Photophysical properties of the intramolecular excited charge-transfer states of π-expanded styryl phenanthrimidazoles – effect of solvent polarity

Abstract

Solvent-dependent electronic structures of selected donor (D)–acceptor (A) phenanthrimidazole derivatives containing styryl as an electron acceptor fragment in fluorescent charge transfer (CT) states have been investigated. Radiative charge recombination [CT → S₀] is discussed in terms of the Mulliken–Murrell model of the CT complexes and the Marcus theory of photoinduced electron transfer (ET). Solvatochromic effects on the fluorescence spectra indicate the CT character of the emitting singlet states and the analysis leads to the electron transfer in the Marcus inverted region. The fluorescence rate constants (k_r) and transition dipole moments (M) indicate that the electronic coupling between the emitting CT state and the ground state is a governing factor of the radiative transitions. Large values of M indicate a nonorthogonal geometry (confirmed by XRD) of the donor and acceptor subunits in the fluorescent states. Emission spectra of N,N-dimethyl-4-(2-styryl-1H-phenanthro[9,10-d]imidazol-1-yl)benzenamine is of interest because of the existence of dual emitting states where a locally excited state is responsible for fluorescence in nonpolar solvents. In polar solvents fluorescence is from a twisted intramolecular charge transfer (TICT) state. Density functional theory (DFT) calculations support the formation of the TICT state. The twist of the –N(CH₃)₂ group and the change in its hybridization in the excited state develops a high dipole moment and thereby stabilizes it to give the TICT fluorescence in polar solvents.

---

1. Introduction

The investigations of intramolecular electron transfer (ET) in donor (D)–acceptor (A) compounds are concerned with the electronic interactions between the lowest excited charge transfer state (CT) and the ground state (S₀ (V₀)) and/or the locally excited states (LE), localized either in the acceptor (V^A_1,3) or in the donor (V^D_1,3). A similar approach can be applied to describe the properties of the singlet CT states, e.g., the transition dipole moments for the CT absorption (¹CT ← S₀) and the fluorescence (¹CT → S₀), as well as the characteristics of the triplet ³CT states.¹ The study of the intramolecular ET reactions is to identify structural elements that promote electronic coupling between an electron donor and an acceptor, and it is a challenging one.^2–7 The values of the electronic coupling elements are determined theoretically by the interactions between the atoms forming the A–D bond.^8,9 Neglecting contributions from the σ orbitals, one can obtain for π-electronic systems

V₀ = C^A_LUMOC^D_HOMO β_AD cos(θ_A–D) + const., (1)

V^A_1,3 = C^A_HOMOC^D_HOMO β_AD cos(θ_A–D) + const., (2)

V^D_1,3 = C^A_LUMOC^D_LUMO β_AD cos(θ_A–D) + const., (3)

where θ_A–D denotes the angle between the planes of the acceptor and donor subunits and C_HOMO and C_LUMO are the Linear Combination of Atomic Orbitals (LCAO) coefficients of the 2p_Z atomic orbitals of the highest occupied molecular orbital (HOMO) and of the lowest unoccupied molecular orbital (LUMO) located on the atoms forming the A–D bond. β is the resonance integral for these AD atoms and const. is related to the electronic interactions between the remaining pairs of atoms in the D–A molecule. In the present study the development of a very high dipole moment of the molecule in the excited state at a twist angle of 90° of the –N(CH₃)₂ group accompanied by a change in the shape of the group is because of a change in the hybridization of the nitrogen atom. TICT in N,N-dimethyl-4-(2-styryl-1H-phenanthro[9,10-d]imidazol-1-yl)benzenamine can make the compound a very sensitive candidate as a biosensor and it shows dual fluorescence. A double bond is incorporated between a phenanthrimidazole ring and a N,N-dimethylaminostyryl ring in such a way that these two rings are in a trans position with respect to the double bond.¹⁰

In this article we report the elegant catalytic synthesis, characterisation and solvatochromism of phenanthrimidazoles with styryl as the acceptor units. Single crystal X-ray diffraction (XRD) supports the nonorthogonal geometry of the newly synthesised 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole. The presented results are based on a study of the solvent effects on the spectral position of absorption and fluorescence spectra as well as on the CT emission quantum yields and excited state depopulation kinetics. We have addressed the influence of solvents on the photophysical properties of the synthesized molecules in terms of hc[small nu, Greek, tilde]^vac_abs, hc[small nu, Greek, tilde]^vac_flu and (hc[small nu, Greek, tilde]^vac_abs − hc[small nu, Greek, tilde]^vac_flu) with solvent polarity function. A photophysical study on the bioactive molecule N,N-dimethyl-4-(2-styryl-1H-phenanthro[9,10-d]imidazol-1-yl)benzenamine with N(CH₃)₂ function is of interest as it shows dual fluorescence. Density functional theory (DFT) calculations have rationalised the experimental data.

2. Experimental

2.1. Spectral and cyclic voltammetric measurements

The ¹H and proton decoupled ¹³C NMR spectra of imidazole derivatives 1–3 were recorded in dimethyl sulphoxide (DMSO) at room temperature using a Bruker 400 MHz NMR spectrometer. The mass spectra of the samples were obtained using a Thermo Fischer LC-mass spectrometer in FAB mode. The UV-vis absorption and fluorescence spectra were recorded in all solvents with a PerkinElmer Lambda 35 spectrophotometer and PerkinElmer LS55 spectrofluorometer, respectively. Fluorescence lifetime measurements were carried out with a nanosecond time correlated single photon counting (TCSPC) spectrometer Horiba Fluorocube-01-NL lifetime system with Nano LED (pulsed diode excitation source) as the excitation source and TBX-PS as detector. The slit width was 8 nm and the laser excitation wavelength was 260 nm. The fluorescence decay was analyzed using DAS6 software. The quantum yield was measured in dichloromethane using coumarin 47 in ethanol as the standard.^11a–k Single crystal XRD has been recorded on an Agilent Xcalibur Ruby Gemini diffractometer. The radiation source is enhanced Mo X-ray source. The gaphite monochromator used is of 10.5081 pixels mm⁻¹ for the detector resolution. The cyclic voltammetry analyses were performed with a CHI 630A potentiostat-electrochemical analyzer at a scan rate of 100 mV s⁻¹ using 0.1 M tetra-(n-butyl)-ammonium hexafluorophosphate as supporting electrolyte with Ag/Ag⁺ (0.01 M AgNO₃) as the reference electrode and Pt electrode as the working electrode under nitrogen atmosphere at room temperature.

2.2. Solvatochromic comparison method (SCM)

To find information about the individual contribution of different solvent effects, a multi parametric approach, the solvatochromic comparison method (SCM) proposed by Kamlet et al.,¹² has been used. This approach separates the dielectric effects of solvents (π*), hydrogen-bond donor ability (α) and hydrogen-bond acceptor ability (β) of the solvents on the spectral properties. The equation describing these effects is:

E = E⁰ + cπ* + aα + bβ (4)

where, a, b, and c are the coefficients and E⁰ is the spectral maximum independent of solvent effects. The values of π*, α and β of different solvents have been taken from the report of Kamlet et al.¹² The values of E are absorption/fluorescence band maxima in terms of cm⁻¹.

2.3. Computational details

The quantum chemical calculations were performed using the Gaussian 03 (ref. 13) package. Computations of the vertical excitations, difference density plots and optimization of the ground and excited states were performed using density functional theory (DFT) and time-dependent DFT (TD-DFT) using B3LYP/6-31G(d,p) basis set, respectively. The ground and excited states’ HOMO and LUMO frontier orbitals of phenanthrimidazole derivatives were calculated by both DFT and TD-DFT methods at the B3LYP/6-31(d,p) level.

2.4. Synthesis of 2-styrylphenanthrimidazoles by InF₃

A mixture of phenanthroquinone (1 mmol), corresponding cinnamaldehyde (1 mmol), substituted aniline (1 mmol), ammonium acetate (1 mmol) and indium trifluoride (InF₃) (1 mol%) was stirred at 80 °C. The progress of the reaction was monitored by TLC (Scheme 1). After completion of the reaction, the mixture was cooled, dissolved in acetone and filtered. The product was purified by column chromatography using benzene : ethyl acetate (9 : 1) as the eluent.

Scheme 1 Possible mechanism for IF₃ catalytic synthesis of imidazoles.

2.4.1. 2-Styryl-1H-phenanthro[9,10-d]imidazole (1). M.p. 202 °C, anal. calcd for C₂₃H₁₆N₂: C, 86.22; H, 5.03; N, 8.74. Found: C, 86.20; H, 4.9; N, 8.71. ¹H NMR (400 MHz, DMSO): δ 7.44–7.53 (m, 4H), 7.77–7.97 (m, 7H), 8.34 (d, J = 8 Hz, 1H), 8.45 (d, J = 7.2 Hz, 1H), 8.97 (t, 2H). ¹³C NMR (100 MHz, DMSO): δ 113.80, 120.05, 120.60, 122.19, 124.11, 124.31, 125.18, 126.58, 127.00, 127.74, 127.87, 127.91, 128.42, 128.72, 128.97, 129.69, 134.91, 135.09, 138.09, 143.78, 161.90. MS: m/z 320 [M⁺].

2.4.2. 1-Phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole (2). M.p. 208 °C, anal. calcd for C₂₉H₂₀N₂: C, 87.85; H, 5.08; N, 7.07. Found: C, 87.81; H, 5.02; N, 7.03. ¹H NMR (400 MHz, DMSO): δ 6.69 (d, J = 16 Hz, 1H), 7.07 (d, J = 7.6 Hz, 1H), 7.33–7.36 (m, 6H), 7.48–7.54 (m, 3H), 7.68–7.87 (m, 8H), 8.20 (d, J = 7.6 Hz, 1H), 8.87 (q, J = 8 Hz, 2H). ¹³C NMR (100 MHz, DMSO): δ 113.82, 120.17, 122.12, 123.65, 124.46, 125.21, 125.81, 126.46, 126.66, 126.84, 127.82, 127.42, 127.79, 128.29, 128.32, 128.66, 128.79, 128.92, 130.38, 130.53, 134.38, 135.69, 136.73, 136.93, 149.24. MS: m/z 396 [M⁺].

2.4.3. 2-Styryl-1-p-tolyl-1H-phenanthro[9,10-d]imidazole (3). M.p. 210 °C, anal. calcd for C₃₀H₂₂N₂: C, 87.77; H, 5.40; N, 6.82. Found: C, 87.72; H, 5.37; N, 6.79. ¹H NMR (400 MHz, DMSO): δ 6.70 (d, J = 16 Hz, 1H), 7.14 (d, J = 7.6 Hz, 2H), 7.32–7.401 (m, 4H), 7.50–7.59 (m, 6H), 7.69 (t, J = 15.2 Hz, 1H), 7.78 (t, J = 14.8 Hz, 1H), 7.86 (d, J = 7.86 Hz, 1H), 8.71 (d, J = 7.6 Hz, 1H), 8.88 (q, J = 8.4 Hz, 2H). ¹³C NMR (100 MHz, DMSO): δ 17.39, 20.99, 113.89, 120.20, 122.10, 122.21, 123.65, 124.44, 125.17, 125.77, 126.51, 126.68, 126.86, 127.24, 127.39, 127.78, 128.17, 128.30, 128.34, 128.76, 128.92, 129.18, 130.98, 134.10, 134.32, 135.74, 136.91, 139.93, 149.37. MS: m/z 410 [M⁺].

2.4.4. 1-(4-Methoxyphenyl)-2-styryl-1H-phenanthro[9,10-d]imidazole (4). M.p. 212 °C, anal. calcd for C₃₀H₂₂N₂O: C, 84.48; H, 5.20; N, 6.57; O, 3.75. Found: C, 84.46; H, 5.19; N, 6.56; O, 3.74. ¹H NMR (400 MHz, DMSO): δ 7.41–7.54 (m, 8H), 7.67–7.87 (m, 4H), 7.97 (d, J = 8.8 Hz, 2H), 8.09 (d, J = 8.4 Hz, 2H), 8.23 (d, J = 7.2 Hz, 3H), 8.34 (d, J = 8.8 Hz, 2H), 8.89 (d, J = 9.2 Hz, 1H). ¹³C NMR (100 MHz, DMSO): δ 55.49, 105.55, 113.81, 118.85, 120.05, 120.60, 122.19, 122.30, 124.11, 126.57, 126.82, 127.74, 127.86, 127.90, 128.04, 128.28, 128.75, 128.96, 129.11, 129.33, 130.53, 131.47, 134.92, 135.09, 135.90, 138.08, 138.75, 143.51, 153.59, 157.22, 161.89. MS: m/z 426 [M⁺].

2.4.5. N,N-Dimethyl-4-(2-styryl-1H-phenanthro[9,10-d]imidazol-1-yl)benzenamine (5). M.p. 215 °C, anal. calcd for C₃₁H₂₅N₃: C, 84.71; H, 5.73; N, 9.56. Found: C, 84.70; H, 5.71; N, 9.54. ¹H NMR (400 MHz, DMSO): δ 2.90 (s, 6H), 6.30 (d, J = 16 Hz, 1H), 6.67 (d, J = 7.6 Hz, 1H), 6.93–6.99 (m, 4H), 7.08–7.16 (m, 3H), 7.28–7.47 (m, 7H), 7.80 (d, J = 7.6 Hz, 1H), 8.47 (q, J = 8 Hz, 2H). ¹³C NMR (100 MHz, DMSO): δ 114.85, 121.27, 123.22, 124.70, 125.56, 126.31, 126.91, 127.56, 127.76, 127.94, 128.96, 128.52, 128.89, 129.39, 129.42, 129.66, 129.89, 129.98, 131.45, 131.59, 135.46, 136.73, 137.85, 138.42, 151.52. MS: m/z 439 [M⁺].

3. Result and discussion

3.1. XRD characterisation of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole (2)

1-Phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole is a orthorhombic crystal and crystallizes in the space group Pbca with cell dimensions a = 15.2615(9) Å, b = 16.0717(7) Å, c = 16.9595(8) Å and the ORTEP diagram is presented in Fig. 1. The imidazole unit forms dihedral angles of 50.5°(2) and 63.1°(2) with the adjacent styryl and phenyl ring [C(1)–N(1)–C(24)–C(29) = −74.5°(2), N(2)–C(15)–C(16)–C(17) = −9.50°(2)]. In the crystal, molecules are consolidated into a three-dimensional architecture by π–π stacking interactions. Optimization of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole has been performed by DFT at B3LYP/6-31G(d,p) level using Gaussian-03. All these XRD data are in good agreement with the theoretical values (Table 1). The optimized bond lengths, bond angles and dihedral angles are slightly higher than that of XRD values. These deviations can be attributed to the fact that the theoretical calculations are of the isolated molecule in the gaseous phase and the XRD results are of the molecule in the solid state.

Fig. 1 ORTEP diagram of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole.

Table 1 Selected bond lengths (Å), bond angles (°) and torsional angles (°) of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole (2)^a

Bond length	XRD	Bond angles	XRD	Torsional angles (°)	XRD (Å)
a Values in parenthesis correspond to theoretical values (Gaussian-03-DFT/B3LYP/6-31G(d,p)).
C(1)–N(1)	1.3876(1.3879)	C(1)–N(1)–C(15)	106.62(106.64)	C(1)–N(1)–C(15)–N(2)	0.23(0.25)
N(1)–C(15)	1.3809(1.3811)	N(1)–C(15)–N(2)	111.88(111.89)	N(1)–C(15)–N(2)–C(14)	−0.35(−0.36)
N(2)–C(15)	1.318(1.319)	C(15)–N(2)–C(14)	104.96(104.98)	C(1)–C(14)–N(1)–C(15)	0.36(0.38)
N(2)–C(14)	1.3662(1.366)	C(1)–C(14)–N(2)	111.93(111.95)	N(1)–C(1)–C(14)–N(2)	−0.22(−0.24)
C(1)–C(14)	1.375(1.377)	C(14)–C(1)–N(1)	104.62(104.64)	C(14)–C(1)–N(1)–C(15)	0.00(0.02)
C(1)–C(2)	1.430(1.432)	C(1)–N(1)–C(24)	128.12(128.14)	C(13)–C(14)–C(1)–C(2)	−1.3(−1.5)
C(14)–C(13)	1.432(1.433)	N(1)–C(24)–C(25)	119.82(119.83)	C(12)–C(13)–C(14)–N(2)	−1.5(−1.6)
C(15)–C(16)	1.443(1.445)	N(1)–C(24)–C(29)	119.46(119.47)	C(8)–C(13)–C(14)–N(2)	178.32(178.34)
C(24)–N(1)	1.4334(1.434)	C(29)–C(24)–C(25)	120.72(120.76)	C(8)–C(13)–C(14)–C(1)	−1.0(−1.2)
C(24)–C(29)	1.370(1.371)	C(14)–C(1)–C(2)	123.37(123.39)	C(12)–C(13)–C(14)–C(1)	179.11(179.13)
C(24)–C(25)	1.365(1.367)	C(1)–C(14)–C(13)	121.64(121.66)	C(3)–C(2)–C(1)–N(1)	1.8(1.9)
C(16)–C(17)	1.318(1.319)	C(13)–C(14)–N(2)	126.43(126.45)	C(2)–C(1)–N(1)–C(24)	−179.48(−179.50)
C(18)–C(17)	1.458(1.459)	N(2)–C(15)–C(16)	125.48(125.49)	C(1)–N(1)–C(24)–C(25)	−1.2(−1.4)
C(18)–C(23)	1.388(1.389)	N(1)–C(15)–C(16)	122.64(122.66)	C(1)–N(1)–C(24)–C(29)	105.30(105.32)
C(18)–C(19)	1.386(1.388)	C(17)–C(18)–C(19)	122.97(122.98)	N(1)–C(24)–C(29)–C(28)	−74.5(−74.7)
C(23)–C(22)	1.375(1.377)	C(19)–C(18)–C(23)	117.79(117.80)	N(1)–C(24)–C(25)–C(26)	178.71(178.73)
C(19)–C(20)	1.374(1.376)	C(18)–C(23)–C(22)	121.36(121.38)	N(28)–C(29)–C(24)–C(25)	−179.79(−179.81)
		C(18)–C(19)–C(20)	120.76(120.79)	C(29)–C(24)–C(25)–C(26)	−1.1(−1.3)
		C(8)–C(13)–C(14)	117.50(117.52)	C(14)–N(2)–C(15)–C(16)	−0.1(−0.2)
		C(1)–C(2)–C(3)	125.30(125.05)	N(2)–C(15)–C(16)–C(17)	179.58(179.60)
		C(1)–C(2)–C(7)	115.84(115.85)	C(15)–C(16)–C(17)–C(18)	−9.5(−9.7)
				C(16)–C(17)–C(18)–C(23)	−179.73(−179.74)
				C(24)–N(1)–C(15)–C(16)	166.04(166.06)
				N(1)–C(15)–C(16)–C(17)	0.4(06)
				C(17)–C(18)–C(23)–C(22)	170.46(170.47)
				C(17)–C(18)–C(19)–C(20)	−178.16(−178.17)
				C(18)–C(19)–C(20)–C(21)	178.69(178.70)
				C(22)–C(23)–C(18)–C(19)	0.0(0.2)
				C(20)–C(19)–C(18)–C(23)	1.7(1.9)

---

3.2. Absorption and luminescence

Room temperature absorption and luminescence maxima of styryl phenanthrimidazoles (1–5) are shown in Fig. 2 and 3, respectively. The spectra show a superposition of the bands corresponding to the donor and acceptor subunits which seem to be only slightly perturbed by their interactions. The three absorption bands in hexane around 26 860, 28 397 and 35 895 cm⁻¹ are assigned to the ¹(π–π*) states that correspond in Platt's notation to the ¹L_b, ¹L_a and ¹B_a excited states. The low and high energy transitions, ¹L_b ← S₀, ¹L_a ← S₀ and ¹B_a ← S₀, respectively, with a relatively high probability can be clearly observed in the absorption spectra of all the compounds studied.^14–16 A detailed inspection of the low-energy absorption region of the D–A imidazole derivatives containing styryl as an electron acceptor clearly indicates the presence of additional charge transfer singlet states. A long wave shoulder attributed to the ¹CT ← S₀ transition is also observed. The magnitudes of the shifts suggest that the ground state of the molecule is polar. The absorption data have been analyzed using the solvent comparison method proposed by Kamlet et al.¹² The equation obtained from this approach for compound 3 (Table 2) is:

E (cm⁻¹) = 37 088 − 10 793π* − 19 844α + 12 388β (5)

where the ‘π*’ denotes the dielectric effects of the solvents, ‘α’ and ‘β’ are hydrogen bond donor ability and hydrogen bond acceptor ability of the solvents, respectively. The regression values obtained for absorption and fluorescence are around 0.97 and 0.99, respectively. Negative values of ‘c’ and ‘a’ for compounds 1–4 indicate that these two parameters contribute to the stabilisation of the ground state. Negative values of ‘a’ indicate that the stabilisation is also due to hydrogen-bond donor ability of solvents. The lone pair of electrons on the tertiary nitrogen atom of the imidazole moiety are available for donation to the hydrogen bonding solvents. For compound 5 (Table 2) the equation obtained is:

E (cm⁻¹) = 31 012 − 998π* + 1256α − 11 210β (6)

Fig. 2 Room temperature absorption spectra of 1–5.

Fig. 3 Room temperature emission spectra of 1–5.

Table 2 Adjusted coefficients ((ν_x)₀, c_a, c_b and c_c) for the multilinear regression analysis of the absorption (ν_abs) and fluorescence (ν_emi) wavenumbers and stokes shift (Δν_ss) of 1–5 with solvent polarity/polarizability, and acid and base capacity using the Kamlet (π*, α and β) scales

Compd	νabs	π*	α	β
1	(3.69 ± 0.59) × 10^4	(11.01 ± 3.20) × 10^3	−(20.42 ± 3.10) × 10^3	(12.94 ± 2.51) × 10^3
2	(3.70 ± 0.05) × 10^4	(11.42 ± 2.48) × 10^3	−(22.19 ± 2.48) × 10^3	(14.46 ± 2.27) × 10^3
3	(3.70 ± 0.05) × 10^4	(10.79 ± 1.02) × 10^3	−(19.84 ± 3.22) × 10^3	(12.38 ± 0.25) × 10^3
4	(3.81 ± 0.06) × 10^4	(12.14 ± 4.82) × 10^3	−(21.04 ± 2.98) × 10^3	(12.92 ± 2.52) × 10^3
5	(3.10 ± 0.05) × 10^4	(0.99 ± 0.02) × 10^3	−(1.25 ± 0.72) × 10^3	(11.21 ± 2.54) × 10^3

---

	νemi	π*	α	β
1	(2.89 ± 0.11) × 10^4	−(18.55 ± 4.20) × 10^3	(27.26 ± 3.38) × 10^3	−(14.66 ± 4.22) × 10^3
2	(2.90 ± 0.10) × 10^4	−(18.42 ± 3.68) × 10^3	(27.66 ± 4.74) × 10^3	−(15.26 ± 4.93) × 10^3
3	(2.91 ± 0.11) × 10^4	−(18.13 ± 1.02) × 10^3	(25.70 ± 3.29) × 10^3	−(13.29 ± 4.71) × 10^3
4	(2.72 ± 0.12) × 10^4	−(18.23 ± 3.21) × 10^3	(26.82 ± 4.21) × 10^3	−(14.98 ± 3.69) × 10^3
5	(2.76 ± 0.14) × 10^4	−(17.96 ± 1.21) × 10^3	(24.20 ± 4.10) × 10^3	−(12.16 ± 2.81) × 10^3

---

The negative values of ‘c’ and ‘b’ indicate stabilization, whereas a positive value of ‘a’ indicates destabilization. The positive value of ‘a’ is consistent with the fact that donation of the lone pair of electrons of the nitrogen atom of the –N(CH₃)₂ group to hydrogen bonding solvent hinders the participation of the lone pair with the π-cloud of the ring and thus destabilizes the system.

The emission spectra of 1–5 show a red-shift from hexane to water. The emission spectra of phenanthrimidazole derivatives correspond to ³(n–π*)¹⁷ state localized in the acceptor styryl unit. The effect of polarity of the medium on the fluorescence maximum is more intense than that on the absorption maximum. This observation suggests that the emitting state is more polar than the ground state.^18,19 The emission yields are more prominent in polar solvents compared to those of non-polar solvents. The linear variation of the Stokes shift with E_T (30) (with solvent polarity parameter) has been shown in Fig. 4a, where a double linear correlation is obtained.²⁰ Polar protic solvents fall on a separate line indicating that the mode of solvation of the emitting state is different from that in the other polar aprotic solvents. For polar protic solvents, a gradual increment of the Stokes shift is due to intermolecular hydrogen bonding interactions. The studied compounds exhibit an overall increase of Stokes shift from non-polar to polar aprotic solvents mainly due to the combined effect of increasing the polarity of the medium and intramolecular charge transfer (CT) state.

Fig. 4 (a) Variation of Stokes shift as a function of E_T (30) values; (b) emission spectra of 3 in methanol as a function of water percentage composition. Curves (a) to (g) correspond to 0%, 20%, 40%, 50%, 60%, 80% and 100% water (v/v).

The influence of specific solvent–fluorophore interactions on emission of 1–5 in methanol–water mixtures of different compositions has been monitored and the spectra are illustrated in Fig. 4b. The addition of water to the methanol–water mixture remarkably enhances the emission with a red shift and this may be due to a combined effect of hydrogen bonding and polarity of the medium. Mixing of the two closely lying lowest singlet states (n, π* and π, π*) of phenanthrimidazole derivatives favours the intersystem crossing.²¹ As the polarity of the medium is increased, the intermolecular hydrogen bonding interaction in the excited state stabilized the π, π* state and enhanced the energy between the two states, which diminishes the mixing between the states (Scheme 2). As a result intersystem crossing from S₁ to T₁ decreases and an enhancement of fluorescence is observed.

Scheme 2 Schematic demonstration of modulation of the close-lying lowest singlet n, π* and π, π* states with solvent polarity in the excited state.

The behaviour of phenanthrimidazole derivatives towards different solvent polarities may be interpreted in terms of the difference in the ground and excited state dipole moments. The emission shift may be attributed to the interaction between the dipole moment of the solute and the polarizability of the solvent. The extent of charge separation on electronic excitation have been determined by measuring the change in the dipole moment (Δμ = μ_e − μ_g) utilizing the spectral shift between the absorption and emission maxima as a function of solvent polarity. According to the Lippert–Mataga eqn (7):²²

[small nu, Greek, macron]_SS = [2(μ_e − μ_g)/hca³]Δf + ν^o_SS (7)

where [small nu, Greek, macron]_SS is the Stokes shift, the superscript “o” indicates the absence of solvent, μ_g and μ_e are dipole moments in the ground state and excited state, respectively, h is Planck's constant, c is the velocity of light and a is the Onsager cavity radius. The orientation polarizability Δf is defined as

Δf = [(ε − 1)/(2ε + 1) − [(n² − 1)/(2n² + 1)]

where ε and n are solvent dielectric constant and refractive index, respectively. The Lippert–Mataga plot is linear for the non-polar and polar/aprotic solvents as shown in Fig. 5a. The geometrical optimization was done by DFT method using Gaussian-03 (ref. 13) to calculate μ_g. Using these μ_g values [5.68 D (1), 7.03 D (2), 6.81 D (3), 6.25 D (4) and 5.18 D (5)] and the slope of Lippert–Mataga plot, the value of μ_e calculated is in the range 14.0–22.0 D for the studied imidazoles.

Fig. 5 (a) Lippert–Mataga plot of Stokes shift against solvent polarity function (Δf) of all solvents; (b) fluorescence maxima in terms of wavenumber (cm⁻¹) and E_T (30) values.

3.3. Solvatochromic comparison method (SCM)

SCM has also been applied to analyze the fluorescence data, which results in the following equation for compound 3 (Table 2):

E (cm⁻¹) = 29 169 − 18 136π* + 25 704α − 13 299β (8)

This equation shows that the parameters, ‘π*’ and ‘β’ contribute to the stabilization of the excited state for compounds 1–4. For compound 5 the equation obtained is:

E (cm⁻¹) = 27 691 − 17 961π* − 242 022α − 12 165β (9)

All three parameters ‘π*’, ‘α’ and ‘β’ contribute to the stabilisation of the excited state. Compared to the ground state, the negative value of ‘b’ indicates that the stabilization is due to hydrogen-bond donor ability of solvents. This favours that, although the lone pair of electrons on the N atom of –N(CH₃)₂ group of compound 5 is not available due to its involvement in the formation of the charge transfer state, but that on the tertiary N atom of the imidazole moiety is available for donation to the hydrogen bonding solvents. This, in turn, favours the stabilisation of the excited state. Eqn (9) shows that greater contribution to the stability of the excited state is also due to dipole–dipole interactions like ground state stabilization. But this contribution is greater in the excited state than that in the ground state, which also supports that the excited state dipole moment is greater than the ground state dipole moment. This explanation supports the high Stokes shift values in polar solvents.²³ The calculated ratios of α over π* [1.84 (ν_abs) & 1.42 (ν_emi) (1), 1.85 (ν_abs) & 1.47 (ν_emi) (2), 1.94 (ν_abs) & 1.50 (ν_emi) (3), 1.73 (ν_abs), 1.35 (ν_emi) (4) and 1.26 (ν_abs), 1.35 (ν_emi) (5)] reveal that interactions between phenanthrimidazole derivatives and solvents with acidity property (α) predominate in the excited state. A good linear variation is also obtained between the fluorescence maxima and E_T (30) values²⁴ as shown by Fig. 5b.

3.4. Free energy change of solvation and reorganisation energies

The free energy change of solvation and reorganization energies in various solvents have been estimated (Table 3). According to Marcus,²⁵ E(A) = ΔG_solv + λ₁ and E(F) = ΔG_solv − λ₀, where E(A) and E(F) are absorption and fluorescence band maxima in cm⁻¹, respectively, ΔG_solv is the difference in free energy of the ground and excited states in a given solvent and λ represents the reorganization energy. The free energy change of solvation and reorganization energies in various solvents have been estimated. Under the condition that λ₀ ≈ λ₁ ≈ λ, we get, E(A) + E(F) = 2ΔG_solv; E(A) − E(F) = 2λ. The ΔG_solv is maximum for hexane since it is completely non-polar and also α and β values of hexane are zero and ΔG_solv is minimum in water. The difference between these values (water and hexane) should give the free energy change required for hydrogen bond formation. The plot of Δ[(ΔG_solv) = (ΔG_hex − ΔG_water)] versus E_T (30) has been depicted in Fig. 6a. The difference in free energy of solvation in hexane and different hydrogen bonding solvents (i.e. ΔG_solv) follow the order of the hydrogen bond energy.²⁶ In aprotic solvents the values are small and interaction of phenanthrimidazole derivatives with those solvents is due to dipolar interactions in the excited state. The reorganization energy values have also been determined in different solvents. The definite values of reorganization energy confirmed the interaction between low frequency motions such as reorientation of solvent cell with low and medium frequency nuclear motion of the solute.

Table 3 Absorption (λ_abs, nm), emission (λ_f, nm), Stokes shift (ν_ss, cm⁻¹), fluorescence quantum yield (Φ_f), lifetime (τ, ns), free energy change of solvation (Δ(ΔG), kcal mol⁻¹), reorganization energy (λ, eV), radiative rate constant (k_r, 10⁸ s⁻¹), nonradiative rate constant (k_nr, 10⁸ s⁻¹) and log(k_r/k_nr) of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole (2) in various solvents

Solvents	λabs	λf	[small nu, Greek, macron]ss	Δ(ΔG)	λ	Φ	τ	kr	knr	log(kr/knr)
Hexane	277	311	3873.37	0.00	5.54	0.76	1.3	0.58	0.18	0.50
Cyclohexane	276	315	4443.29	0.50	6.35	0.75	1.3	0.58	0.19	0.48
Dioxane	274	320	5191.07	0.74	7.42	0.75	1.3	0.57	0.19	0.48
Carbon tetrachloride	273	330	6253.28	1.87	8.94	0.74	1.3	0.57	0.20	0.45
Benzene	272	359	8887.40	5.12	12.70	0.7	1.3	0.54	0.24	0.37
Ether	270	366	9734.69	5.41	13.91	0.7	1.4	0.5	0.21	0.37
Chloroform	266	371	10 660.3	5.24	15.24	0.7	1.4	0.5	0.21	0.37
Ethyl acetate	264	377	11 451.5	5.40	16.37	0.75	1.4	0.54	0.18	0.48
THF	261	385	12 291.2	5.62	17.57	0.74	1.4	0.53	0.19	0.45
Dichloromethane	259	389	12 925.6	5.56	18.47	0.71	1.4	0.51	0.21	0.39
Propanol	257	397	13 625.6	5.96	19.47	0.73	1.6	0.45	0.17	0.43
Butanol	255	401	14 262.7	5.96	20.39	0.74	1.6	0.46	0.16	0.45
2-Propanol	255	415	15 146.0	7.01	21.65	0.75	1.6	0.47	0.16	0.48
Acetone	254	427	16 021.9	7.74	22.90	0.63	1.5	0.42	0.25	0.23
Ethanol	252	431	16 421.7	7.71	23.47	0.65	1.6	0.41	0.22	0.27
Methanol	251	432	16 634.5	7.61	23.78	0.63	1.6	0.39	0.23	0.23
Acetonitrile	251	435	16 901.9	7.63	24.16	0.62	1.6	0.39	0.24	0.21
Ethylene glycol	249	436	17 202.5	7.40	24.59	0.59	1.6	0.37	0.26	0.16
DMSO	248	439	17 468.9	7.39	24.97	0.61	1.5	0.41	0.26	0.19
Water	247	442	17 921.4	7.24	25.61	0.52	1.4	0.37	0.34	0.04
Glycerol	245	445	18 287.5	7.09	26.14	0.61	1.6	0.38	0.24	0.19

---

Fig. 6 (a) Variation of difference in free energy with E_T (30); (b) variation of quantum yield with solvent polarity parameter E_T (30).

3.5. Fluorescence lifetime and quantum yield

The time correlated single photon counting (TCSPC) results fit to the biexponential decay,

f(t) = α₁ exp(−t/τ₁) + α₂ exp(−t/τ₂) (10)

where α₁ and τ₁ are respectively, the pre-exponential factor and lifetime of the various excited states involved. This model is based on the assumption that one, two or three fluorescent substances are present in the solution. If N_i molecules are excited at time zero, the fluorescence quantum yield of the i^th component α_i is proportional to the ratio α_iτ_i/N_i and the α_i factors are related to the absorbance of the various substances at the excitation wavelength. The fluorescence decay curves of phenanthroimidazoles were recorded in ethanol (Fig. 6b). Laser excitation was set at 270 nm and the fluorescence signal was measured at the emission wavelength of the individual compound. DAS6 software was used for the fit and the χ² values are always less than 1.2. The absolute PL quantum yields were measured by comparing fluorescence intensities (integrated areas) of a standard sample (Coumarin 46) and the unknown sample using the formulawhere, Φ_unk is the fluorescence quantum yield of the sample, Φ_std is the fluorescence quantum yield of the standard; I_unk and I_std are the integrated emission intensities of the sample and the standard, respectively. A_unk and A_std are the absorbances of the sample and the standard at the excitation wavelength, respectively. η_unk and η_std are the indexes of refraction of the sample and standard solutions.

3.6. Radiative and non-radiative rate constants

The radiative and non-radiative decay of the excited state have been obtained using the quantum yields and lifetimes. The formula employed to calculate the radiative (k_r) and non-radiative (k_nr) rate constants is k_r = Φ/τ; k_nr = (1/τ) − (Φ/τ); τ = (k_r + k_nr)⁻¹, where k_r and k_nr are the radiative and non-radiative deactivation, τ is the lifetime. A typical set of k_r and k_nr values are tabulated (Table 3). The radiative and non-radiative rate constants show that in most of the cases the radiative emission is predominant over non-radiative transitions. The variation of φ with solvent polarity parameter E_T (30) is depicted in Fig. 7a. The φ values in various solvents are sensitive towards solvent polarity. The remarkable increase of φ in polar protic medium is compared with that in polar aprotic medium. This may be due to different contributions of CT and hydrogen bonding interactions. The logarithm of (k_r/k_nr) is plotted against the solvent polarity parameter E_T (30) which is shown in Fig. 7b. Two different straight lines are obtained, one for aprotic solvents and the other for protic solvents. In both the cases, upon increasing the polarity, the logarithm ratio of radiative to nonradiative rate decreases, but a steeper slope is obtained in the case of protic solvents. It indicates that the radiative and nonradiative rates are more sensitive towards protic solvents. It may be that the hydrogen bonding interaction in polar protic environment enhances the stabilization of the excited state.²⁷

Fig. 7 (a) Lifetime decay of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole in CH₂Cl₂; (b) log(k_r/k_nr) vs. E_T (30).

3.7. Excited state dipole moments

An interesting result is provided by the blue shift of the CT absorption bands with increasing solvent polarity (Fig. 8a).²⁸ With the assumption that point dipole is at the center of the spherical cavity (Fig. 8b) and the mean solute polarizability (α) is insignificant, it follows,

hc[small nu, Greek, tilde]_abs ≈ hc[small nu, Greek, tilde]^vac_abs − 2μ_g(μ_e − μ_g)/a³_o[(ε − 1/2ε + 1) − ½(n² − 1/2n² + 1)] (12)

where μ_g and μ_e are the dipole moments of the solute in the ground and excited states, respectively, ν_abs and [small nu, Greek, tilde]^vac_abs are the spectral positions of a solvent-equilibrated absorption maximum and the value extrapolated to the gas-phase, respectively, a_o is the effective radius of the Onsager cavity,²⁹ and ε and n are the static dielectric constant and the refractive index of the solvent, respectively. In the case of the well-separated CT absorption bands, eqn (12) is used to determine the values of μ_g(μ_e − μ_g)/a³_o and [small nu, Greek, tilde]^vac_abs. In the excited state, the negative and positive ends of the electric dipole are localized nearly in the centres of the donor and acceptor fragments, respectively. Red shift of their spectral position, increase of the Stokes shift and enlargement of emission bandwidth with increasing solvent polarity in fluorescence spectra, point to the CT character of the fluorescent states and indicate that the absolute values of μ_e are much higher than those of μ_g. The excited state dipole moments μ_e can be estimated by the fluorescence solvatochromic shift method due to the fact that the excited states live sufficiently long with respect to the orientation relaxation time of the solvent.^30–33 Under the same assumptions as used for expression (12), it follows that

hc[small nu, Greek, tilde]_flu ≈ hc[small nu, Greek, tilde]^vac_flu − 2μ_e(μ_e − μ_g)/a³_o[(ε − 1/2ε + 1) − ½(n² − 1/2n² + 1)] (13)

where [small nu, Greek, tilde]_flu and [small nu, Greek, tilde]^vac_flu are the spectral positions of the solvent equilibrated fluorescence maxima and the value extrapolated to the gas-phase, respectively. The compounds studied show a satisfying linear correlation between the energy hc[small nu, Greek, tilde]_flu and the solvent polarity function in a polar environment and also in all the solvents (Fig. 8c).³⁴ The values of μ_e(μ_e − μ_g)/a³_o extracted from the data measured in polar media are somewhat larger than those resulting from the analysis of the data obtained for the whole range of the solvents. This finding can be explained only by the dependence of the electronic structure of the fluorescent states on solvation. Due to a relatively small energy gap between the lowest internal charge transfer (ICT) states and the states excited locally in the nonpolar solvents, this leads to an increase of the contribution of the (π, π*) character to the wave function of the CT states. It leads to a lowering of energy with respect to a pure CT state because of the stabilizing character of such interactions and red shift obtained in the fluorescence spectra.

Fig. 8 (a) Solvatochromic effects on the spectral position of the CT absorption maxima; (b) direction of ground and excited state dipoles; (c) CT fluorescence maxima vs. solvent polarity function.

Under the assumption that the CT fluorescence corresponds to the state reached directly upon excitation, the quantity (μ_e − μ_g)²/a³_o can be evaluated from the solvation effects on the Stokes shift,

hc([small nu, Greek, tilde]_abs − [small nu, Greek, tilde]_flu) = hc(hc[small nu, Greek, tilde]^vac_abs − hc[small nu, Greek, tilde]^vac_flu) + 2(μ_e − μ_g)²/a³_o[(ε − 1/2ε + 1) − ½(n² − 1/2n² + 1)] (14)

The compounds studied show a satisfying linear correlation between the energy hc[small nu, Greek, tilde]_abs − hc[small nu, Greek, tilde]_flu and the solvent polarity function in a polar environment and also in all the solvents studied; the values of (μ_e − μ_g)/a³_o are 1.30 eV (1), 2.05 eV (2), 1.85 eV (3), 1.49 eV (4) and 1.05 eV (5). The eqn (12)–(14) relate the measured quantities to the excited state dipole moments μ_e. Under the assumption that μ_e ≫ μ_g and with the effective spherical radius (a₀) of the molecules, 5.49 Å (1), 5.90 Å (2), 5.98 Å (3), 6.05 Å (4) and 6.13 Å (5), as estimated from the molecular dimensions of the compounds calculated by molecular mechanics, eqn (13) and (14) yield very similar values of μ_e as 15.91 D (1), 22.21 D (2), 20.68 D (3), 17.56 D (4) and 14.23 D (5) for the studied molecules. The large values of Δμ = μ_e − μ_g ≈ 10.23 D (1), 15.18 D (2), 13.87 D (3), 11.31 D (4) and 9.05 D (5) correspond to a charge separation of about 0.3 nm, 0.4 nm, 0.5 nm, 0.5 nm and 0.4 nm, which roughly agrees with the centre-to-centre distance between the donor and acceptor moieties of the compounds.

This conclusion is in agreement with a linear relationship found between the CT fluorescence energies and the differences in the redox potentials corresponding to the oxidation of the donor subunit E_oxi(D) and the reduction of the acceptor moiety E_red(A) in the D–A molecules. The correlation is shown in Fig. 9. In this correlation the values of E_oxi(D) − E_red(A) are taken from the electrochemical data obtained for the given D–A molecule in ACN containing 0.1 M TBAPF₆. These values are very similar to those expected from the electrochemical properties of the donor and acceptor alone. The small shift of E_red(A) to more negative potentials can be explained by the electron donating properties of substituted imidazole fragment bonded to the acceptor subunit. Correspondingly, the small shift of E_oxi(D) to more positive potentials arises from the electron withdrawing character of the acceptor moiety. The E_red(A) and E_oxi(D) values indicate (in agreement with the absorption spectra) that both subunits of all the D–A molecules studied interact very weakly.

Fig. 9 Correlation between the energy of the fluorescence maxima of the 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole derivatives.

These finding points to a different electronic structure of the emitting singlet state ¹CT. The spectroscopic ¹CT state can be regarded as a linear combination of the zero-order ET state (¹ET) with: (i) various locally excited ¹(π, π*) states and (ii) with the ground state (S₀):^2–4

Ψ¹_CT ≊ C_ETΦ¹_ET + C_aΦ¹_a + C_bΦ¹_b + C₀Φ¹₀ (15)

where Φ¹₀, Φ¹_ET, Φ¹_a and Φ¹_b represent the closed-shell configuration of the ground state, the zero-order wave functions of the pure ¹ET state and the ¹L_a and ¹L_b states of the donor moieties, respectively. The ¹ET state is described by a full ET from the occupied HOMO orbital of the donor to the vacant LUMO orbital of the acceptor. The values of Δμ suggest that the wave functions Ψ¹_CT of the ¹CT states are in the order 2 > 3 > 4 > 1 > 5. This finding is in agreement with the value of hc[small nu, Greek, tilde]^vac_flu is of the order 2 > 3 > 4 > 1 > 5. It is also in agreement with the magnitude of the electronic coupling elements V^D₁ estimated from eqn (3). Therefore one can expect that the contribution of the ¹(π, π*) character to the wave function Ψ¹_CT should decrease in the order, 2 > 3 > 4 > 1 > 5.

3.8. Electronic coupling elements

The electronic coupling elements between the lowest excited ¹CT state and the ground state (V₀) or the locally excited state lying most closely in energy (V₁) can be estimated from the CT absorption and fluorescence investigations.^35,36 Applying a simple kinetic model of an irreversible excited CT state formation (with 100% efficiency), the radiative and non-radiative rate constants, k_r and k_nr, and the resulting electronic transition dipole moments (M_flu) are given by:^37,38The electronic transition dipole moments (M_flu) can be expressed by the following relation^35,36

The first term in the above equation corresponds to the Mulliken two-state model^39,40 (i.e., the interactions between the solvent equilibrated fluorescent CT state and the Franck–Condon ground state), the second one represents the Murrell ‘borrowing’ mechanism^4,35,36 (i.e., the contributions from the locally excited states i of energy E_i). The electronic transition dipole moments correspond to radiative transitions ¹LE_i → S₀ (states i in the studied systems to the ¹L_a and ¹L_b states of the donor moieties). Neglecting the second term in eqn (17) and assuming that the C_ET value is close to unity, the V₀, 0.25 eV (1), 0.20 eV (2), 0.27 eV (3), 0.27 eV (4) and 0.18 eV (5) values calculated with Δμ = μ_e − μ_g ≈ 10.23 D (1), 15.18 D (2), 13.87 D (3), 11.31 D (4) and 9.05 D (5). These quantities are very close to the values calculated from eqn (1).

The inner reorganization energy (λ_i) corresponds to the high-frequency motions associated with the changes in the solute bond lengths and angles. The values of the low-frequency reorganization energy (λ₀) depend on the solvent polarity (Fig. 10), as expected. Within the continuum dielectric model of solvation, an analysis of the solvatochromic effects on λ₀ is possible according to the following expression:

λ₀ ≈ δλ₀ + λ₀ ≈ δλ₀ + (μ_e − μ_g)²/a³_o[(ε − 1/2ε + 1) − ½(n² − 1/2n² + 1)] (18)

Fig. 10 Dependence of the reorganization energy λ₀ related to the low-frequency solvent and solute motions accompanying the excited-state electron transfer on the solvent polarity function.

This relation is suitable for the direct comparison with the results of the investigations of the solvatochromic effects on the spectral position of the CT fluorescence maxima (Fig. 3). The values of (μ_e − μ_g)²/a³_o (1.30 eV (1), 2.05 eV (2), 1.85 eV (3), 1.49 eV (4) and 1.05 eV (5)) obtained from the mean slope of the plots corresponding to eqn (18) agree with those calculated by eqn (14). This finding supports again the hypothesis that the wave functions Ψ¹_CT of the ¹CT states are of the order 2 > 3 > 4 > 1 > 5. The ΔG_CT values are related to the energy corresponding to the sum of the CT absorption and fluorescence maxima −ΔG_CT = ½hc(ν_abs^−CT + ν_flu^−CT). This relation allows us to estimate the spectral position of CT absorption maxima. The evaluated values support the significant contribution of the ¹CT ← S₀ transition to these bands. Moreover the estimated change of the spectral position of the CT absorption agrees well with the observed solvatochromic effects of the lowest absorption band. It indicates that the similar electronic structures of both corresponding states i.e., the state initially reached upon excitation and the equilibrated fluorescent one. The energetical relations are visualized in Scheme 3 for a shallow ground state (S₀) potential and a steep ¹CT potential. This is a typical situation for electron transfer systems because the force constant for the solute–inner solvation shell interaction determining the steepness of the potential with respect to the solvent coordinate is larger in the polar ¹CT state than in the ground state.⁴¹

Scheme 3 Gibbs energy change of fluorescence back charge transfer process.

The electrochemical properties of the imidazoles (1–5) have been examined by cyclic voltammetry and the redox potentials have been measured from the plot of potential versus current. The energies of the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) have been calculated using the relation, E_HOMO = 4.8 + E^oxi_½; E_LUMO = E_HOMO − 1239/λ_onset. The 3D plot of HOMO–LUMO orbital picture is shown in (Fig. 11). The LUMO energies have been deduced from the HOMO energies and the lowest-energy absorption edges of the UV-vis absorption spectra. The calculated energy gap (E_g = E_HOMO − E_LUMO) of 1–5 are 1.54 eV, 2.84 eV, 3.99 eV, 1.65 eV and 0.82 eV. Therefore, the HOMO stability and the emission energy gap are controlled by the nature and substituent present in the imidazole moiety.

Fig. 11 Computed HOMO–LUMO orbitals contour map.

3.9. Effect of [H⁺] on absorption and luminescence

The absorption spectrum of N,N-dimethyl-4-(2-styryl-1H-phenanthro[9,10-d]imidazol-1-yl)benzenamine recorded at pH 0.25 exhibits two bands, the peak at 289 nm is red shifted and the other at 240 nm is blue shifted with respect to that of the neutral molecule (λ^ab_max = 255 nm). The fluorescence spectra at different pH values are shown in Fig. 12. The fluorescence spectra obtained by exciting at the respective band maxima of the above mentioned two absorption bands are also different. Similar to the absorption spectrum, the peak at 492 nm is red shifted and the other at 391 nm is blue shifted with respect to that of the neutral molecule (λ^fl_max = 454 nm). The fluorescence excitation spectrum corresponding to the blue-shifted fluorescence band matches exactly with the short wavelength absorption band while that for the red-shifted fluorescence band resembles the long wavelength absorption band. This suggests the formation of two different kinds of monocation species I and II (Scheme 4). The red-shifted absorption and emission bands correspond to the monocation I, formed by the protonation of the heterocyclic nitrogen of the imidazole ring and the blue-shifted absorption and emission bands are associated with the monocation II formed as a result of protonation at the dimethylamino nitrogen of the styryl ring. Since the lowest energy transition of the molecule is π → π* in nature, the protonation at the heterocyclic nitrogen atom will shift the absorption and fluorescence bands to the red, whereas the same at the nitrogen atom of the N(CH₃)₂ group will shift the band to the blue with respect to that of the neutral molecule. The large red shift for the monocation I can be explained by the resonance interaction of the N(CH₃)₂ group with the styryl part of the molecule that leads to structure I′ responsible for the stabilization of the species. The formation of both types of monocations I and II indicates the involvement of the lone pair of electrons of the N(CH₃)₂ group in the resonance interaction, thus reducing the charge density on the nitrogen atom of the N(CH₃)₂ group and enhancing the same on the heterocyclic nitrogen atom.¹⁰

Fig. 12 Fluorescence spectra of 1-phenyl-2-styryl-1H-phenanthro[9,10-d]imidazole in aqueous solution of different pH; (a) 0.25, (b) 0.80, (c) 1.70, (d) 2.28, (e) 2.70, (f) 4.01, (g) 5.60, (h) 7.00.

Scheme 4 Effect of protonation.

As the pH of the medium increases (pH 1.70), the intensity of the fluorescence band corresponding to species I decreases whereas that of the species II increases. Decreasing the concentration of hydrogen ions means the possibility of formation of monocation I is reduced. Above pH 1.70, the intensity of the monocation II starts decreasing and at the expense of this, a new band starts appearing at the wavelength 408 nm lying between the above mentioned two bands; (492 nm and 391 nm), which is almost the same as λ^fl_max of the TICT band in water. The appearance of TICT emission band at pH 2.28 is shown in Fig. 12, the intensity of which becomes maximum at pH ∼ 4.01. All these observations support the equilibrium monocation II neutral molecule III in the pH range of 1.70–4.01. The pK* (excited state protonation constant) of this equilibrium, calculated by fluorimetric titration method^42–44 is (monocation II neutral species III) 2.62. The monocation II shows its maximum accumulation at a pH of 1.70. With increasing pH, the equilibrium concentration of the neutral species increases in the equilibrium between monocation II neutral species III. This results in an increase in fluorescence intensity of species III. It can be seen in Fig. 12 that when the intensity of species II at pH 4.01 is very low, the intensity of species III becomes close to the intensity of species II at pH 1.70. The pK^*_a of this equilibrium is found to be 2.62, whereas the ground state protonation constant (pK_a) of the same equilibrium is 4.92. The higher value of pK_a compared to pK^*_a can be explained by the fact that the basicity of the N atom of N(CH₃)₂ decreases due to a greater extent of charge transfer from the same N atom on excitation, which results in the lowering of the equilibrium concentration of species II in the excited state.^42,45 The appearance of a new band at pH 2.28 suggests that this band is due to the TICT emission, which is the only predominant emission that can exist in water. This also confirms that monocation II is formed as a result of protonation of the N atom of N(CH₃)₂ group, for which a TICT band is not expected due to the engagement of lone pair of electrons with the H⁺ ion. Therefore no TICT band appears at pH 0.25 for monocation I, although it involves greater charge transfer from N atom of N(CH₃)₂ group to the protonated phenanthrimidazole ring. The protonation of N atom in this ring facilitates the conjugation of the N(CH₃)₂ group with the π-system, which in turn leads to the appearance of a double bond character of the phenyl carbon–nitrogen bond and thus reduces the rotational relaxation of the N(CH₃)₂ group in the excited singlet state.

The TICT fluorescence intensity reaches its maximum value at pH ∼ 4.01 after which the fluorescence intensity starts decreasing and becomes constant at pH ∼ 7.8. Therefore, from this observation it can be concluded that the H-bonding effect of water becomes important only when the pH of the solution is greater than 4.01. Increasing the pH of the solution above 4.01 drives more and more H₃O⁺ ions to lose a proton to form H₂O. As a consequence, the efficiency of H-bonding interactions increases, thus decreasing the TICT fluorescence intensity, which attains its minimum constant value at around pH 7.8. This observation indicates the sensing efficiency towards H-bonding interaction.⁴⁶ pK_a and pK^*_a values of monocation II neutral species equilibrium have been calculated and compared. Moreover, the fluorescence quenching of neutral molecule has been observed above pH 4.01 and the possible reason for this has been explained above.

3.10. Quantum chemical calculations

To rationalize the experimental findings and also to locate the TICT state, DFT calculations^47–51 have been carried out to determine the optimized geometries of the different conformations to explain their properties in the ground and excited states. The donor moiety is flexibly linked to the acceptor moiety by the C–N bond. In the Franck–Condon or locally excited (LE) state there is an increase in planarity of the molecule due to the mutual conjugation of nitrogen lone pair orbital and the acceptor π sub system that may have a fractional internal charge transfer. Rotational motion around the C–N bond is described by the torsion angle φ, which is equal to zero when the amino lone pair orbital is perpendicular to the molecular plane of the acceptor sub system. When the two moieties are orbitally decoupled due to the twisting of the dimethylamino group with respect to the styryl ring of acceptor part (φ = 90°), then full intramolecular charge transfer (ICT) takes place in the excited state. In the ground state the –N(CH₃)₂ group shows a pyramidal shape because of the sp³ hybridization of the nitrogen atom and changes to the trigonal plane in the excited state due to a change in hybridization of the nitrogen atom to sp².⁴⁹ The calculation suggests the requirement of the twisting of the donor group N(CH₃)₂ perpendicular to the plane of the rest of the molecule for the creation of stable TICT state in polar medium. In this process the overlapping orbitals in the planar geometry get decoupled and as a result of this complete charge transfer takes place.

3.11. TICT state vs. DFT calculations

The TICT state can be well examined through energy evolutions of several excited singlet states as a function of the twist angle (φ) between the donor and the acceptor parts of the molecule. Such an examination allows ascertaining the dual fluorescence and can be interpreted in terms of TICT state. This type of analysis has been done for several molecules using a wide range of experimental and theoretical methods.^49–56 DFT calculations have been used to find out the excited singlet state that is responsible for the increasing TICT emission. On rotating the N(CH₃)₂ moiety in the molecule, keeping the rest of the molecule in the trans planar form, it can be observed that at an angle of 90°, the S₃ state of the molecule is acquiring a very high dipole moment (14.23 D). The change in the dipole moments of the different singlet states with the variation of the torsion angle is illustrated by Fig. 13. The overall polarity of the synthesized imidazole derivatives was small when their dipole moments aligned in a parallel fashion (Fig. 14). When the electric field is removed, the parallel alignment of the molecular dipole moments begins to deteriorate and eventually the imidazole derivative loses its NLO activity. The ultimate goal in the design of polar materials is to prepare compounds which have their molecular dipole moments aligned in the same direction.⁵⁷

Fig. 13 Change in the dipole moments of the different singlet states with the variation of the torsion angle.

Fig. 14 Orientation of dipole moments.

The remarkable lowering of energy of the S₃ state has been observed at a twist angle, φ = 90°, which is consistent with the increase in dipole moment of the S₃ state at the same twist angle. The effect of solvation by each polar solvent results in the lowering of energy of the S₃^TICT state, even lower than S₁^LE as well as S₁^TICT states, which makes S₃^TICT state responsible for anomalous fluorescence at a critical polarity. These calculations suggest the presence of TICT state and highly Stokes shifted fluorescence in polar solvents and also support the experimental observations.^58,59

Further confirmation of the TICT phenomenon comes from the frontier molecular orbital (FMO) pictures. The nature of the molecular orbitals involved is illustrated by Fig. 11. In the planar geometry, the HOMO and LUMO have considerable delocalization over the whole π-system. Both the orbitals are primarily located on the donor and acceptor parts of the molecule. The S₀ → S₃^LE transition thus occurs only with a fraction of a full electron transfer from the nitrogen atom of N(CH₃)₂ fragment to the remainder of the molecule, which results in a calculated dipole moment at the S₃^LE state. At the molecular twist of 90°, π orbitals of the N(CH₃)₂ group are completely decoupled from the remaining π orbitals, so that the HOMO → LUMO excitation entails a full electron transfer from the donor to the acceptor. This leads to the formation of the S₃^TICT state with a very high dipole moment value of 14.23 D, which on stabilization due to solvation with a solvent of high polarity, becomes responsible for the highly Stokes shifted fluorescence. In the gas phase, the energies of TICT states are greater than the respective planar S₁ and S₂ states, whereas the energy of S₃^TICT state is lower than the S₃ planar state. The dipole moment of only the S₃^TICT state among the first four singlet states is greater than the respective planar state. This dipole moment is large enough relative to the dipole moments of the remaining states to locate the TICT state by the dielectric continuum estimation in polar solvents.⁶⁰

4. Conclusion

Solvent-induced transformation of the electronic structure of the ¹CT states of selected D–A derivatives of phenanthrimidazole derivatives containing styryl as an electron acceptor has been analysed by combining the Mulliken–Murrell theory of the CT complexes and the Marcus theory of the radiative charge recombination ¹CT → S₀. Analysis of the electronic coupling elements show that the differences in the photophysical properties of the studied molecules can be interpreted in terms of the different electronic interactions between the CT state and the ground state (V₀) and/or ¹(π, π*) excited states most probably localized in the donor moiety V^D₁. This approach can be used to explain the solvatochromic effects on the spectral position of CT fluorescence spectra as well as on transition dipole moment values. The conformation of the investigated D–A systems in the fluorescent CT states seems to be similar to that in the ground state. The decrease of TICT fluorescence of N,N-dimethyl-4-(2-styryl-1H-phenanthro[9,10-d]imidazol-1-yl)benzenamine with increasing pH above 4.01 suggests that water molecules become efficient for hydrogen bonding interactions only above this pH. This prompts us to suggest that this molecule may be used as a sensor to sense the hydrogen-bonding efficiency of the medium.

Acknowledgements

One of the authors, Prof. J. Jayabharathi is thankful to DST [no. SR/S1/IC-73/2010], DRDO (NRB-213/MAT/10-11), UGC (F. no. 36-21/2008 (SR)) and CSIR (no. 3732/NS-EMRII) for providing funds to this research study.

References

1. J. Herbich, A. Kapturkiewicz and J. Nowacki, Phosphorescent intramolecular charge-transfer states, Chem. Phys. Lett., 1996, 262, 633–642.
2. R. S. Mulliken, J. Am. Chem. Soc., 1952, 74, 811–824.
3. R. S. Mulliken and W. B. Person, Molecular Complexes: A Lecture and Reprint Volume, Wiley, New York, 1969.
4. J. N. Murrell, J. Am. Chem. Soc., 1959, 81, 5037–5043.
5. M. Bixon, J. Jortner and J. Verhoeven, J. Am. Chem. Soc., 1994, 116, 7349–7355.
6. I. R. Gould, R. H. Young, L. J. Mueller, A. C. Albrecht and S. Farid, J. Am. Chem. Soc., 1994, 116, 8188–8199.
7. J. Herbich and A. Kapturkiewicz, Electronic structure and molecular conformation in the excited charge transfer singlet states of p-acridyl and other aryl derivatives of aromatic amines, J. Am. Chem. Soc., 1998, 120, 1014–1029.
8. R. R. Dogonadze, A. M. Kuznetsov and T. A. Marsagishvili, The present state of the theory of charge transfer in condensed phase, Electrochim. Acta, 1980, 25, 1–28.
9. N. Mataga, Y. Kaifu and M. Koizumi, The Solvent Effect on Fluorescence Spectrum. Change of Solute-Solvent Interaction During the Lifetime of Excited Solute Molecule, Bull. Chem. Soc. Jpn., 1955, 28, 690.
10. J. Dey, E. L. Roberts and I. M. Warner, J. Phys. Chem. A, 1998, 102, 301–305.
11. (a) J. Jayabharathi, V. Thanikachalam, K. Jayamoorthy and N. Srinivasan, Synthesis, spectral studies and solvatochromism of some novel imidazole derivatives – ESIPT process, Spectrochim. Acta, Part A, 2013, 105, 223–228; (b) J. Jayabharathi, V. Thanikachalam and R. Sathishkumar, Characterization, physiochemical and computational studies of the newly synthesized novel imidazole derivative, Spectrochim. Acta, Part A, 2012, 97, 582–588; (c) J. Jayabharathi, V. Thanikachalam and K. Jayamoorthy, Computational studies of 1,2-disubstituted benzimidazole derivatives, Spectrochim. Acta, Part A, 2012, 97, 131–136; (d) J. Jayabharathi, V. Thanikachalam, N. Srinivasan and K. Jayamoorthy, Synthesis and physico-chemical studies of cyclometalated heteroleptic iridium(III) complexes, Spectrochim. Acta, Part A, 2012, 93, 240–244; (e) J. Jayabharathi, V. Thanikachalam and M. Venkatesh Perumal, Photophysical studies of fused phenanthrimidazole derivatives as versatile π-conjugated systems for potential NLO applications, Spectrochim. Acta, Part A, 2012, 92, 113–121; (f) J. Jayabharathi, V. Thanikachalam and K. Jayamoorthy, Physicochemical studies of chemosensor imidazole derivatives: DFT based ESIPT process, Spectrochim. Acta, Part A, 2012, 89, 168–176; (g) J. Jayabharathi, V. Thanikachalam and K. Jayamoorthy, Photophysical studies of some heterocyclic chromophores as potential NLO materials, Spectrochim. Acta, Part A, 2012, 89, 301–307; (h) J. Jayabharathi, V. Thanikachalam, N. Srinivasan and M. Venkatesh Perumal, Synthesis, spectral studies and solvatochromic analysis of novel imidazole derivatives, Spectrochim. Acta, Part A, 2012, 89, 194–200; (i) J. Jayabharathi, V. Thanikachalam, K. Brindha Devi and M. Venkatesh Perumal, Optical properties of organic nonlinear optical crystal – A combined experimental and theoretical study, Spectrochim. Acta, Part A, 2012, 86, 69–75; (j) J. Jayabharathi, V. Thanikachalam, N. Srinivasan, M. Venkatesh Perumal and K. Jayamoorthy, Physicochemical studies of molecular hyperpolarizability of imidazole derivatives, Spectrochim. Acta, Part A, 2011, 79, 137–147; (k) J. Jayabharathi, V. Thanikachalam, K. B. Devi and M. Venkatesh Perumal, Kamlet-Taft and Catalan Studies of Some Novel Y-shaped Imidazole Derivatives, J. Fluoresc., 2012, 22, 737–744.
12. M. J. Kamlet, J.-L. M. Abboud, M. H. Abraham and R. W. Taft, J. Org. Chem., 1983, 48, 2877–2887.
13. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03 (Revision E.01), Gaussian, Inc., Wallingford, CT, 2004.
14. R. W. Bigelow and G. E. Johnson, J. Chem. Phys., 1977, 66, 4861–4871.
15. M. S. Gudipati, J. Daverkausen, M. Maus and G. Hohlneicher, Higher electronically excited states of phenanthrene, carbazole and fluorene, Chem. Phys., 1994, 186, 289–301.
16. K. Chiba, J. I. Aihara, K. Araya and Y. Matsunaga, Bull. Chem. Soc. Jpn., 1980, 53, 1703.
17. S. P. McGlynn, T. Azumi and M. Kinoshita, Molecular spectroscopy of the triplet state, Prentice Hall, Englewood Cliffs, NJ, 1969.
18. D. M. Willard, R. E. Riter and N. E. Levinger, Dynamics of Polar Solvation in Lecithin/Water/Cyclohexane Reverse Micelles, J. Am. Chem. Soc., 1998, 120, 4151–4160.
19. S. Saha and A. Samanta, Influence of the structure of the amino group and polarity of the medium on the photophysical behavior of 4-amino-1,8-naphthalimide derivatives, J. Phys. Chem. A, 2002, 106, 4763.
20. C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, VCH, Weinheim, 2^nd edn, 1988.
21. J. S. S. de Melo, R. S. Becker and A. L. Macanita, Photophysical Behaviour of Coumarins as a Function of Substitution and Solvent. Evidence for the Existence of a Lowest Lying (n*) State, J. Phys. Chem., 1994, 98, 6054–6058.
22. E. Lippert, Spektroskopische bistimmung des dipolmomentes aromatischer verbindugen imersten angeregten singulettzustand, Z. Electrochem., 1957, 61, 962–975.
23. M. A. Haidekker, T. P. Brady, D. Lichlyter and E. A. Theodorakis, Bioorg. Chem., 2005, 33, 415–425.
24. C. Reichardt, Empirical parameters of solvent polarity as linear free energy relationships, Angew. Chem., Int. Ed. Engl., 1979, 18, 98–110.
25. R. A. Marcus, J. Chem. Phys., 1963, 38, 1858.
26. G. W. Castellan, Physical Chemistry, Narosa Publishing House, Delhi, 3^rd edn, 1985.
27. S. Dhar, D. K. Rana, S. S. Roy, S. Roy, S. Bhattacharya and S. C. Bhattacharya, Effect of solvent environment on the photophysics of a newly synthesized bioactive 7-oxy(5-selenocyanato-pentyl)-2H-1-benzopyran-2-one, J. Lumin., 2012, 132, 957–964.
28. W. Rettig and M. Zander, Chem. Phys. Lett., 1982, 87, 229.
29. L. Onsager, J. Am. Chem. Soc., 1936, 58, 1486–1486.
30. C. J. F. Böttcher, in Theory of Electric Polarization, ed. O. C. Van Belle, P. Bordewijk and A. Rip, Elsevier, Amsterdam, 1973, vol. I.
31. E. Z. Lippert, Naturforscher, 1955, 10, 541.
32. N. Mataga, Y. Kaifu and M. Koizumi, Bull. Chem. Soc. Jpn., 1955, 28, 690.
33. W. Liptay and E. C. Lim, in Excited States, Academic Press, New York, 1974, p. 129.
34. E. G. McRae, J. Phys. Chem., 1957, 61, 562–572.
35. A. Kapturkiewiez, J. Herbich, J. Karpiuk and J. Nowacki, J. Phys. Chem. A, 1997, 101, 2332.
36. J. Herbich and A. Kapturkiewiez, J. Am. Chem. Soc., 1998, 120, 1014–1029.
37. J. B. Birks, Photophysics of Aromatic Molecules, Wiley, New York, 1970, p. 51.
38. J. Michl and E. W. Thulstrup, Spectroscopy with Polarized Light, VCH, New York, 1986, pp. 28 and 75.
39. R. S. Mulliken, J. Am. Chem. Soc., 1952, 74, 811–824.
40. R. S. Mulliken and W. B. Person, Molecular Complexes: A Lecture and Reprint Volume, Wiley, New York, 1969.
41. R. A. Marcus, J. Phys. Chem., 1989, 93, 3078.
42. S. K. Saha and S. K. Dogra, J. Mol. Struct., 1998, 470, 301–311.
43. P. Purkayastha, S. C. Bera and N. Chattopadhyay, J. Mol. Liq., 2000, 88, 33–42.
44. S. K. Saha, S. Pandey and S. K. Dogra, Indian J. Chem., Sect. A: Inorg., Bio-inorg., Phys., Theor. Anal. Chem., 1995, 34, 771–777.
45. M. Swaminathan and S. K. Dogra, J. Am. Chem. Soc., 1983, 105, 6223–6228.
46. T. A. Fayed and S. S. Ali, Spectrosc. Lett., 2003, 36, 375–386.
47. J. J. P. Stewart, J. Comput. Chem., 1989, 10, 221–264.
48. C. J. F. Bottcher, Theory of Electronic Polarization, Elsevier, Amsterdam, 1983, vol. 1.
49. A.-D. Gorse and M. Pesquer, J. Phys. Chem., 1995, 99, 4039–4049.
50. P. Purkayastha and N. Chattopadhyay, Phys. Chem. Chem. Phys., 2000, 2, 203–210.
51. P. Purkayastha and N. Chattopadhyay, Int. J. Mol. Sci., 2003, 4, 335–361.
52. Z. R. Grabowski, K. Rotkiewicz, A. Siemiarczuk, D. J. Cowley and W. Baumann, Nouv. J. Chim., 1979, 3, 443–454.
53. S. Kato and Y. Amatatsu, J. Chem. Phys., 1990, 92, 7241–7257.
54. J. Lipinski, H. Chojnacki, Z. R. Grabowski and K. Rotkiewicz, Chem. Phys. Lett., 1980, 70, 449–453.
55. S. Marguet, J. C. Mialocq, P. Millie, G. Berthier and F. Momicchioli, Chem. Phys., 1992, 160, 265–279.
56. J. P. LaFemina, C. B. Duke and A. Paton, J. Chem. Phys., 1987, 87, 2151–2157.
57. D. Steiger, C. Ahlbrandt and R. Glaser, J. Phys. Chem. B, 1998, 102, 4257–4260.
58. A. E. Nikolaev, G. Myszkiewicz, G. Berden, W. L. Meerts, J. F. Pfanstiel and D. W. Pratt, J. Chem. Phys., 2005, 122, 084309.
59. Z. R. Grabowski, K. Rotkiewicz and W. Rettig, Chem. Rev., 2003, 103, 3899–4032.
60. Y. Amamatsu, Theor. Chem. Acc., 2000, 103, 445–450.

---

Footnote

† CCDC 960194. For crystallographic data in CIF or other electronic format see DOI: 10.1039/c3ra44994g

---