the 13566

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A

ARTICLE

Figure 2. Molecular structure of 1 showing the atom labeling scheme. Hydrogen atoms are omitted for clarity.

Table 2. Calculated BSSE Corrected Binding Energies (DeBSSE, kcal/mol) of Pristine Molecules 1 and 2 with Different Anions at B3LYP/6-31G(d)/LanL2DZ level basis sets 6-31G* 6-31G*/LanL2DZ

1F


1CN



54.280


58.672

2F


2CN



58.798


61.057

Pt
N3 and Pt
N4 decrease. With the injection of the second electron, a similar structural change is found for 122
; that is, the elongated bonds further increase and the shortened bonds further decrease compared with those of 2
. This implies that the reduction processes of 2 also have a significant influence on geometry. Then, we believe that the reduction process may occur on the B atom and bpy according to the major change on structure. It also can be found the change of bond length for 322
is similar to that of other reduced forms. Finally, we calculated the bond length alternation21 (BLA, the difference between single and double bond lengths) of two pyridines for 2 and its reduced forms, and the results are listed in Table 1. It is found that the BLA values of pyridines in 322
are 0.012 and 0.015, which are significantly smaller than that of other systems, and the distance C1
C2 shortens to 1.444 Å. This implies that the conjugation of bpy is enhanced obviously. The decreasing distances of two Pt
N bonds suggest that the interaction between metal ion Pt(II) and ligand is also enhanced. The counterpoise (CP) correction scheme of Boys
Bernardi has been used to take into account the effect of basis set superposition error (BSSE) in optimizations and calculations of binding energies for F
/CN
compounds (Figure 1). As shown in Table 1, besides C7
B13 and C8
N14, the optimized bond distances of 1F
and 1CN
change slightly compared with those of 1 and 2, respectively. The C7
B13 and C8
N14 bonds elongate and the C1
C2 bond changes slightly, which implies that the introduction of F
and CN
reduce the conjugation between boron and pyridine, and have no influence on the conjugation of bpy. 3.2. Responses to Fluorides and Cyanides. One particular feature of 1 and its Pt(II) complex 2 that distinguishes them from previously push
pull Pt(II) complexes is that the three-coordinate boron center can bind to anions. In this work, we have calculated the binding energy of 1F
, 1CN
, 2F
, and 2CN
, which are shown in Table 2. The calculated binding energy

Figure 3. LUMO of 1 and 2, and the spin density plots (contour value = 0.0025) obtained at the B3LYP/6-31G(d)/LanL2DZ level for oneelectron reduced species of 1 (a) and one-electron (b) and two-electron (c) reduced species of 2.

Table 3. Comparison of Calculated at the B3LYP/6-31G(d)/ LanL2DZ Level and Experimental Reduction Potentials versus FcCp20/+ in DMF for the Ligand 1 and Pt(II) Complex 2 system

a

E1/2red1 (V)

1


1.99

2


1.62 a

a

E1/2red2 (V)

ΔE (V)


2.38

b


1.92b


2.10 a


2.32b

0.48a

0.40b

b

Experimental values. Calculated values.

of
54.28/
58.67 kcal/mol for 1F
/CN
is slightly larger than
58.80/
61.06 kcal/mol for 2F
/CN
, which illustrates that the ability of binding fluoride or cyanide anions to the boron center in the metal complex and the free ligand is in the order 2CN
(2F
) > 1CN
(1F
), consistent with the previously report.5 This reveals that the chelation of the Pt(II) center enhances the electron accepting ability of the boron atom, which may cause a change on the NLO responses. Moreover, the bond length B
X decreases with the increasing binding energy, as shown in Table 1. 3.3. Redox Properties of 1 and 2. Changes of the electronic properties are related to the highest occupied orbitals (HOMO) and affect the lowest unoccupied orbitals (LUMO), thus leading to the change in the redox properties. For the studied reduction process, as shown in Figure 3, the LUMO of 1 mainly delocalizes on the B atom and bpy, and the LUMO of 2 concentrates on the B atom, bpy, and Pt(II) ion. The second reduction of 2 can occur at two possible sites, the B atom, bpy, and the Pt(II) ion (β-LUMO in 2
) or BMes2, bpy, the Pt(II) ion, and NPh2 (α-LUMO). Associating with the frontier molecular orbital, hence, suggests that the B atom and bpy may be the reduction 13567

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A

ARTICLE

Table 4. Experimental and Theoretical Data Calculated at the PBE1PBE/6-31+G(d)/LanL2DZ Level (Transition Energy E (eV), Oscillator Strength (f), and the Corresponding Dominant MO Transitions) for 1, 2, and F
/CN
Compoundsa system

λexpt (nm)

λcalc (nm)

f

E (eV)

400

415 (434)

major transition H
0 f L+0(96%) (H
0 f L+0)

0.8032 (0.8746)

2.99 (2.86)

1F


337

0.2502

3.68

1CN


316

0.8005

3.93

H
2 f L+0(44%)

H
1 f L+0(27%)

356

348

0.0289

3.57

H
0 f L+3(36%)

H
3 f L+1(34%)

H
1 f L+2(8%)

H
0 f L+2(8%)

430

425

0.5968

2.92

H
3 f L+0(49%)

H
5 f L+0(19%)

H
4 f L+0(16%)

H
6 f L+0(10%)

313 376

0.1303 0.2998

3.96 3.30

H
7 f L+2(63%) H
9 f L+0(31%)

H
10 f L+2(17%) H
7 f L+0(26%)

H
8 f L+0(12%)

H
5 f L+1(9%)

1

2

2F


H
2 f L+1(30%) H
2 f L+2(10%)

H
0 f L+1(21%) H
3 f L+1(10%)

H
0 f L+0(8%)

H
1 f L+0(5%)

H
4 f L+1(6%) 2CN


312 368

0.1411

H
8 f L+2(45%)

3.98

0.1848

3.37

H
6 f L+2(20%)

H
5 f L+2(8%)

H
9 f L+2(10%)

H
9 f L+0(36%)

H
3 f L+1(25%)

H
6 f L+0(25%) a

The bold data in parentheses are TDDFT results in CH2Cl2 solution for system 1.

Figure 4. Molecular orbitals of 1 and 2 involved in the dominant electron transitions obtained at the PBE1PBE/6-31+G(d)/ LanL2DZ level. The black arrow represents the intense absorption peak and red arrow for the weak one.

centers of 1, and the first one-electron-reduction processes of 2 occur on the B atom, bpy, and the Pt(II) ion. Fortunately, the spin density distributions for 1
(a) and 2
(b) confirm these predictions (Figure 3). So the second one-electron-reduction process of 2 also occurs on the B atom, bpy, and Pt(II) ion, as shown by the β-LUMO in 2
according to the spin density of 22
(c). We proceed to reproduce the electrochemical behavior of the present studied systems. It has been proved that the oxidation
reduction potentials can be predicted theoretically in solution. The theoretical prediction of the redox potential of 1 and 2 requires the determination of the free energy associated

with the process. BNbpy ox ðaqÞ þ e
f BNbpy red ðaqÞ ΔG

PtðBNbpyÞPh2ox ðaqÞ þ e
f PtðBNbpyÞPh2red ðaqÞ ΔG

The term ΔG represents the free energy of the reduction process in solution, which can be obtained by frequency calculation of 1, 1
and 2, 2
, 122
in DMF solution. The free energy change associated with the reference normal hydrogen electrode half-reaction has been recalculated to be
4.28 V.22 So the reference electrode Ag/AgCl (FcCp20/+ was used as the standard E1/2 = 0.55 V in ref 5) should 13568

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A

Figure 5. Total and partial density of states (TDOS and PDOS) around the HOMO
LUMO gap obtained at the B3LYP/6-31G(d) level for 1 and 1
.

be
4.83 V. When the Nernst equation E0 =
ΔG0/nF is combined with the reference value
4.83 V, the reduction potentials (Ecal) can be predicted. Table 3 summarizes calculated and experimental reduction potentials. The E1/2red1 potential of the 2 is much more positive than the free ligand, again supporting that chelation to the Pt(II) center enhances the electron accepting ability of the boron center. 3.4. Absorption Spectra of 1, 2, and the Corresponding F
/CN
Compounds. The binding to anions not only can be exploited for anion sensing but more importantly can be used to probe the various electronic transition pathways such as MLCT in the complex. We therefore investigated and compared the responses of 1 and the 2 toward fluoride and cyanide ions in UV
vis. The UV
vis absorption spectra of 1 and 2 have been reported. To rationalize the observed spectral properties, the electron absorption spectra of 1, 1F
, 1CN
, 2, 2F
, and 2CN
have been calculated within the TDDFT framework at the CAMB3LYP and PBE1PBE levels with 6-31+g* (LanL2DZ basis set on metal ion). CAM-B3LYP has demonstrated that there is an improvement on charge transfer excitations.23 PBE1PBE, also known as PBE0, which is obtained by casting the functional and correction of Perdew, Burke, and Erzenrhof in a hybrid HF/DFT scheme with a fixed 1/4 ratio,24 has been shown to improve the accuracy of excitation energies and charge transfer bands in metal

ARTICLE

Figure 6. Total and partial density of states (TDOS and PDOS) around the HOMO
LUMO gap obtained at the B3LYP/6-31G(d)/LanL2DZ level for 2 and 2
.

complexes for both gas phase and solution calculations.25 TDPBE1PBE calculations account well for the experimental linear optical spectroscopic features, and a satisfactory agreement between calculated and experiment is found (Table 4). The solvent effect exerts critical influence on the absorption spectra according to the earlier reports.26 In this work, therefore, the electronic excitation computations were also performed in dichloromethane (CH2Cl2) solution for ligand 1 by using PCM model. The major absorption with maximal oscillator strength and the transition energy on the basis of TDDFT calculations are listed in Table 4. The spectrum of 1 exhibits an intense absorption band with the lowest energy in the UV
vis region at 415 nm, which is attributed to the transition from the NPh2 moiety to the B atom (Figure 4) and therefore assigned as an intraligand charge transfer (ILCT) transition. Moreover, the HOMO and LUMO molecular orbitals involving the major transition are localized on the bpy moiety and display π orbital character. Therefore, the absorption band has also a significant π f π* character. A small bathochromic shift of wavelength with maximal absorption is found in CH2Cl2 solution according to the PCM model calculation. Compared with ligand 1, 1F
and 1CN
both display blue shifts on intense absorption that are attributed to the BMes2 f NPh2 ILCT and π f π* transition (see Figure S1 of the 13569

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A

ARTICLE

Table 5. Dipole Moment of Ground State (6-31G(d)/ LanL2DZ Basis Sets) and Static First Hyperpolarizability (6-31+G(d)/LanL2DZ Basis Sets)a B3LYP system

μx

μz

1


2.0

0.04 2.0

1


μ


10.3
0.2 10.3

functional


20.5
0.1 20.5

1CN

21.5
1.8 21.6 2 2
1 2


2

3 2


2

2F



2.9

8.9

9.4


7.0 13.1 14.8
12.3 16.2 20.3
13.2 17.7 22.2
19.7 14.8 24.7

2CN

20.2 16.7 26.4 10


1.8

2.6

3.2

βz

βvec

CAM-B3LYP

75.9

1.5


75.9 (
75.6)

mPWPW91*

98.5

1.9


98.4

CAM-B3LYP
529.9 11.7 mPWPW91*

1F


βx


566.4

4.3

529.5 (509.5) 566.2

CAM-B3LYP


37.0
0.03

37.0

mPWPW91*


42.7

0.1

42.7

CAM-B3LYP


32.3

1.0

32.0 (32.1)

mPWPW91*


37.3

1.2

37.0
1.6

(36.9)

CAM-B3LYP

71.6 21.4

mPWPW91*

91.7 24.5

CAM-B3LYP

138.5 39.3


30.3

mPWPW91*

102.4 38.7


13.9 (
14.2)

CAM-B3LYP

81.5 71.3

7.4

mPWPW91*

84.4 86.5

17.7 (15.3)


4.7 (
3.8)

CAM-B3LYP
846.8 168.5

649.0

mPWPW91*
2014.9 169.4

1349.4 (1508.3)

CAM-B3LYP


22.3 17.8

mPWPW91*


25.8 19.9

32.8 (33.3)

CAM-B3LYP


16.1 19.3

24.9

mPWPW91*


18.2 21.8

CAM-B3LYP

59.8

1.6

28.8

28.1 (28.8)
32.0

a

The data in parentheses are calculated by using the 6-311+G(d,p) basis set.

Supporting Information). Notably, for ligand 1, the B atom is electron deficient because of its vacant p-orbital, and the vacant p-orbital effectively overlaps with adjacent π-skeleton of bpy in LUMO (Figure 4). However, the p
π* conjugation is broken because the p-orbital becomes occupied by the attack of F
/ CN
, which causes a change in delocalization pattern of LUMOs. That is, the electronic densities of LUMOs move toward the π-conjugated bpy and NPh2 that act as auxiliary acceptors whereas the Mes2 acts as an electron-donating group (see Supporting Information, Figure S1). For the spectrum of 2, intense and weak absorption bands are found in the UV
vis region at 356 and 430 nm, respectively. The two bands can both be characterized as a mixed ligand to ligand charge transfer (LLCT) and metal to ligand charge transfer (MLCT) transition. The LLCT transition involves the electron excitation from phenyl rings of
PtPh2 to the BNbpy moiety, and the MLCT transition is described as electron redistribution from the d orbital of platinum to the unoccupied set of π* orbitals belonging to the bpy. More specifically, the LLCT transition of the weak absorption band is from the phenyl rings of PtPh2 to the B atom and bpy, whereas it is from the phenyl rings of PtPh2 to the B atom, N atom, and bpy for intense absorption. From Table 4, blue shifts of two absorption bands are observed in 2F
and 2CN
compared with 2. Importantly, as shown in Figure S2 (Supporting Information), when F
and CN
were introduced, the two absorption bands are also attributed to LLCT and d f π* MLCT transitions, and the LLCT transition can be described as an excitation from the B atom and bpy moiety to
NPh2. For complex 2, the B atom has a major contribution in the formation

of its LUMO as well as 1, and the Pt(II) ion has the ability of an donating electron. As with the ligand 1, the F
/CN
compounds of 2 have similar delocalization patterns of LUMOs by the breaking of p
π* conjugation. 3.5. Density of States Calculation. For the purpose of more detailed comparative study of the electronic structures for reduced forms, the total density of states (TDOS) and projected partial density of states (PDOS) calculations have been carried out by using the Aomix program. As plotted in Figure 5, it reveals that in ligand 1 the boron atom plays a significant role in the formation of LUMOs, whereas it mainly contributes to αHOMO and a little to α-LUMO in the reduced form 1
. This illustrates that in the reduced form there would still be p
π* conjugation between the B atom and the adjacent π-skeleton such as trimethylbenzene or bpy in LUMO, which is helpful for the first hyperpolarizability in the x-axis. A similar situation is found for Pt(II) complex 2 and its first one-electron-reduced form 2
, as shown in Figure 6. For 2, the boron atom plays a significant role in the formation of LUMO, whereas it the major formation of α-HOMO and little to α-LUMO in 2
. And the Pt(II) ion mainly focuses on the HOMOs in both forms. However, the B atom is only a small proportion in the HOMO of 22
(Figure S3, Supporting Information). The analysis suggests that the B atom has capability of donating an electron after the injection of an extra electron. In summary, for a series of 2, Pt(II) mainly displays electron donor character, and the B atom plays a role in withdrawing electron in the neutral form but donating an electron in reduced forms. In addition, the Mulliken populations of atomic orbitals in the isolated fragments and in the whole molecule and their difference reaction have been presented in Table S1 (Supporting Information). As data show, the p-atomic orbitals contributed to the boron atom and the d-atomic orbitals to Pt(II), which visualized the reaction pattern of the B atom in the reduction process. 3.6. Static and Frequency-Dependent First Hyperpolarizability. The calculated dipole moment of the ground state and the static first hyperpolarizability component βx, βz, and βvec values of the studied systems have been listed in Table 5. To investigate the effect of DFT methods on NLO responses, the β values were obtained with the two different functionals on the basis of the same geometry. As the data show, all the β values of the CAM-B3LYP functional are slightly smaller than that of the mPWPW91* except for 2
, but the two functionals show the same trend in β values. Then, we will discuss the β values of a series of 2 at the mPWPW91* level and use the CAM-B3LYP functional for systems without Pt(II). First, the βx value of 2 is 71.6  10
30 esu at the CAM-B3LYP level, which is close to 75.9  10
30 esu for ligand 1. However, the βz value, 21.4  10
30 esu, is ∼14.3 times as large as that of 1 (the mPWPW91* results is similar to that of CAM-B3LYP). Moreover, the βx value of cis-ligand 10 is 59.8  10
30 esu and the βz value is 1.6  10
30 esu. This suggests that the trans- is better than cis-ligand for NLO material and introduction of metal ion Pt(II) increases the distance of charge transfer in the z-axis, thus enhancing the NLO responses. The βx value of the oneelectronic reduced species, 1
, is ∼7.0 times as large as that of 1, which is attributed to the enhanced conjugation in the x-axis due to the charge distribution after the injection of an electron; the βx and βz values of the one-electronic-reduced species 2
are slightly larger than that of 2 due to enhanced conjugation, which results in an increase on the βvec value of 2
, ∼3.0 times as large as that of the neutral form; for the stable two-electron-reduced 13570

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A species, 122
, the βz value is ∼3.1 times as large as that of 2. Taken together, we can judge that the second reduction process mainly focuses on the metal ion Pt(II) in the z-axis of 22
and the MLCT transition is helpful for the NLO responses. It is also interesting to notice that the components βx and βz of twoelectron-reduced species 322
are significantly larger than that of neutral 2 and one-electron reduced 2
; thus the βvec value, 1349.4  10
30 esu, is ∼286.1 times larger than that of 2, ∼96.1 times larger than that of 2
. This indicates that the incorporation of electron injection one-by-one causes an enhancement in the molecular second-order NLO response. From a structure
property point of view, the BLA value of 322
is obviously smaller than that of other reduced systems of 2, as mentioned above, which means that the bridge conjugation of system 322
in the x-axis is enhanced. And the interaction between metal Pt(II) and ligand BNbpy is also enhanced according to the shortened distance of Pt
N. Both of these amplify the NLO responses with the enhanced conjugation. The βx value of 1F
is
37.0  10
30 esu, and
32.3  10
30 esu for 1CN
, which are as low as half of 1 (75.9  10
30 esu). That is, the binding of F
and CN
to 1 decreases the absolute value of the β value. Recovering its causes, the π f π* transition, which is helpful for NLO responses, was reduced due to the breaking of p
π* conjugation between the B atom and the adjacent π-skeleton. Another reason is that the conjugation of bpy was not enhanced. More interestingly, the sign of the β value becomes positive from negative in the x-axis, which must be related to the change of CT direction. It also illustrates that the F
and CN
compounds produce similar photoelectrical changes in 1 according to the analogical absolute and sign of β. A similar result is found for Pt(II) complexes; that is, in comparison with the value for 2, the βx value of 2F
is as low as
25.8  10
30 and
18.2  10
30 esu for 2CN
, respectively, and the βz values change slightly. Moreover, the β values of F
/CN
compounds of 1 are ∼2.0 times as large as that of F
/CN
compounds of 2. We know that the chelation of the Pt(II) center enhances the electron accepting ability of the boron atom. So, the decrease of the β value in F
/ CN
compounds of 2 can be explained by the stronger interaction between the boron atom and F
/CN
, the more serious breaking of p
π* conjugation between the B atom and the adjacent π-skeleton. We have also calculated the βvec values of all systems using the 6-311+G(d, p) basis to check the effects of the triplet-ζ basis set and adding a set of polarization p functions for nonmetal elements on NLO responses. The results listed in Table 5 show that the larger 6-311+G(d, p) basis set has little influence on the β values for all systems compared with the 6-31+G(d) basis set. Moreover, the hyperpolarizability is related to the energy and oscillator strength (f) of permitted UV
vis excitations according to the two-level model.27 Indeed, it is worth noting that all the calculated βvec values vary according to λmax, namely, the highest λmax corresponding to the highest βvec. The orders of the βvec value, 1 > 1F
(1CN
) and 2 > 2F
(2CN
), testify that the red shift of λmax is helpful for enhancing the NLO responses. Finally, we calculated the frequency-dependent first hyperpolarizability for our studied systems because βvec is the quantity that is sampled in electric-field-induced simple harmonic (EFISH) measurements. The β(
2ω;ω,ω) values have been investigated by frequency-dependent coupled-perturbed (CP) DFT,28 and the same basis sets with static βvec value calculation are employed. Generally, the molecular hyperpolarizabilities, β(
2ω;ω,ω), have been measured in a fundamental incident

ARTICLE

wavelength which has a second harmonic far enough from the absorption bands to avoid the overmeasure on the βvec value due to resonance effects. Hence, according to the maximum wavelength of the absorption band (430 nm) of the studied systems, we investigated the frequency dispersion at a near resonant wavelength of 1.340 μm (ω = 0.0340) and a nonresonant wavelength of 1.907 μm (ω = 0.0239). The calculation results are presented in Table S2 (Supporting Information) and show the expected increase in magnitude of the first hyperpolarizabilty with increasing frequency. Only one exception is β1.907 of one-electronreduced form 1
, which could produce a second harmonic resonant with the weak absorption around the second harmonic.

4. CONCLUSION Systematic DFT calculations have been carried out on ligand 1 and its Pt(II) chelate complex 2. The electron spectra of 1F
/ CN
and 2F
/CN
and second-order NLO properties of all systems have been analyzed in detail. The calculations indicate that the binding of F
/CN
significantly affects the direction and pattern of CT between BMes2 and NPh2 along the x-axis and thus changes the absolute value and sign of the static first hyperpolarizablities. Then, a decrease of the βx value in F
/ CN
compounds has been observed because the binding of F
/ CN
breaks the p
π* conjugation between the B atom and the adjacent π-skeleton and does not enhance the bpy conjugation. Moreover, the β value of 1F
/CN
is ∼2.0 times as large as that of 2F
/CN
, because the chelation of the Pt(II) center enhances the electron accepting ability of the boron atom. On the basis of the excellent redox properties of 1 and 2, the redox switching of the NLO responses has been studied. The results show that the reduction process affects the geometries resulting in a better conjugation in bpy than neutral and thus heightens the first hyperpolarizabilities βvec. The βvec value of one-electronreduced species 1
is ∼7.0 times as large as that of its neutral form 1, and that of one- and two-electron-reduced species is ∼3.0 times and ∼3.8 times as large as that of their neutral form 2, respectively. And the βvec value of two-electron-reduced species 3 2
2 is as large as 1349.4  10
30 esu, which is ∼286.1 times larger than that of the neutral form. Therefore, we conclude that the D
π
A system BNbpy containing three-coordinate organoboron and its metal Pt(II) chelate complex could be excellent redox switchable NLO materials. ’ ASSOCIATED CONTENT

bS

Molecular orbitals of 1F
, 1CN , 2F , and 2CN involved in the dominant electron transitions (Figure S1 and S2), DOS for 122
(Figure S3), Mulliken populations of atomic orbitals (Table S1), and the frequency dependent first hyperpolarizability values (Table S2) for studied systems. This information is available free of charge via the Internet at http://pubs.acs.org. Supporting Information.








’ AUTHOR INFORMATION Corresponding Author

*Fax: (+86) 431 85098768. E-mail: [email protected].

’ ACKNOWLEDGMENT We gratefully acknowledge the financial support from the National Natural Science Foundation of China (No. 20873017) 13571

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572

The Journal of Physical Chemistry A and the Natural Science Foundation of Jilin province (20101154).

’ REFERENCES (1) (a) Williams, D. J.; Prasad, P. Introduction to Nonlinear Optical Effects in Molecules and Polymers; Wiley: New York, 1991. (b) Dalton, L. R.; Harper, A. W.; Ghosn, R.; Steier, W. H.; Ziari, M.; Fetterman, H.; Shi, Y.; Mustacich, R. V.; Jen, A. K.-Y.; Shea, K. J. Chem. Mater. 1995, 7, 1060–1081. (c) Marks, T. J.; Ratner, M. A. Angew. Chem., Int. Ed. Engl. 1995, 34, 155–173. (d) Di Bella, S.; Fragal
a, I.; Ledoux, I.; Diaz-Garcia, M. A.; Marks, T. J. J. Am. Chem. Soc. 1997, 119, 9550–9557. (e) Nakano, M.; Fujita, H.; Takahata, M.; Yamaguchi, K. J. Am. Chem. Soc. 2002, 124, 9648–9655. (f) Srinivas, K.; Sitha, S.; Jayathirtha Rao, V.; Bhanuprakash, K. Optical Materials. 2006, 28, 1006–1012. (2) Hudson, Z. M.; Wang, S. N. Acc. Chem. Res. 2009, 42, 1584–1596. (3) (a) Glogowski, M. E.; Williams, J. L. R. J. Organomet. Chem. 1981, 218, 137–146. (b) Williams, J. L. R.; Grisdale, P. J.; Doty, J. C. Org. Chem. 1971, 36, 544–549. (c) Doty, J. C.; Babb, B.; Grisdale, P. J.; Glogowski, M. E.; Williams, J. L. R. J. Organomet. Chem. 1972, 38, 229–236. (d) Glogowski, M. E.; Williams, J. L. R. J. Organomet. Chem. 1981, 216, 1–8. (e) Glogowski, M. E.; Zumbulyadis, N.; Williams, J. L. R. J. Organomet. Chem. 1982, 231, 97–107. (f) Kaim, W.; Schultz, A. Angew. Chem., Int. Ed. Engl. 1984, 23, 615–616. (g) Schultz, A.; Kaim, W. Chem. Ber 1989, 122, 1863–1868. (h) Fiedler, J.; Zalis, S.; Klein, A.; Hornung, F. M.; Klaim, W. Inorg. Chem. 1996, 35, 3039–3043. (4) (a) Entwistle, C. D.; Marder, T. B. Chem. Mater. 2004, 16, 4574–4585. (b) Yuan, Z.; Collings, J. C.; Taylor, N. J.; Marder, T. B.; Jardin, C.; Halet, J.-F. J. Solid State Chem. 2000, 154, 5–12. (c) Entwistle, C. D.; Marder, T. B. Angew. Chem., Int. Ed. 2002, 41, 2927–2931. (5) Sun, Y.; Wang, S. N. Inorg. Chem. 2009, 48, 3755–3767. (6) (a) Cheng, L. T.; Tam, W.; Eaton, D. F. Organometallics 1990, 9, 2856–2857. (b) Kains, D. R.; Lacroix, P. G.; Ratner, M. A.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10089–10102. (c) Roberto, D.; Ugo, R.; Bruni, S.; Cariati, E.; Cariati, F.; Fantucci, P. C.; Invernizzi, I.; Quici, S.; Ledoux, I.; Zyss, J. Organometallics 2000, 19, 1775–1788. (d) Coe, B. J.; Jones, L. A.; Harris, J. A.; Brunschwig, B. S.; Asselberghs, I.; Clays, K.; Persoons, A.; Garín, J.; Orduna, J. J. Am. Chem. Soc. 2004, 126, 3880–3891. (e) Coe, B. J.; Harries, J. L.; Helliwell, M.; Jones, L. A.; Asselberghs, I.; Clays, K.; Brunschwig, B. S.; Harris, J. A.; Garín, J.; Orduna, J. J. Am. Chem. Soc. 2006, 128, 12192–12204. (7) (a) Coe, B. J.; Harris, J. A.; Jones, L. A.; Brunschwig, B. S.; Song, K.; Clays, K.; Garín, J.; Orduna, J.; Coles, S. J.; Hursthouse, M. B. J. Am. Chem. Soc. 2005, 127, 4845–4859. (b) Le Bouder, T.; Maury, O.; Bondon, A.; Costuas, K.; Amouyal, E.; Ledoux, I.; Zyss, J.; Le Bozee, H. J. Am. Chem. Soc. 2003, 125, 12284–12299. (c) Maury, O.; Viau, L.; Senechal, K.; Corre, B.; Guegan, J.-P.; Renouard, T.; Ledoux, I.; Zyss, J.; Le Bozee, H. Chem.—Eur. J. 2004, 10, 4454–4466. (d) Baccouche, A.; Peigne, B.; Ibersiene, F.; Hammout
ene, D.; Boutarfa€ia, A.; Boucekkine, A.; Feuvrie, C.; Maury, O.; Ledoux, I.; Le Bozec, H. J. Phys. Chem. A. 2010, 114, 5429–5438. (8) (a) LeCours, S. M.; Guan, A.-W.; DiMagno, S. G.; Wang, C. H.; Therien, M. J. J. Am. Chem. Soc. 1996, 118, 1497–1503. (b) de la Torre, G.; Vzquez, P.; Agull-Lpez, F.; Torres, T. Chem. Rev. 2004, 104, 3723–3750. (c) Liu, C. G.; Guan, W.; Song, P.; Yan, L. K.; Su, Z. M. Inorg. Chem. 2009, 48, 6548–6554. (9) (a) Lacroix, P. G.; Lepetit, C.; Daran, J. C. New J. Chem. 2001, 25, 451–457. (b) Spassova, M.; Kolev, T.; Kanev, I.; Jacquemin, D.; Champagne, B. J. Mol. Struct. (THEOCHEM) 2000, 528, 151–157. (c) Muhammad, S.; Xu, H. L.; Janjua, M. R. S. A.; Su, Z. M.; Nadeem, M. Phys. Chem. Chem. Phys. 2010, 12, 4791–4799. (10) (a) Coe, B, J. Acc. Chem. Res. 2006, 39, 383–393. (b) Liu, C. G.; Guan, W.; Song, P.; Yan, L. K.; Su, Z. M. Inorg. Chem. 2009, 48 6548–6554. (11) (a) Loucif-Sa€ibi, R.; Nakatani, K.; Delaire, J. A.; Dumont, M.; Sekkat, Z. Chem. Mater. 1993, 5, 229–236. (b) Nakatani, K.; Delaire, J. A. Chem. Mater. 1997, 9, 2682–2684.

ARTICLE

(12) (a) Green, K. A.; Cifuentes, M. P.; Christopher Corkery, T.; Samoc, M.; Humphrey, M. G. Angew Chem. Int. Ed. 2009, 48, 7867–7870. (b) Paul, F.; Costuas, K.; Ledoux, I.; Deveau, S.; Zyss, J.; Halet, J. F.; Lapinte, C. Organometallics 2002, 21, 5229–5235. (13) (a) Yamaguchi, S.; Shirasaka, T.; Akiyama, S.; Tamao, K. J. Am. Chem. Soc. 2002, 124, 8816–8817. (b) Zhou, G.; Baumgarten, M.; M€ullen, K. J. Am. Chem. Soc. 2008, 130, 12477–12484. (c) Sun, Y.; Wang, S. N. Inorg. Chem. 2010, 49, 4394–4404. (14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.1; Gaussian, Inc.: Wallingford, CT, 2009. (15) (a) Miertuss, S.; Tomasi, J. Chem. Phys. 1982, 65, 239–245. (b) Cossi, M.; Barone, V. J. Chem. Phys. 2000, 112, 2427–2435. (16) Sim, F.; Chin, S.; Dupuis, M.; Rice, J. E. J. Phys. Chem. 1993, 97, 1158–1163. (17) Chopra, P.; Carlacci, L.; King, H. F.; Prasad, P. N. J. Phys. Chem. 1989, 93, 7120–7130. (18) Jacquemin, D.; Perp
ete, E. A.; Medved’, M.; Scalmani, G.; Frisch, M. J.; Kobayashi, R.; Adamo, C. J. Chem. Phys. 2007, 126, 191108(1
4). (19) (a) Lin, J.; Wu, K. C.; Zhang, M. X. J. Comput. Chem. 2009, 30, 2056–2063.(b) Accoding to the parameter in ref 19a, we worked out the iops keywords: IOp(3/76=1000004000), IOp(3/77= 0600006000), IOp(3/78=1000010000). (20) Yanai, T.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393, 51–57. (21) Jacquemin, D.; Perp
ete, E. A.; Ciofini, I.; Adamo, C. J. Comput. Chem. 2008, 29, 921–925. (22) Lewis, A.; Bumpus, J. A.; Truhlar, D. G.; Cramer, C. J. J. Chem. Educ. 2004, 81, 596–603. (23) (a) Rudberg, E.; Salek, P.; Helgaker, T.; Ågren, H. J. Chem. Phys. 2005, 123, 184108(1
7). (b) Peach, M. J. G.; Helgaker, T.; Keal, T. W.; Lutnæs, O. B.; Tozer, D. J.; Handy, N. C. Phys. Chem. Chem. Phys. 2006, 8, 558–562. (24) (a) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865–3868. (b) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78, 1396–1396. (25) (a) Adamo, C.; Barone, V. Chem. Phys. Lett. 1999, 314, 152–157. (b) Adamo, C.; Scuseria, G. E.; Barone, V. J. Chem. Phys. 1999, 110, 6158–6170. (26) (a) Cundari, T. R.; Kurtz, H. A.; Zhou, T. J. Phys. Chem. A 2000, 104, 4711–4717. (b) Inerbaev, T. M.; Belosludov, R. V.; Mizuseki, H.; Takahashi, M.; Kawazoe, Y. J. Chem. Theory Comput. 2006, 2, 1325–1334. (27) (a) Oudar, J. L.; Chemla, D. S. J. Chem. Phys. 1977, 66, 2664–2668. (b) Oudar, J. L. J. Chem. Phys. 1977, 67, 446–457. (28) (a) Sekino, H.; Bartlett, R. J. J. Chem. Phys. 1986, 85, 976–989. (b) Rice, J. E.; Amos, R. D.; Colwell, S. M.; Handy, N. C.; Sanz, J. J. Chem. Phys. 1990, 93, 8828–8839.

13572

dx.doi.org/10.1021/jp206003n |J. Phys. Chem. A 2011, 115, 13564–13572