Synthesis, photophysical studies, solvatochromic analysis and TDDFT calculations of diazaspiro compounds

Accepted Manuscript

Synthesis, photophysical studies, solvatochromic analysis and TDDFT calcu-

lations of Diazaspiro compounds

Komal Aggarwal, Jitender M. Khurana

PII:

S1386-1425(15)00146-8

http://dx.doi.org/10.1016/j.saa.2015.02.004

SAA 13293

DOI:

Reference:

Spectrochimica Acta Part A: Molecular and Biomo-

lecular Spectroscopy

To appear in:

Received Date:

Revised Date:

Accepted Date:

6 June 2014

31 January 2015

2 February 2015

Please cite this article as: K. Aggarwal, J.M. Khurana, Synthesis, photophysical studies, solvatochromic analysis

and TDDFT calculations of Diazaspiro compounds, Spectrochimica Acta Part A: Molecular and Biomolecular

Spectroscopy (2015), doi: http://dx.doi.org/10.1016/j.saa.2015.02.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis, photophysical studies, solvatochromic analysis and TDDFT calculations of

Diazaspiro compounds

Komal Aggarwal and Jitender M. Khurana*

Department of Chemistry, University of Delhi, Delhi-110007, India

*Corresponding author. Tel.: +91 11 2766772; fax: +91 1 27667624

E-mail addresses: jmkhurana1@yahoo.co.in; jmkhurana@chemistry.du.ac.in

Abstract

Diazaspiro[5.5]undecane-1,3,5,9-tetraones and 3-thioxo-diazaspiro[5.5]undecane-1,5,9-trione

have been synthesized via double Michael addition of 1,5-diphenyl-1,4-pentadien-3- one with

active methylene heterocycles N,N-dimethyl barbituric acid, barbituric acid and thiobarbituric

acid in water:ethanol (1:1) using TBAB as catalyst. The solvent effect on photophysical

behavior of these compounds showed that stokes shift increases with increase in polarity of

solvents. The solvent effect on the spectral properties has been investigated by using the

Lippert–Mataga and Reichardt methods. The solvatochromism is analyzed by linear solvation

energy relationship using the new four-parameter Catalán polarity scales. The relative

fluorescence quantum yield of these diazaspiro compounds varies in solvents of different

polarity. The HOMO and LUMO energies have been calculated by TDDFT (B3LYP/6-311G

(d, p)) approach. TDDFT calculations were also used to compare the experimental and

theoretical absorption spectra.

Keywords

Diazaspiro compounds, Catalán polarity scale, dipole moment, HOMO-LUMO, TDDFT

1. Introduction

The effect of solvent on the absorption and fluorescence properties of organic compounds has

been a subject of several investigations [1-5]. These investigations have considerable

importance in the field of photophysical studies and photochemistry. The effect of solvents on

absorption and fluorescence spectra can be used to determine the electric dipole moment of

solute molecules which is a direct measure of the electron distribution in a molecule. The

excited state dipole moment of a molecule reveals information on the electronic and

geometrical structures of the molecule in the short-lived excited state. Knowledge of dipole

moments of first electronically excited singlet state of the molecules is quite useful in

designing materials with nonlinear optical properties [6], and elucidating the nature of the

1

excited states. It also indicates the charge distribution in the molecule and is also useful for

predicting the site of attack by electrophilic and nucleophilic reagents in some photochemical

reactions.

Solvatochromism is an experimentally simple and widely accepted method [7-12], as it does

not use any external field [13-14]. This method [15-18], based on the shift of the absorption

and fluorescence maxima in various media, provides a reliable way to estimate ground and

excited state dipole moment compared to conventional methods like electric dichroism [19],

fluorescence polarisation [20], stark splitting of rotational levels [21], microwave

conductivity[22], and thermochromic shift method [23].

Extensive photophysical investigation of newly synthesized fluorophores is an essential tool

to extract the necessary information about the molecules. Heterocyclic spiro compounds

containing nitrogen have interesting conformational features and also show pharmaceutical

properties [24-25]. Barbitutrates have shown a wide array of CNS drugs such as sedative,

anticonvulsant, anxiolytic, anesthetic, antiepileptic [26-28]. They have also been used as a

disperse dye with strong fluorescence, as yellow organic pigment [29] and investigated as

stain developers for the identification of nucleic acids [30-31]. Similarly,

diazaspiro[5.5]undecane-1,3,5,9-tetraones and 3-thioxo-2,4 diazaspiro[5.5]undecane-1,5,9-

triones show wide range of biological and therapeutic properties such as antibacterial [32],

fungicidal [33], potent sedative-hypnotic [34], CNS depressants [35] and anticonvulsant [36].

The various biological applications of barbiturates and diazaspiro compounds encouraged us

to investigate the solvatochromic behaviour of these molecules.

Literature survey revealed that there are no reports available on the photophysical studies and

determination of µ_gand µ_evalues of the investigated molecules. In this paper, we report a

simple

and

efficient

synthesis

of

diazaspiro

compounds

containing

barbiturates/thiobarbiturates via the double Michael addition of dibenzalacetone derivatives

with N,N-dimethylbarbituric acid/barbituric acid/thiobarbituric acid in ethanol: water (1:1)

using 10 mol% of TBAB under reflux. Solvatochromic behaviour of three diazaspiro

compounds with different substituents on barbituric acid moiety was studied. The results were

used to obtain the ground and excited state dipole moments. The solvent effects were

analyzed using Lippert-Mataga, Reichardt equation and also with the new four-parameter

Catalán polarity scales. We have also carried out the TDDFT (time-dependent density

functional theory) calculations of spiro derivatives to calculate the electronic excitation

energies and a comparative study of the experimental absorption spectra with TDDFT

calculations.

2

2. Experimental

All the starting materials were of GR (Guaranteed Reagent) quality of Merck and all solvents

used were HPLC grade. Melting points were determined on a Tropical Lab equip apparatus

1

and are uncorrected. The H NMR spectrum was recorded on Jeol JNM ECX-400P at 400

MHz using TMS as an internal standard. The chemical shift values are recorded on δ scale

and the coupling constants (J) are in Hertz. Solvents were checked in steady-state

fluorescence apparatus for any fluorescence impurities in the wavelength ranges of interest.

UV-vis absorption spectra were recorded on an Analytikjena Specord 250 spectrophotometer.

Fluorescence spectra were measured at Cary Eclipse fluorescence spectrophotometer. All

spectroscopic measurements were done at room temperature. HPLC purification was done on

Shimadzu analytical HPLC system of synthesized diazaspiro compounds using 90:10 (water:

acetonitrile) solvent system with flow rate 1 mL/min. The photophysical properties of the

diazaspiro compounds (3a-c) were recorded in dilute concentrations (5 × 10⁻⁵M).

2.1. General procedure for the synthesis of diazaspiro compounds

In a general procedure, dibenzylidene acetone (1 mmol), N.N-dimethyl barbituric acid/

barbituric acid/ thiobarbituric acid (1 mmol) and 4 mL of ethanol: water (1:1) were taken in a

50 mL round-bottomed flask. 10 mol% of Tetrabutyl ammonium bromide (TBAB) was added

to the mixture, and the contents were stirred. The reaction mixture was refluxed and the

progress of the reaction was monitored by TLC using ethyl acetate: petroleum ether (30: 70)

as eluent for disappearance of active methylene compounds. After completion of the reaction,

the reaction mixture was allowed to cool to room temperature and diluted with water (5 mL).

The solid obtained was filtered at pump and washed with water: ethanol (2:1). The product

was recrystallized with ethanol. The products were characterized by their spectral data.

2.2. Spectral data

2.2.1. 7,11-Diphenyl-2,4-diazaspiro[5.5]undecane-1,3,5,9-tetraone (3a)[37(a)]

1

White solid, Yield: 86%, M.p. 275-277°C (Lit. M.p. 272-273°C); H NMR (400 MHz,

DMSO-d₆) : 2.44, 2.48 (dd, 2H, J = 15.4 Hz, 4.4 Hz), 3.49 (t, 2H, J = 14.6 Hz), 3.96, 3.99

(dd, 2H, J = 14.3 Hz, 4.4 Hz), 7.13 (d, 4H, Ar-H, J = 7.3 Hz), 7.27 - 7.34 (m, 6H, Ar-H),

11.20 (s, 1H, -NH), 11.46 (S, 1H, -NH).

2.2.2. 2,4-Dimethyl-7,11-diphenyl-2,4-diazaspiro[5.5]undecane-1,3,5,9-tetraone (3b) )[33]

3

1

White solid, Yield: 89%, M.p. 150-152^oC (Lit. M.p. 150-152°C); H NMR (400 MHz,

CDCl₃) δ : 2.59, 2.63 (dd, 2H, J = 14.6 Hz, 4.4 Hz), 2.86 (s, 3H, -NCH₃), 3.01 (s, 3H, -

NCH₃), 3.69 (t, 2H, J = 14.6 Hz), 3.99, 4.03 (dd, 2H, J = 13.9 Hz, 4.4 Hz), 7.05-7.07 (m, 4H,

Ar-H), 7.23-7.26 (m, 6H, Ar-H).

2.2.3. 7,11-Diphenyl-3-thioxo-2,4-diazaspiro[5.5]undecane-1,5,9-trione (3c) )[37(b)]

1

White solid, Yield: 84%, M.p. 254-256°C (Lit. M.p. 250-252°C); H NMR (400 MHz,

CDCl₃) δ : 2.62, 2.66 (dd, 2H, J = 15.4 Hz, 4.4 Hz), 3.64 (t, 2H, J = 15.4 Hz), 3.95, 3.99 (dd,

2H, J = 14.7 Hz, 4.4 Hz), 7.12-7.15 (m, 4H, Ar-H), 7.27 - 7.28 (m, 6H, Ar-H), 8.35 (s, 1H, -

NH), 8.55 (S, 1H, -NH).

2.3. Computational details

The hybrid density functional B3LYP (Becke–Lee–Young–Parr composite of exchange-

correction functional) method [38] and the standard 6–311G (d, p) basis set were used for

structure optimization. All the theoretical calculations are performed with GAUSSIAN 09W

software package [39]. The theoretical electronic properties (HOMO–LUMO energies,

absorption wavelengths) were performed using the TDDFT [40-42] in vacuum using the same

B3LYP level and basis set.

2.4. Methods

Experimental calculations of ground state and excited state dipole moments

We have employed the fluorescence solvatochromic shift method [15-18] to measure the

stabilization of the excited states of 3a-c. The change of magnitudes for dipole moments

between ground and excited states, i.e., ∆µ = µ_e- µ_g, can be estimated by the Lippert-Mataga

equation (1) and expressed as [43-44]:

ꢂ

ꢆ^ꢇ

2 ꢃ_ꢄ− ꢃ_ꢅ

ῡ_ꢀ− ῡ_ꢁ=

(∆ꢋ) + constantꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(1)

hcꢈ_ꢉ^ꢊ

where, ῡ_ꢀand ῡ_ꢁare the wavenumbers of the absorption and emission maxima respectively.

ꢍ − 1

2ꢍ + 1 2ꢎ^ꢇ+ 1

ꢎ^ꢇ− 1

ꢇ

( )

∆ꢋ = ꢋ ꢍ − ꢋ(ꢎ ) =

−

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(2)

When specific fluorophore/solvent interactions such as hydrogen bonding or electron-pair

donor/electron-pair acceptor interactions also contribute significantly in addition to the non-

specific interactions, then the Lippert-Mataga equation is no longer applicable. In that case,

the dipole moment change between the excited and ground state can be estimated by

4

correlating the Stokes shift with the microscopic solvent polarity parameter ꢏ_ꢐ^ꢑas proposed by

Reichardt [45] and developed by Ravi et al. [46].

In empirical molecular-microscopic solvent polarity parameter(ꢏ_ꢐ^ꢑ), the problem associated

with the Onsager’s radius estimation can be minimized since a ratio of two Onsager’s radii is

involved according to eq. (3):

ꢇ

ꢊ

ꢒꢃ

ꢈ

ꢖ ꢗ ^ꢕꢘ ꢙ

ꢈ

ꢓꢔ

ꢒῡ = ῡ_ꢀ− ῡ_ꢁ= 11307.6

ꢏ_ꢐ^ꢑ+ ꢚꢛꢎꢜꢝꢈꢎꢝꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(3)

ꢒꢃ_ꢕ

where ∆ꢃ_ꢕꢌ(9 D) and a_B(6.2 A°) are the dipole moment change (∆µ = µ_e- µ_g) and Onsager’s

radius, respectively, for a pyridinium N-phenolate betaine dye used to determine the

ꢏ_ꢐ^ꢑvalues, whereas ∆µ and a are the corresponding quantities for the molecule under study.

The change in dipole moment ∆µ can be evaluated from the slope of the Stokes shift versus

ꢏ_ꢐ^ꢑplot and is given by eq. (4):

ꢟꢌ × ꢌ81

ꢂ

ꢆ

ꢞ

=

ꢒꢃ = ꢃ_ꢄ− ꢃ_ꢅ

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(4)

(

ꢠ )^ꢊ

11307.6ꢌ 6.2 ꢈ

where ‘m’ is the slope obtained from the plot of Stokes shift versus microscopic solvent

polarity (ꢏ_ꢐ^ꢑ) using eq. (3).

The following equations have been used to determine the excited state (ꢃ_ꢄ) and ground state

(ꢃ_ꢅ) dipole moments by solvatochromic method. These equations have been obtained by

employing the quantum-mechanical second order perturbation theory of a spherical solute and

also considering the Onsager’s model of reaction field for a polarizable dipole moment.

Bakhshiev’s equation [10]

(

)

ῡ_ꢀ− ῡ_ꢁ= ꢟ_ꢡꢢ ꢍ, ꢎ + ꢚꢛꢎꢜꢝꢈꢎꢝꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(5)

ꢡ

Kawski-Chamma-Viallet equation [11-12]

ῡ_ꢀ+ ῡ_ꢁ

(

)

= −ꢟ_ꢇꢢ ꢍ, ꢎ + ꢚꢛꢎꢜꢝꢈꢎꢝꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(6)

ꢇ

2

where, ῡ_ꢀand ῡ_ꢁare wavenumbers of the absorption and emission maxima, respectively, and

ꢎ^ꢇ+ 1 ꢍ − 1 ꢎ^ꢇ− 1

ꢎ^ꢇ+ 2 ꢍ + 2 ꢎ^ꢇ+ 2

(

)

ꢓ

ꢙ

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(7)

ꢢ ꢍ, ꢎ =

−

ꢡ

2ꢎ^ꢇ+ 1 ꢍ − 1 ꢎ^ꢇ− 1

2(ꢎ^ꢇ+ 2) ꢍ + 2 ꢎ^ꢇ+ 2

3

2

ꢎ^ꢣ− 1

(

)

ꢓ

ꢙ

ꢓ

ꢙ

ꢌꢌꢌꢌꢌꢌꢌꢌ(8)

ꢇ

ꢢ_ꢇꢍ, ꢎ =

−

+

ꢇ

(

)

ꢎ + 2

5

where, F₁(ε, n) and F₂(ε, n) are Bakhshiev polarity function and Kawski-Chamma-Viallet

polarity function, respectively.

ꢂ

ꢆ^ꢇ

2 ꢃ_ꢄ− ꢃ_ꢅ

ꢟ_ꢡ=

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(9)

ℎꢚꢈ_ꢉ^ꢊ

ꢇ

ꢂ

^ꢇꢆ

2 ꢃ_ꢄ− ꢃ_ꢅ

ℎꢚꢈ_ꢉ^ꢊ

ꢟ_ꢇ=

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(10)

where, m₁and m₂are the slopes which can be calculated from equations (5) and (6),

respectively, h is Planck’s constant, c is the speed of light in vacuum, a_ois the radius of the

cavity in which the fluorophore resides, and µ_eand µ_gare the dipole moments in the excited

and ground states, respectively, ε and n are the dielectric constant and the index of refraction

of the solvents, respectively. Onsager cavity radii (a_o) for investigated molecules were

determined theoretically according to their optimized geometry using DFT at B3LYP/6-311G

(d, p) basis set.

If the ground state and excited state dipole moments are parallel, the following expressions

are obtained on the basis of (9) and (10):

ꢡ/ꢇ

^ꢉꢦ

ꢟ_ꢇ− ꢟ_ꢡℎꢚꢈ^ꢊ

ꢤ

ꢥ

ꢃ_ꢅ=

ꢃ_ꢄ=

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(11)

2

2ꢟ_ꢡ

ꢡ/ꢇ

ꢟ_ꢇ+ ꢟ_ꢡℎꢚꢈ^ꢊ

^ꢉꢦ

2ꢟ_ꢡ

ꢤ

ꢥ

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(12)

2

Therefore, the ratio of excited state dipole moment to the ground state dipole moment can be

expressed as (eq. 13):

ꢃ_ꢄ

ꢃ_ꢅ

ꢟ_ꢇ+ ꢟ

^ꢡꢧ

ꢧ

=

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(13)

ꢟ_ꢇ− ꢟ_ꢡ

For the determination of the relative fluorescence quantum yield (Φ_f) of 3a, 3b and 3c in

solvents having different polarity and hydrogen bonding ability, only dilute solutions with an

absorbance below 0.1 at the excitation wavelength (λ_ex= 270 nm for 3a, λ_ex= 272 nm for 3b

and λ_ex= 280 nm for 3c) were used. For the Φ_fdeterminations of all the three compounds, 2-

amino pyridine in 1N sulphuric acid (Φ_f= 0.60) was used as fluorescence standard and

calculated on the basis of the equation (14) [46], where n₀and n are the refractive indices of

the solvents, A⁰and A are the absorbances, ꢨ_ꢁ^ꢉand ꢨ_ꢁare the fluorescence quantum yields and

6

the integrals denote the area of the fluorescence band for the standard and the sample,

respectively.

ꢎ^ꢇꢪ^ꢩ

ꢬ

ꢂ

ꢆ

ꢭ_ꢁꢮꢭ_ꢁ

ꢫ

ꢁ

Ф_ꢁ= Ф_ꢁ^ꢩ

ꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌꢌ(14)ꢌꢌꢌꢌꢌꢌꢌ

ꢎ^ꢇꢌꢪ

ꢬ

^ꢩꢂ

ꢆ

ꢭ_ꢁꢮꢭ_ꢁ

ꢁ

ꢫ

ꢩ

3. Results and discussion

Nitrogen containing spiro heterocycles having barbituric acid, N,N-dimethyl barbituric acid

and thiobarbituric acid moiety were particularly chosen in this study. The diazaspiro

compounds were synthesized as outlined in Scheme 1. The molecular structures of all three

diazaspiro compounds (3a-c) as identified by spectral data and the optimized molecular

geometries calculated by quantum chemical calculations at DFT/B3LYP (6-311G (d,p)) are

shown in Fig. 1 and Fig. 2, respectively.

O

R'

10 mol% TBAB

N

+

H₂O: EtOH(1:1)

reflux

X

O

N

R'

2

X

N

O

R'

1

3

R' = H, X= O (3a)

R' = CH₃, X = O (3b)

R' = H, X= S (3c)

R' = H, X= O (2a)

R' = CH₃, X = O (2b)

R' = H, X= S (2c)

Scheme 1: Double Michael addition of dibenzylidene acetone with barbituric acid, N,N-

dimethyl and thiobarbituric acid.

O

H

S

H

O

H₃C

O

N

H

N

H

N

O

CH₃

3c

3a

3b

Fig. 1: Structures of compounds 3a, 3b and 3c.

7

Fig. 2: Optimized geometries for the lowest-lying state (S_o) of 3a, 3b and 3c at the B3LYP/6-

311G (d,p) level.

8

UV-visible absorption maxima (λ_abs/nm), fluorescence emission maxima (λ_em/nm), Stokes

shift (ῡ_a- ῡ_f/cm^-1) and quantum yield (Φ_f) of 3a, 3b and 3c in a series of solvents of varying

polarity are given in Table 1.

Table 1: UV-vis spectroscopic and photophysical properties of 3a, 3b and 3c in different

solvents

Comp.

Solvents

ῡ_a/cm^-1

(λ_abs/nm)

ῡ_f/cm^-1

(λ_em/nm)

ῡ_a- ῡ_f/cm^-1(ῡ_a+ ῡ_f)/2 cm^-1

Φ_f

No.

3a

MeOH

ACN

36765 (272)

24096 (415)

12669

12552

12451

12330

12285

11987

11859

30430

30489

30674

30765

30894

31181

31245

0.076

0.082

0.086

0.095

0.077

0.069

0.065

36765 (272)

36900 (271)

37037 (270)

37175 (269)

24213 (413)

24449 (409)

24630 (406)

24752 (404)

25188 (397)

25316 (395)

DMF

DCM

CHCl₃

DIOX

n-HEPT

3b

MeOH

ACN

36232 (276)

36363 (275)

36232 (276)

36496 (274)

36765 (272)

37037 (270)

24038 (416)

24271 (412)

24154 (414)

24449 (409)

24752 (404)

24813 (403)

25125 (398)

12192

12092

12077

12047

12012

11951

11919

30134

30317

30192

30472

30758

30788

31081

0.067

0.050

0.056

0.074

0.062

0.052

0.045

DMF

DCM

CHCl₃

DIOX

n-HEPT

3c

MeOH

ACN

35971 (278)

35587 (281)

35461 (282)

35714 (280)

23752 (421)

23980 (417)

24154 (414)

24213 (413)

24509 (408)

24630 (406)

25380 (394)

12219

11607

11433

11248

10952

10831

10333

29861

29780

29870

29837

29985

30045

30547

0.081

0.062

0.073

0.078

0.058

0.046

0.042

DMF

DCM

CHCl₃

DIOX

n-HEPT

3.1. Solvent effect on the absorption and fluorescence spectra

The solvatochromic properties of 3a, 3b and 3c, their UV–vis and fluorescence spectra were

measured in solvents of different polarity. Typical absorption and fluorescence spectra of 3a,

3b and 3c in different solvents are shown in Figs. 3, 4 and 5, respectively. All the three

diazaspiro compounds (3a-c) showed a broad absorption band in 250–300 nm spectral range

with a peak position centered at 270 nm. The absorption band showed moderate sensitivity

(4-10 nm) towards both solvent polarity and the substituents. Thus it can be concluded that

the substitution and solvent effects are negligible in the ground state of the molecules.

9

On increasing solvent polarity from n-heptane to methanol, the emission peak maxima of 3a,

3b and 3c are more readily shifted towards longer wavelengths i.e. for 3a it red shifted from

395 to 415 nm, for 3b from 398 to 416 nm and for 3c from 394 to 421 nm corresponding to

20, 18 and 23 nm bathochromic shift of the emission maximum of the 3a, 3b and 3c,

respectively, confirming a π- π* transition. The shift of the emission maxima towards longer

wavelengths could be due to the marked difference between the excited state and ground state

charge distribution in the solute, resulting in stronger interaction with polar solvents in the

excited state and also to a higher dipole moment in the excited state.

Fig. 3: Absorption and fluorescence spectra of compound 3a in MeOH, ACN, DMF, DCM,

CHCl₃, DIOX and n-HEPT. Excitation wavelength: 270 nm.

Fig. 4: Absorption and fluorescence spectra of compound 3b in MeOH, ACN, DMF, DCM,

CHCl₃, DIOX and n-HEPT. Excitation wavelength: 272 nm.

10

Fig. 5: Absorption and fluorescence spectra of compound 3c in MeOH, ACN, DMF, DCM,

CHCl₃, DIOX and n-HEPT. Excitation wavelength: 280 nm.

It can be observed from Table 1 that increasing solvent polarity from n-heptane to methanol

resulted in an increase of Stokes shift by 810, 273 and 1886 cm^-1for 3a, 3b and 3c,

respectively, which is indicative of charge transfer transition and increase in the dipole

moment on excitation. The large magnitude of Stokes shift also indicates that the excited state

geometry could be different from ground state. It can be inferred from Table 1 that the higher

Stokes shift observed in polar protic methanol compared to polar aprotic acetonitrile,

indicates that hydrogen bonding interactions predominate over dipolar interactions.

3.2. Correlation of Solvatochromic shift with solvent polarity parameters

The plot of the Stokes shift (ῡ_a- ῡ_f) as a function of ∆f is sufficiently linear with excellent

correlation for 3a-c (Fig. 6(a)). Accordingly, ∆µ = µ_e- µ_gvalues have been calculated as 5.97

D, 3.81 D and 7.93 D for 3a, 3b and 3c, respectively. The linear correlation of Stokes shifts vs

ꢏ_ꢐ^ꢑplotted for the solvents used in this study (Fig. 6(b)), is a clear evidence of the existence of

general solute-solvent interactions and H-bonding interactions in the polar solvents.

11

Fig. 6 (a) Plot of Stokes shift vs. Orientation polarizabitily (∆f) of 3a, 3b and 3c. (b) Plot of

Stokes shift vs. Riechardt’s solvent polarity function ꢏ_ꢐ^ꢑꢌof 3a, 3b and 3c.

The slopes obtained by plotting Stokes shift vs. orientation polarizability (∆f) (Lippert-

Mataga Correlation) (Fig. 6a)/ molecular-microscopic solvent polarity parameter(ꢏ_ꢐ^ꢑ)

(Reichardt Correlation) (Fig. 6b)/ solvent polarity function F₁(ε, n) (Bakhshiev Correlation)

(Fig. 7a) and spectral shift (ῡ_a+ ῡ_f)/2 vs. solvent polarity function F₂(ε, n) (Kawski–Chamma–

Viallet Correlation) (Fig. 7b), of compounds 3a, 3b and 3c are given in Table 2.

Fig. 7: (a) Variation of Stokes shift with F₁(ε, n) by using Bakhshiev’s equation in different

solvents for 3a, 3b and 3c. (b) The variation of arithmetic mean of ῡ_aand ῡ_fwith F₂(ε, n) by

using Kawski–Chamma–Viallet’s equation in different solvents for 3a, 3b and 3c.

Table 2: Slopes obtained by Lippert-Mataga, Reichardt, Bakhshiev and Kawski–Chamma–Viallet (K-

C-V) Correlations

Comp.

Slope (cm^-1)

Intercept (cm^-1)

Correlation

coefficient (r)

0.98

No. of

data

7

Lippert-Mataga Correlation

Reichardt Correlation

Bakhshiev Correlation

K-C-V Correlation

3a

3b

3c

2229.25

676.69

11897.62

11916.86

10508.35

0.93

0.95

7

6

3460.10

3a

3b

3c

1157.66

387.45

11910.78

11908.58

10374.13

0.94

0.99

7

2520.60

3a

3b

3c

749.96

227.06

1153.35

11922.50

11924.71

10545.84

0.96

0.91

0.94

7

6

3a

3b

3c

1412.55

1557.53

1250.32

31592.10

31395.68

30639.05

0.97

0.96

0.84

7

6

12

From the slope m (eq. 4, Table 2), the difference between dipole moments (∆µ = µ_e−µ_g) are

2.36, 1.58 and 3.75 D for 3a, 3b and 3c, respectively (Table 3). The linear behavior of Stokes

shift versus solvent polarity functions (F₁and F₂) indicates general solvent effects as a

function of the dielectric constant and refractive index. The slopes (m₁and m₂) were taken

into account for calculating the ground and excited states dipole moments (µ_gand µ_e) of all

the three compounds (3a-c).

The µ_gand µ_evalues obtained from (eq. 11) and (eq. 12) and the ratio of µ_eand µ_gobtained

from (eq. 13) are presented in Table 3. The change in dipole moment values, ∆µ, estimated

using the Lippert-Mataga, Reichardt’s, Bakhshiev’s and Kawski-Chamma-Viallet correlations

are also given in Table 3. The ground state dipole moments of 3a-c were also computationally

optimized by quantum chemical calculations using Gaussian 09 program [39] at DFT/

B3LYP (6-311G (d, p)) and the corresponding optimized molecular geometries are shown in

Fig. 2.

Table 3: Onsager cavity radius (a_o), ground state (µ_g), excited state (µ_e), dipole moment

change (∆µ) and dipole moment ratio (µ_e/µ_g) of compounds 3a, 3b and 3c.

ꢆ^ꢲ

ꢃ_ꢅ^ꢀ

ꢃ_ꢄ^ꢰ

ꢒꢃ^ꢱ

ꢒꢃ^ꢄ

ꢒꢃ^ꢅ

ꢒꢃ^ꢁ

ꢠ

a_o(Å)

ꢃ_ꢅ^ꢯ

ꢂ

ꢃ_ꢄꢃ_ꢅ

3a

3b

3c

5.44

6.00

5.68

2.77

3.59

1.97

1.53

6.46

0.39

4.99

8.67

4.77

5.97

3.81

7.93

2.36

1.58

3.73

3.46

2.21

4.58

4.75

5.78

4.77

3.26

1.34

12.23

Note: 1 Debye (D) = 3.335 64 × 10⁻³⁰C.m

^aCalculated by Gaussian09 software.

^bThe experimental ground states dipole moment calculated from equation (11).

^cThe experimental excited state dipole moment calculated from equation (12).

^dThe change in dipole moment calculated from Lippert’s equation (1).

^eThe change in dipole moment calculated from Reichardt’s equation (4).

^fThe change in dipole moment calculated from Bakhshiev’s equation (9).

^gThe change in dipole moment calculated from Chamma−Viallet’s equation (10).

^hThe ratio of excited state and ground state dipole moments values calculated using (13).

It is observed from Table 3 that, ∆µ values are highest by Lippert–Mataga method compared

to all other methods. This is due to the fact that, this method does not consider polarizability

of the solute. The charge separation in a compound 3b led to the highest excited state dipole

moment (Table 3). In case of compounds 3a and 3c only tautomeric structures are possible

(Fig. 8) resulting in a lower excited state dipole moment compared to 3b. All the results show

a significant increase in the dipole moment on excitation. Thus, a relatively large charge

transfer takes place during the excitation of the diazaspiro compounds.

13

O

HO

N

O

H

O

N

HO

N

H

O

N

H

N

H

O

3a

^-O

O

N

^-O

N

O

N

O

N

O

3b

O

HO

N

O

H

N

S

N

HS

N

S

N

H

O

H

3c

Fig. 8: Possible resonating or tautomeric structures of 3a, 3b and 3c.

The multi parametric approach of Kamlet and Taft has been used [48] to determine the

contribution of various modes of solvation (i.e. specific and nonspecific) towards the Stokes

shift values. This method involves the effect of individual solvent parameters (hydrogen bond

donor, hydrogen bond acceptor and dielectric effects) on the spectral properties of the

compounds. The Kamlet–Abboud–Taft relationship is given by eq. 15:

∆ν = ∆ν₀+ aα + bβ + cπ*

(15)

where α is an index of hydrogen bond donating ability of the solvent, β is an index of

hydrogen bond accepting capacity of the solvent and π* is the polarity/polarizability

parameter of the solvents. The coefficients a, b and c are measure of the sensitivity to each

individual contributing parameter. The disadvantage of the Kamlet-Abboud-Taft approach is

that the dipolarity and polarizability of the solvent are included in only one parameter (π*).

Recently, Catalán [49] proposed another multiparameter equation, which comprises two

parameters for specific and two parameters for non-specific interactions including separate

consideration of solvent polarizability (SP) and dipolarity (SdP) (eq. 16).

14

y = y₀+ a.SA + b.SB + c.SP + d.SdP

(16)

where y denoted a solvent-dependent physicochemical property in a given solvent and y₀the

statistical quantity corresponding to the value of property in the gas phase, SA and SB

correspond to the Kamlet–Taft parameter α and β.

Multivariate regression for 3a, 3b and 3c were calculated to fit the values of absorbance

maxima, emission maxima and the Stokes shift to the Catalán relationships, the estimated

Catalán coefficients (ꢈ_ꢳꢴ, ꢵ_ꢳꢶ, ꢚ_ꢳꢷ, ꢮ_ꢳꢸꢷ) and correlation coefficients (r) are listed in Table 4.

Table 4: Adjusted coefficients (y_o,ꢈ_ꢳꢴ, ꢵ_ꢳꢶ, ꢚ_ꢳꢷ, ꢮ_ꢳꢸꢷ) and correlation coefficients (r) for the

multiple linear regression analysis of the ῡ_a, ῡ_fand Stokes shift (ῡ_a- ῡ_f) of 3a, 3b and 3c as a

function of Catalán four-parameter solvent scales.

ꢹ_ꢺꢻ

ꢼ_ꢺꢽ

ꢾ_ꢺꢿ

ꣀ_ꢺꣁꢿ

y_o

y

ῡ_a

ῡ_f

r

3a

3b

3c

36629.6 163.0

24419.8 165.4

12145.5 154.6

-52.3 85.9

-389.3 87.1

355.1 81.5

145.3 72.1

250.7 73.2

-162.2 68.4

846.9 234.2

1381.4 237.6

-432.1 222.2

-471.8 43.6

-1178.3 44.2

730.2 41.3

0.99

∆ῡ

ῡ_a

ῡ_f

36837.9 332.1

24857.8 349.7

12002.1 27.7

-47.3 175.0

-244.9 184.3

191.2 14.6

-416.5 147.0

-409.5 154.8

-10.1 12.2

398.4 477.2

506.8 502.5

-133.7 39.8

-645.5 88.8

-835.9 93.5

178.0 7.4

0.99

∆ῡ

ῡ_a

ῡ_f

36471.8 489.0

25499.1 1081.2

10970.0 601.2

467.9 257.7

-661.0 569.8

1129.3 316.8

103.9 216.4

-141.7 478.5

245.8 266.1

-1296.1 707.6

-358.3 1553.6

-934.9 863.8

-57.7 130.8

-1162.3 289.3

1105.2 160.8

0.91

0.95

0.99

∆ῡ

The linear correlation between the experimental absorbance, fluorescence and Stokes shift

with the corresponding calculated values using all the parameters of Catalán four parameter

solvent scale are shown in Figs. 9, 10 and 11, respectively.

15

Fig. 9: Correlation between the experimental absorption values with the calculated values

obtained by a multicomponent linear regression using the SA, SB, SP and SdP – scale

(Catalán) solvent parameters for 3a, 3b and 3c.

Fig. 10: Correlation between the experimental fluorescence values with the calculated values

obtained by a multicomponent linear regression using the SA, SB, SP and SdP – scale

(Catalán) solvent parameters for 3a, 3b and 3c.

Fig. 11: Correlation between the experimental Stokes shift values with the calculated values

obtained by a multicomponent linear regression using the SA, SB, SP and SdP – scale

(Catalán) solvent parameters for 3a, 3b and 3c.

Compound 3a showed a satisfactory fit for the solvent dependence of absorption maxima (r =

0.99) as the absorption maxima depend mainly on the solvent polarizability (ꢹ_ꢺꢻ) out of four

Catalán parameters indicating that the increased solvent polarizability stabilizes 3a in their

ground state. According to the eq. 16, the change of absorption maximum of 3b depends on

all the four solvent parameters with good correlation r = 0.99 but the solvent polarizability

has the highest impact on the absorption maxima of 3b indicating that the absorbance maxima

shifted to lower energies with increasing polarizability of the solvents. Similarly, compound

3c also showed a good correlation for the solvent dependence of the absorption maxima r =

0.91 in which absorption maxima depend on the solvent acidity (ꢹ_ꢺꢻ) indicating that there is

hydrogen bonding between the C=O group of 3c with the hydrogen of polar protic solvents.

A good correlation of 3a, 3b and 3c was observed r = 0.99, 0.99 and 0.95, respectively for the

solvent dependence of fluorescence maxima which depends on the two solvent parameters

(

ꢾ_ꢺꢿ, ꣀ_ꢺꣁꢿ)ꢌand negligabily on other two coefficients (ꢹ_ꢺꢻ, ꢼ_ꢺꢽ). However, the solvent

16

polarizability (ꣂ_ꢺꢿ)ꢌhas higher effect than solvent dipolarity (ꣀ_ꢺꣁꢿ), suggesting that the

emission maxima of all the three compounds is stabilized by increase in solvent polarizability

(ꣂ_ꢺꢿ).

The Stokes shift of compounds 3a, 3b and 3c also showed an excellent fit with r = 0.99, 0.99

and 0.99, respectively. The Stokes shift depends not only on the dipolarity but also on the

solvent acidity. However, the impact of solvent acidity is lower than that of solvent dipolarity

indicating that the Stokes shift increases with dipolarity.

3.3. Comparison of the calculated absorption spectra with experiment

On the basis of fully optimized ground-state structures (Fig. 2) , TDDFT calculations using

B3LYP-6-311G (d, p) basic set have been used to determine the HOMO-LUMO energy gap

and also the electronic transitions for 3a, 3b and 3c. The calculated results of oscillator

strength (f), frequency, wavelength and energies of HOMO-LUMO orbitals are listed in Table

5.

Table 5: Selected vertical excitations calculated by the TDDFT (B3LYP) method for 3a, 3b

and 3c

Exp.

TD-DFT (B3LYP/6-311G (d,p))

Energies (a.u.)

λ

_exc(nm)

λ

_exc(nm)

E_exc(eV)

4.3103

4.4494

4.6789

Character

H → L+2 (57%), H → L+5 (30%) E_HOMO= -0.25795

F

3a

3b

3c

287.65

278.66

264.98

0.0006

0.0029

0.0001

270

H → L (70%)

E_LUMO= -0.01863

→ L+1 (11%), H → L (64%),

-5

H-6

∆E= -0.23932

H-1 → L (20%)

→ L+2 (56%), H → L+5 (33%)

→ L (C=O) (70%)

287.53

273.37

266.06

4.3120

4.5355

4.6600

0.0006

0.0035

0.0000

H

E_HOMO= -0.25663

E_LUMO= -0.01630

∆E= -0.24033

272

H-5 → L (65%), H-4 → L (16%),

→ L (15%)

→ L (69%)

H-1

386.73

331.52

3.2060

3.7398

0.0000

0.0043

E_HOMO= -0.23094

E_LUMO= -0.09586

∆E= -0.13508

282

H → L (69%), H-1→ L (13%),

→ L (10%)

H-2

H-2 → L (62%)

302.43

4.0996

0.0013

The TDDFT calculations in gas phase show a strong absorption band at 278 nm for 3a, 273

nm for 3b and 331 nm for 3c which are in good agreement with the experimental absorption

bands at 270 nm, 272 nm and 282 nm in chloroform, respectively. These absorption bands

17

indicate the existence of HOMO to LUMO transition. On the basis of the calculated

molecular orbital coefficients, H→L electronic transition has been assigned as π→π* for 3a,

→π* for

3b and n

3c because of the presence of lone pair on sulphur in 3c.

Fig. 12 shows 3D surface plots of the highest occupied molecular orbitals (HOMO), the

lowest unoccupied molecular orbitals (LUMO) for compounds 3a, 3b and 3c. In case of 3a

and 3b, it can be seen that the HOMOs are mainly localized on the phenyl ring, whereas the

LUMOs are localized over the barbituric acid moiety, suggesting a strong propensity for

intramolecular charge transfer from the electron-donor phenyl ring to the electron acceptor

C=O of amidic linkage, and therefore correspond to π→π* transitions. In case of 3c, HOMOs

are mainly localized over the lone pair on sulphur atom and LUMOs are localized over the

thiobarbituric acid moiety, indicating the occurance of intramolecular charge transfer from

lone pair of sulphur atom to the C=O of thiobarbituric acid, confirming the n→π* transition.

Fig. 12: The distribution of the HOMO and the LUMO over the molecules of 3a, 3b and 3c

calculated with TDDFT (B3LYP/6-311G (d,p)).

4. Conclusion

18

In conclusion, a new protocol for the synthesis of diazaspiro compounds has been developed

using TBAB in water:ethanol under reflux. Photophysical properties of three diazaspiro

compounds have been studied in the solvents of different polarity and the results showed that

these fluorophores exhibit large Stokes shifts in polar protic solvent methanol. It has been

found that the absorption and the fluorescence spectra of the compounds shift to lower energy

as solvent polarity increases, confirming a π → π* transition. Lippert–Mataga plot and the

Reichardt’s plot of these compounds in solvents of different polarity illustrated linearity with

good correlation. We found that all the three molecules possess higher dipole moment values

in the excited state than the ground state. Moreover, the new four-parameter Catalán solvent

scale which takes into account acidity, basicity, polarizability and dipolarity of the solvent,

was also applied. The solvent parameter, namely solvent polarizability (ꢾ_ꢺꢿ)ꢌcontributes

majorly in stabilizing diazaspiro compounds in the excited state. The frontier molecular

orbitals have been visualized and the HOMO–LUMO energy gap has been calculated. The

calculated HOMO and LUMO energies show that charge transfer occur within the molecule.

Acknowledgments

KA thanks CSIR, New Delhi, India for the grant of Junior and Senior Research Fellowships.

The authors also thank Gaussian for providing us Gaussian 09 software.

References

[1] M.F. Broglia, S.G. Bretolotti, C.M. Previtali, H.A. Montejano, J. Photochem. Photobiol. A

180 (2006) 143-149.

[2] V.E. Nicotra, M.F. Mora, R.A. Iglesias, A.M. Baruzzi, Dyes and Pigments 76 (2008) 315-

318.

[3] N.R. Patil, R.M. Melavanki, S.B. Kapatkar, N.H. Ayachit, J. Saravanan, J. Fluoresc. 21

(2011) 1213-1222.

[4] A.A. El-Rayyes, T. Htun, Spectrochim. Acta Part A 60 (2004) 1985-1989.

[5] (a) K. Aggarwal, J.M. Khurana, J. Photochem. Photobiol., A 276 (2013) 71-82. (b) K.

Guzow, A. Ceszlak, M. Kozarzewska, W. Wiczk, Photochem. Photobiol. Sci. 10 (2011)

1610-1621.

[6] A. Mallick, S. Maiti, B. Haldar, P. Purkayastha, N. Chattopadhyay, Chem. Phys. Lett. 371

(2003) 688-693.

[7] M.K. Saroj, N. Sharma, R.C. Rastogi, J. Mol. Struct. 1012 (2012) 73-86.

19

[8] R. M. Melavanki, N.R. Patil, S.B. Kapatkar, N.H. Ayachit, S. Umapathy, J.

Thipperudrappa, A.R. Nataraju, J. Mol. Liq. 158 (2011) 105-110

[9] M.S. Zakerhamidi, S.A. di-Kandjani, M. Moghadam, E. Ortyl, S. Kucharski, Spectrochim.

Acta Part A 85 (2012) 105-110.

[10] Y.T. Varma, S. Joshi, D.D. Pant, J. Mol. Liq. 179 (2013) 7-11.

[11] J. Jayabharathi, V. Kalaiarasi, V. Thanikachalam, K. Jayamoorthy, J. Fluoresc. 24

(2014) 599-612.

[12] J.S. Kadadevarmath, G.H. Malimath, N.R. Patil, H.S. Geethanjali, R.M. Melavank, Can.

J. Phys. 91 (2013) 1107-1113.

[13] (a) S.D. Choudhury, S. Basu, Chem. Phys. Lett. 371 (2003) 136-140; (b) R. Kakkar, V.

Katoch, J. Mol. Struct. (Theochem) 578 (2002) 169-175.

[14] A. Grofsik, M. Kubinyi, A. Ruzsinszky, T. Veszpremi, W.J. Jones, J. Mol. Struct. 555

(2000) 15-19.

[15] E. Lippert, Z. Naturforsch. 10 (1955) 541-545.

[16] N.G. Bakshiev, Opt. Spektrosk. 16 (1964) 821-832.

[17] A. Kawski, Z. Naturforsch. 57a (2002) 255-262.

[18] A. Chamma, P. Viallet, C.R. Acad. Sci. Paris Ser. C 270 (1970)1901-1910.

[19] J. Czekalla, Chimia 15 (1961) 26-31.

[20] J. Czekalla, Zeitschrift fuer Elektrochemie und Angewandte Physikalische Chemie 64

(1960) 1221-1228.

[21] J.R. Lombardi, J. Am. Chem. Soc. 92 (1970) 1831-1833.

[22] M.P. De Haas, J.M. Warman, Chem. Phys. 73 (1982) 35-53.

[23] A. Kawski, B. Kukliński, P. Bojarski, Z. Naturforsch. 55a (2000) 550-554.

[24] R Pradhan, M. Patra, A. K. Behera, B. K. Mishra, R. K. Behera, Tetrahedron 62 (2006)

779-828.

[25] (a) S. Sivakumar, R.R. Kumar, M.A. Ali, T.S. Choon, Eur. J. Med. Chem. 65 (2013)

240-248; (b) G. Bhaskar, Y. Arun, C. Balachandran, C. Saikumar, P.T. Perumal, Eur. J. Med.

Chem. 51 (2012) 79-91.

[26] L.L. Brunton, J.S. Lazo, K.L. Parker, I. Buxton, D. Blumenthal (2006) Goodman and

Gilman’s The pharmacological basis of therapeutics. 11th eds, The McGraw- Hill Companies

Inc.

[27] M.W. Johns, Drugs 9 (1975) 448-478.

[28] C. Uhlmann, W. Froscher, CNS Neurosci. Ther. 15 (2009) 24-31.

20

[29] (a) D. Theford, A.P. Chorton, J. Hardman, Dyes Pigments 59 (2003) 185-191. (b) S.

Wang, S.-H. Kim, Dyes Pigments 80 (2009) 314-320.

[30] D. Popov, B. Thorell, Stain Technol. 53 (1978) 246-247.

[31] D. Popov, B. Thorell, Stain Technol. 57 (1982) 143-150.

[32] R. K. Behera, A. K. Behera, R. Pradhan, A. Pati, M. Patra, Synth. Commun. 36 (2006)

3729-3742.

[33] R. K. Behera, A. K. Behera, R. Pradhan, A. Pati, M. Patra, Phosphorus, Sulfur Silicon

Relat. Elem. 184 (2009) 753-765.

[34] S. Kesharwani, N.K. Sahu, D.V. Kohli, Pharm. Chem. J. 43 (2009) 315-319.

[35] B. Goel, S. Sharma, K. Bajaj, D. Bansal, T. Singh, N. Malik, S. Lata, C. Tyagi, H.

Panwar, A. Agarwal, A. Kumar, Indian J Pharm Sci. 67 (2005) 194-199.

[36] A.N. Osman, M.M. Kandeel, M.M. Said, E.M. Ahmed, Indian J. Chem. Sect B. 35B

(1996) 1073-1078.

[37] (a) D.B. Ramachary, M. Kishor, G.B. Reddy, Org. Biomol. Chem. 4 (2006) 1641-1646;

(b) J.M. Khurana, R. Arora, S. Satija, Heterocycles 71 (2007) 2709-2716.

[38] A.D. Becke, J Chem Phys 98 (1993) 5648-5650.

[39] M.J. Frisch, G.W. Trucks, H.B. Schlegal, et al. (2009) GAUSSIAN 09, Revision A.02,

Gauss-ian Inc., Wallingford, CT.

[40] S. Miertus, E. Scrocco, J. Tomasi, Chem. Phys. 55 (1981) 117-129.

̃

[41] S. Miertus, J. Tomasi, Chem. Phys. 65 (1982) 239-245.

̃

[42] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 105 (2005) 299-3000.

[43] E. Lippert, Z Elektrochem 61 (1957) 962-975.

[44] N. Mataga, Y. Kaifu, M. Koizumi, Bull. Chem. Soc. 29 (1956) 465-470.

[45] C. Reichardt, T. Welton, Solvents and Solvent Effects in Organic Chemistry (2011)

Wiley VCH, Weinheim.

[46] M. Ravi, T. Soujanya, A. Samanta, T.P. Radhakrishnan, J. Chem. Soc., Faraday Trans.

91 (1995) 2739-2742.

[47] S. Jana, S. Dalapati, S. Ghosh, S. Kar, N. Guchhait, J. Mol. Struct. 998 (2011) 136-143.

[48] M.J. Kamlet, J.L.M. Abboud, M.H. Abraham, R.W. Taft, J. Org. Chem. 48 (1983) 2877-

2880.

[49] J. Catalán, J. Phys. Chem. B 113 (2009) 5951-5960.

21

Figures Caption