Exploring the isomerization paths of push–pull hexatrienes

Abstract

This computational study is an attempt to reveal the mechanism of the isomerization processes happening in donor (D)–acceptor (A) hexatriene systems. The photo-excitation of all-trans isomers of these conjugated systems with terminal donor (amine, methoxy) and acceptor (cyano) groups populates the first (S₁) and second (S₂) singlet excited states which correspond to transitions with reasonably high oscillator strength values. The S₁ state of the amine (D), cyano (A)-substituted system forms a twisted and couple of slightly off-planar minima connected by low-energy transition states with configurations equally dominated by HOMO → LUMO and HOMO² → LUMO² excitations. Two important low-lying S₀/S₁ conical intersections have been identified in this system at 7–8 kcal mol⁻¹ and 8–11 kcal mol⁻¹ above the twisted excited state minima. The first one has been identified as the source of a cis–trans–trans isomer while the latter one may be responsible for the trans–cis–trans isomer. In comparison, the presence of a weaker donor group (methoxy) produces a more stable cis–trans–trans isomer from a lower energy S₀/S₁ conical intersection, situated around 20–23 kcal mol⁻¹ below the vertically excited geometry. Both the isomers have an alternate thermal route of formation from the all-trans isomer through ground state transition states with activation energy values close to 50 kcal mol⁻¹.

---

1. Introduction

In recent years, extensive studies of the non-linear optical (NLO) properties of push–pull polyene systems^1–7 have been attempted. The effects of different donor–acceptor groups on two sides of the polyene chains are analyzed through the study of several properties, such as bond length alteration (BLA), polarizabilities, hyperpolarizabilities and so on. Semiempirical and ab initio quantum mechanical studies have been employed^1,3,4,6 to understand the electronic origin of the non-linear responses of these systems. Correlations between their HOMO–LUMO energy gaps and their hyperpolarizabilities^3,6 have shown interesting trends depending on the choice of donor and acceptor groups on the two ends. In spite of the huge number of NLO-related studies on simple non-aryl or non-heteroaryl donor–acceptor group substituted conjugated polyenes like hexatrienes, no significant photochemical investigations involving their low-lying excited states have been reported so far. On the other hand, the photochemistry and photophysics of the p,p′-disubstituted-1,6-diphenyl-1,3,5-hexatriene (DPH) systems^8–11 are experimentally well-studied; under photo-excitation they are known to give several isomeric products. However, it is quite interesting that the isomerization studies on these latter types of hexatriene systems have mostly been done on either donor–donor or acceptor–acceptor-substituted phenyl rings and, therefore, the systems reported are symmetrical. Only a few studies of the photo-isomerizations of the donor–acceptor substituted DPH systems^12,13 have been experimentally reported so far. To the best of our knowledge, no proper mechanism of the isomerization processes of the donor–acceptor hexatrienes has been investigated in detail which includes their non-radiative decay channels. The huge computational cost related to the substituted DPH systems for carrying out accurate photochemical studies at a meaningful level of calculations, such as CASSCF, might well be the cause behind the unexplored mechanism of their photo-isomerization paths. The possible participation of the whole system in the photo-isomerization process restricts the use of a hybrid scheme in this case, as well. However, there remains an alternate way of roughly understanding the whole process; attaching donor and acceptor groups directly (instead of donor, acceptor-substituted phenyl rings at the two ends of the hexatriene moiety) may reduce the computational cost. The actual mechanism here may be slightly different from the DPH systems, but it will certainly give us the flexibility to analyze in detail the excited state topographies which may not be significantly different from the phenyl hexatriene systems. Moreover, this study may also reveal the photo-isomerization paths of the simple non-aryl donor, acceptor-substituted hexatriene systems which are largely known for their NLO properties, as discussed earlier.

In our present computational study, we have chosen two simple donor–acceptor substituted hexatriene systems (Fig. 1A) where the donor groups are amino and methoxy while the acceptor group has been chosen as cyano. The photo-excitation processes of these polyenes are expected not to be exactly similar to the normal hexatriene system (Fig. 1B), and their photochemistry may lie more towards polar conjugated systems, such as 2,4-pentadien-1-iminium cation (PSB), the model compound (Fig. 1B) of a retinyl iminium ion. To elaborate this point, we need to discuss the difference in the photo-isomerization paths of non-polar hexatriene and a polar conjugated iminium ion. Their first two excited singlet (S₁ and S₂) states are exactly opposite in nature. This is due to the electronegativity of the nitrogen atom in the iminium ion which drags the electron cloud towards itself.¹⁴ The S₁ state of the conjugated triene^15–19 has a biradical characteristic (A_g symmetry), while this state in the PSB system is ionic in nature with B_u-like symmetry. In the hexatriene system, the S₀–S₁ is a forbidden transition. Here the initial photo-excitation populates the S₂ (B_u, ionic) state and this is followed by an ultrafast relaxation to the S₁ state which then undergoes a conical intersection (kinked CI) with the ground state through a Hula-twist motion.^20–22 On the other hand, the allowed vertical excitation to the S₁ state in the protonated Schiff base is followed by an ultrafast non-radiative decay through a barrier-less process of one-bond flip (OBF)^14,23–27 around a single bond (C–C) in this state which was originally a double bond in the ground state. The stability of the S₁ state increases during this rotation while the ground state becomes destabilized. Around a 90° torsion angle value, their energies match and a conical intersection, commonly known as twisted intramolecular charge transfer CI (TICT CI) between these two surfaces happens and, as a consequence, a radiation-less transition occurs to the ground state. In recent times, we have thoroughly investigated the photochemistry of zwitterionic conjugated open-chain nitrone systems^28,29 which include the N-alkyl retinyl nitrones and their model compounds. Their first photo-excited singlet states are mixed ionic-biradical in nature and their transition properties lie in between the non-polar polyenes and polar protonated Schiff bases. Here the lowest-energy conical intersection geometries were found to have a terminally-twisted CNO moiety, which roughly resembles the H₁-bridging conical intersections found in the terminal part of the non-polar conjugated polyene systems. There also exist Hula-twist-type kinked CI and OBF-type TICT CI geometries in these nitrones which involve the central part of the chain of these molecules; however, they are all energetically much higher than the CNO-kinked CI geometry. Our targeted systems in this present work have strong donor groups on one side (NH₂, OCH₃) while the other side has a strong acceptor cyano group. The latter will drag the π-cloud towards itself and this will be facilitated by the donor moieties. The situation seems to be quite similar to that of the iminium cations where the positively charged nitrogen attracts the electron cloud, and, therefore, these polyenes may have some overlapping features with them. However, the pushing of the donor groups may have significant influence on the charge transfer process operating in these systems and may cause the cis–trans isomerization paths to deviate significantly from those of the conjugated iminium ions.

Fig. 1 (A) The two currently studied push–pull systems and (B) non-polar hexatriene and polar protonated Schiff base systems.

2. Computational details

Most of the computational studies carried out in this work are based on the Complete Active Space Self-Consistent Field (CASSCF)^30–35 level of calculations with 6-31G* basis sets using the Gaussian 09 (ref. 36) suite of programs. The minimum energy geometries, transition states and conical intersection points on the potential energy surfaces (PES) are optimized at this CASSCF level of theory using a (10,10) active space size. Some of the active space orbitals are shown in the ESI (Fig. S1†). For locating the ground and excited transition states, the normal TS technique, based on the Berny-algorithm³⁷ has been employed. Dynamic correlation treatment at the CASPT2 level^38–40 has been done through single point studies on top of the CASSCF-optimized geometries using the MOLPRO⁴¹ package.

The radiative transition properties^42,43 for the low-lying vertical transitions at the ground state equilibrium geometries are studied at the Graphical Unitary Group Approach (GUGA) CI^44–47 level through the GAMESS⁴⁸ program package. visualization software programs such as Chemcraft⁴⁹ and GaussView are employed to analyze the output files in the work. All these different levels of quantum mechanical studies are utilized to establish a certain mechanism for the photo-isomerization processes of the studied polyene systems. However, it should be also mentioned in this context that from the point of view of the methodology employed in our work to track the correct decay paths through minima and CIs, there remains scope for improvement using other methods^50,51 which may increase the level of accuracy further.

3. Results and discussions

3.1. Excited state minima and transition states

The ground state and excited state geometries are optimized at the CASSCF (10, 10) level using 6-31G* basis sets. The ground state geometries of some stable isomers of both polyenes are shown in Fig. 2. The order of stability of the isomers in polyene I is found to be TTT > CTT > TCT > TCC > CCT > CTC (where C = cis and T = trans). On the other hand, the stability of the CTT isomer in polyene II is highest while the TCT and TTT isomers come next in the order. The ΔE_CTT–TCT value for polyene I is roughly 0.5 kcal mol⁻¹ at both CASSCF and CASPT2 levels of calculation; this value is slightly higher (2.5 kcal mol⁻¹) in the case of polyene II. The vertical excitations to their S₁ states have high transition moment values (∼8 debye); the S₀–S₂ transitions (Table 2) are also found to be moderately strong, with a roughly 2.5 debye transition moment for the all-trans isomers (TTT) in both systems. It has been noticed that the transition moment values of the latter transitions (S₀–S₂) are higher in the other two stable isomers (CTT and TCT), while their S₀–S₁ values for the same transitions are lower, especially in the TCT isomer (Table S1†). In this present work, we have studied the photo-excitation of the all-trans isomers (TTT) only. In polyene I, the vertical transition to the first excited singlet state is followed by relaxation to the non-planar optimized geometries (ES_Ia, ES_Ib, ES_Ic) situated around 26–28 kcal mol⁻¹ below the vertically excited state at the CASSCF level (Tables 1 and S2†) while the CASPT2 stabilizations were found to be slightly lower (∼16 kcal mol⁻¹). The dihedral angle, 〈N–C–C–C has a value of −162° in ES_Ia (Fig. 3), which changes to −55° in ES_Ib through a clockwise rotation and passes through a low-barrier transition state (TS_Ia) having a value of −106° for the same angle. This twisted relaxed excited state geometry (ES_Ib) is followed by another low-barrier transition state (TS_Ib) with the dihedral angle value further reduced to −38°, which becomes −13° in the subsequent minimum energy geometry (ES_Ic). The turn of this dihedral angle resembles the one-bond-flip motion normally found in the central part of the ionic S₁ states of conjugated iminium ion systems. However, we were not able to track any TICT-CI-type S₀/S₁ intersection geometry as a result this flip which is observed in the protonated Schiff base systems. Considering the fact that the bond lengths of the three minimum energy structures are almost similar, we have tested the energy of a guess geometry at CASPT2 level with 〈N–C–C–C dihedral angle value of 90° (usually the TICT emissive state geometry for several push–pull systems), keeping the bond length parameters the same as in the ES_Ib geometry. The ground state energy value at this geometry is found to be 21 kcal mol⁻¹ higher than that in the ES_Ib geometry and the S₀–S₁ energy gap is around 38–40 kcal mol⁻¹ (Table S2†). The latter gap is almost 23 kcal mol⁻¹ lower than this value in the ES_Ib and ES_Ia geometries (63–64 kcal mol⁻¹). Analysis of the dominating configurations in the different excited state geometries (Franck Condon & ES_Ia, ES_Ib, ES_Ic) has revealed an interesting fact; the S₁ state near the vertically excited point is dominated by HOMO → LUMO excitation, while the S₂ state is dominated by HOMO−1 → LUMO and HOMO² → LUMO² excitations. However, as the S₁ state goes towards the relaxed geometries (ES_Ia, ES_Ib and ES_Ic), almost equal shares of the HOMO² → LUMO² and HOMO → LUMO contributions are found; on the other hand, in the third singlet state, the dominance of the HOMO → LUMO clearly increases.

Fig. 2 Optimized ground state geometries of different isomers of (A) polyene I and (B) polyene II.

Table 1 Absolute energies (E; in hartree) and relative energies (ΔE; in kcal mol⁻¹) of some important points on the ground and excited state surfaces (VEE = vertical excitation energy)

Geometry	Polyene I				Polyene II
	CASSCF		CASPT2		CASSCF		CASPT2
	E	ΔE	E	ΔE	E	ΔE	E	ΔE
GS (TTT)	−378.7355	0	−379.7652	0	−437.5604	0	−438.7522	0
VEE	−378.5358	125.81	−379.5964	105.91	−437.3805	112.89	−438.5749	111.26
ESa	−378.5805	97.26	−379.6226	89.51	−437.4324	86.58	−438.6051	92.29
TSa	−378.5703	103.66	−379.6204	90.85	—	—	—	—
ESb	−378.5781	98.77	−379.6215	90.17	—	—	—	—
TSb	−378.5779	98.89	−379.6087	98.20	—	—	—	—
ESc	−378.5808	97.07	−379.6185	92.11	−437.4331	79.88	−438.6053	92.18
CIa	−378.5647	107.17	−379.6106	96.98	−437.4175	89.6	−438.6052	92.22
GS (CTT)	−378.7338	1.06	−379.7618	2.12	−437.5881	−17.4	−438.7543	−1.34
GS (CTC)	−378.7289	4.14	−379.7594	3.64	—	—	—	—
VEE (S2)	—	—	−379.5920	108.71	—	—	−438.5734	112.61
ES (S2)	−378.5381	123.8	−379.6073	99.1	−437.3819	112.0	−438.57802	109.29
CI (S2/S1)	−378.5413	121.8	−379.5825	114.6	−473.3637	123.4	−438.56777	115.72
CIb	−378.5693	104.29	−379.6231	89.14	—	—	—	—
GS (CCT)	−378.7310	2.82	−379.7613	2.42	—	—	—	—
GS (TCT)	−378.7329	1.63	−379.7609	2.69	−437.5843	−14.99	−438.7507	0.9
TSGS(TCT)	−378.6596	47.62	−379.6937	44.8	−437.5148	28.61	−438.6807	44.87
TSGS(CTT)	−378.6551	50.45	−379.6881	48.38	−437.4886	45.05	−438.6719	50.38
CId	−378.5590	110.75	−379.6086	98.2	−437.3826	111.57	−438.5960	98.01
GS (TCC)	−378.7322	2.07	−379.7632	1.25	—	—	—	—

---

Table 2 Comparative transition properties of different polyene systems at their respective ground state equilibrium geometries^a

Polyene	Transition moment (debye)			Oscillator strength			Einstein's coefficient (s^−1)
	S0–S1	S0–S2	S1–S2	S0–S1	S0–S2	S1–S2	S0–S1	S0–S2
a The values in parentheses are the powers to the base 10.
Polyene I	8.44	2.39	2.80	2.42	0.29	0.02	8.41 (+9)	8.49 (+8)
Polyene II	8.24	2.58	2.93	2.44	0.25	0.02	9.54 (+9)	1.12 (+9)
Hexatriene	0.96	7.76	4.30	0.02	1.86	0.01	7.77 (+7)	5.35 (+9)
Conjugated N-methyl nitrone	4.86	0.11	0.06	0.70	0.00	0.00	2.14 (+9)	1.05 (+6)
Protonated Schiff base	7.54	1.89	1.93	1.69	0.10	0.02	4.53 (+9)	3.11 (+8)

---

Fig. 3 Optimized geometries of excited state minima and transition states of polyene I and schematic representation of the excited state surface (values in the bracket are relative energies in kcal mol⁻¹ w.r.t. the TTT ground state energy).

In polyene II, we have not obtained any twisted ES_b-type geometry where only ES_a-type (〈OCCC = −156°) and ES_c-type geometries (ES_IIa and ES_IIc) were found. Interestingly, the latter geometry is almost planar (〈OCCC = −3°) and heavily dominated by HOMO² → LUMO² and HOMO−1 → LUMO excitations while the contribution from HOMO → LUMO is insignificant. At this point we can compare them with the leading configurations observed in the first two excited singlet states of the non-polar polyenes and protonated Schiff base systems. In non-polar hexatriene, both the vertically excited and the relaxed excited S₁ states have mixed HOMO² → LUMO² and HOMO−1 → LUMO contributions while the excited S₂ state is throughout dominated by the ionic HOMO → LUMO configuration. On the contrary, the latter configuration is largely responsible for the first excited singlet state of the protonated Schiff bases, with some contributions from the doubly excited HOMO² → LUMO² configuration in the relaxed geometry.

However, the second excited singlet state of this iminium ion has a mixed dominance from the configurations arising from HOMO² → LUMO² and HOMO−1 → LUMO excitations. This comparative analysis indicates that the relaxed S₁ states of both push–pull polyenes are quite unique in nature; while all minima in polyene I have almost equal shares of the HOMO → LUMO and HOMO² → LUMO² contributions, the second almost planar minimum-energy geometry in polyene II has a resemblance to the first singlet excited state of a normal hexatriene system. However, we were unable to find any connectivity between the ES_IIa and ES_IIc minima in this polyene, which indicates that the latter geometry might result from photo-excitation of the CTT isomer instead of TTT. To gain an idea of the energy of a twisted ES_b-type geometry (which has not been obtained by optimization) in this polyene, we have checked single point CASPT2 energies at 〈N–C–C–C dihedral angle values of −55° and −90°, keeping the bond length parameters similar to the ES_IIa geometry. In comparison to polyene I, the energy values of polyene II have shown a contrasting feature; it has been observed that the energy value of the ground state steeply rises (by 18 kcal mol⁻¹) from the ES_a-type geometry to the ES_b-type geometry, while in the other polyene this value changes by only 1.5 kcal mol⁻¹. Overall, it seems that in polyene I there are at least 2 possible energy minima (ES_Ia and ES_Ib) while in polyene II there exists only one such minimum energy geometry (ES_IIa).

3.2. Conical intersections & their related isomerization routes

(a) CTT and CTC isomers. In polyene I, following the OBF motion which changes the 〈N–C–C–C dihedral angle from −162° to −13°, we have not obtained any completely planar stable geometry of the S₁ state; rather we have obtained an optimized S₀/S₁ conical intersection geometry (CI_Ia) at 7–9 kcal mol⁻¹ above the ES_a/b/c geometries with a planar 〈N–C–C–C dihedral angle (Fig. 4A), however, interestingly with an out-of-plane C–C–C kink on the other side of the chain. In all the relaxed geometries, this part of the chain was not found to leave planarity. This type of kink is a characteristic of the biradicaloid S₁ states of neutral hexatrienes which happens in the middle part of the kinked-CI (S₀/S₁) geometry and arrives through a Hula-twist motion. This CI_Ia geometry in polyene I is found to lead to the CTT isomer following the direction of its gradient difference vectors (Fig. 4B); the energy of this isomer is found to be slightly above (∼1 kcal mol⁻¹) that of the all-trans isomer (TTT). In case, the kink part follows a similar motion to the Hula-twist type in the kinked-CI of normal hexatrienes, so we can also expect a CTC isomer (Fig. S2†), which is found to be the most unstable isomeric form. The latter geometry is slightly non-planar to avoid the H–H interaction.

Fig. 4 (A) Some optimized conical intersection (S₀/S₁) geometries of polyene I and polyene II and (B) their corresponding gradient difference and derivative coupling vectors.

The CASSCF calculations have shown that the CI_Ia geometry is situated at 8 kcal mol⁻¹ above the TS_Ib/c geometries (Table 1) while the dynamic correlation treatment has predicted its energy to be around 1 kcal mol⁻¹ below the transition state energies. This similar type of conical intersection in polyene II (Fig. 4) is found to lie at a much lower energy level (0–3 kcal mol⁻¹ below ES_IIa), which also seems to be heading towards the most stable CTT isomer like the other polyene. A comparison reveals that the energy difference between the CI_a intersection geometries and the CTT isomers are almost similar (∼106 kcal mol⁻¹ at CASSCF and 94 kcal mol⁻¹ at CASPT2) in both polyene systems.

(b) CCT & TCC isomers. Another important photochemical path of the TTT isomer is related to the S₂ photo-excited state. A relaxed geometry of this state is situated around 7–9 kcal mol⁻¹ below the vertically excited state. This geometry seems to be followed by an S₂/S₁ conical intersection (CI_S2/S1), which involves a twist around the N–C–C–C portion (Fig. S3†), but it is not similar to the OBF-type turn noticed in the S₁ state. The energy of this intersection is found to vary with respect to the S₂ minimum, depending on the level of calculation employed; at the CASSCF level, this intersection is 2 kcal mol⁻¹ below the S₂ minimum, while at the CASPT2 level it is roughly 14 kcal mol⁻¹ above this geometry in polyene I. This S₂/S₁ conical intersection is followed by the lowest energy S₀/S₁ conical intersection geometry (CI_Ib), which lies around 18–25 kcal mol⁻¹ below the former one. This lowest-energy CI (Fig. 4) leads to the CCT isomer, which lies above (∼2.5 kcal mol⁻¹) the TTT isomer and is found to be less stable than both TCT and CTT isomers in this polyene. A third conical intersection geometry (CI_Ic) has been detected (Fig. S3†) at a substantially high-energy level (156 kcal mol⁻¹ above the TTT ground state) and seems to be heading towards the TCC isomer; the latter is found to be the fourth stable isomer of polyene I. Calculations corresponding to the conical intersections, CI_b and CI_c are not attempted for polyene II.

(c) TCT isomer. The TCT isomer is expected to arise from a centrally kinked S₀/S₁ CI geometry (CI_Id), identified at 3 kcal mol⁻¹ above CI_Ia in polyene I (Fig. 4). In the other polyene, the energy gap between CI_IIa and CI_IId is much higher. Usually, the normal hexatrienes are characterized by this type of conical intersection during their TTT–TCT isomerization through the Hula-twist motion of the middle part of the conjugated chain; however, for the current polyenes, the GD and DC vectors are found to be substantially different from those CIs and formation of the TCT isomer from CI_d is actually questionable.

3.3. Ground state isomerizations

We have obtained alternate thermal routes for the TCT and CTT isomer formations from the all-trans system. The less stable (by 1.6–2.6 kcal mol⁻¹ than TTT) TCT isomer in polyene I requires roughly 45–47 kcal mol⁻¹ of activation energy for the TTT–TCT conversion (Fig. 5) through a transition state having an imaginary frequency of 556i cm⁻¹. On the other hand in polyene II, the less stable TTT isomer gets converted to the more stable TCT form through a transition state situated at 28 kcal mol⁻¹ (at CASSCF level) above the TTT isomer, having an imaginary frequency of 440i cm⁻¹; however, the CASPT2 value of this barrier is 45 kcal mol⁻¹. Similar isomerization paths have also been predicted for the TTT–CTT isomerization (Fig. S4†), where 45–50 kcal mol⁻¹ activation energy is required and the frequencies are nearly 450i cm⁻¹ in both polyenes.

Fig. 5 Schematic representation of the possible thermal cis–trans isomerization (TTT isomer to TCT isomer) of (A) polyene I and (B) polyene II.

3.4. Comparison with other conjugated systems

A comparison of the transition moment data of these polyenes with other conjugated systems reveals that they are quite similar to the protonated Schiff base system; the transition moment (and corresponding oscillator strength) values of the strongly allowed S₀–S₁ and weakly allowed S₀–S₂ transitions are in fact higher than those of the conjugated iminium ions. Their nature is completely different from the normal hexatrienes, where the S₀–S₁ is totally forbidden and S₀–S₂ is strongly allowed. This similarity with the iminium ion is due to the fact that at the vertical excitation point, the configurations dominating the S₁ (HOMO → LUMO) and S₂ (HOMO² → LUMO² and HOMO−1 → LUMO) states for both types of systems are similar. However, as we go towards the relaxed excited state geometries of the push–pull polyenes, the contribution of the HOMO² → LUMO² configuration increases significantly, which does not happen in the PSB systems. The presence of such a doubly excited configuration is found to dominate the relaxed excited state (S₁) geometry of the normal hexatriene, which undergoes the Hula-twist motion to produce a kink in the middle unlike the OBF motion of iminium ions. The unique characteristic of the push–pull hexatrienes with equal dominance of HOMO → LUMO and HOMO² → LUMO² may well be the reason for producing a conical intersection (CI_a) having an OBF-type turn on one side (the donor side) and a Hula-twist-type kink on the other side (the acceptor side), which subsequently produces the stable CTT isomer. The S₂ states of these polyenes are not populated to the same extent as the normal hexatrienes, but they are not completely ignorable and the S₂/S₁ conical intersection is followed by the vertical excitation to these states; this is subsequently followed by the lowest-energy S₀/S₁ CI (CI_b), which should produce the slightly sterically hindered CCT isomer. None of the conical intersections are found to lie below the relaxed excited state geometries and therefore they are not easily accessible after relaxation from the FC geometry of S₁. It is possible that similar to CI_b, the other S₀/S₁ intersections may be connected to some S₂/S₁ CIs which we were not able to track, and these may facilitate the photo-isomerization paths to some extent. However, it must be kept in mind that the S₂ population is not significantly high in these polyenes. Overall, there are enough indications that the isomerization processes are not very efficient in these types of push–pull hexatriene systems.

3.5. Donor, acceptor-substituted diphenyl hexatriene systems

Experimental studies by Sonoda et al.^12,13 on the photo-excitations of all-trans-p,p′-substituted-1,6-diphenyl-1,3,5-hexatrienes (Fig. 6) with donor–acceptor groups on the two ends have reported the presence of CTT and TCT isomers as the only products. The photo-excitation of the TTT isomer of p-methoxy-p′-nitro-substituted-1,6-diphenyl-1,3,5-hexatriene (DPH1) in toluene was found to give only the CTT isomer and a good yield of fluorescence. On the other hand, in acetonitrile, the fluorescence yield was very poor which is in contrast to the trend found in symmetrical donor, donor or acceptor, acceptor-DPH systems. In this polar solvent, CTT and TCT isomers were obtained for DPH1 through inefficient isomerizations. The fluorescence peak was found to be red shifted with increasing solvent polarity. Replacing the –OMe group by the –NMe₂ group in the above-mentioned DPH system (DPH2) was found to produce dual emission peaks. The lower wavelength peak was predicted to be for the normal fluorescence from the planar LE state, which remains unaltered with solvent polarity while the higher wavelength one is the anomalous fluorescence peak arising from the TICT state. It must be added here that our computational studies based on the DFT method have revealed that for both these di-substituted DPH systems, the TTT isomers are the most stable, followed by the TCT isomers, situated at 2 kcal mol⁻¹ above the former ones. The CTT isomer is the least stable of the three and lies around 1.5 kcal mol⁻¹ above the TCT isomers.

Fig. 6 Donor, acceptor-substituted DPH systems.

It is quite obvious that the photochemical and photophysical behavior of the molecules with substituted phenyl rings on the hexatriene might be somewhat different from the currently investigated simple donor, acceptor group-substituted hexatrienes. However, it is possible to get an approximate idea which may justify the experimental observations on the above-mentioned DPH systems. In any solvent, the ES_a-type minima will be easily accessible after relaxation from the Franck–Condon geometry and radiative emission may happen from this state. At the same time, this state seems to be connected to the CI_a geometry, which will give a CTT isomer through a non-radiative decay channel. Therefore, following this scheme, in a low-polarity medium, both isomerization (TTT–CTT) and fluorescence can be expected for the substituted DPH systems, as reported by experimental results. In more polar solvents the more twisted structure ES_b can be reached by crossing the low-barrier excited transition state while the non-radiative path leading to the CTT isomer (from ES_a) through CI_a can still be effective. In DPH1, the low fluorescence yield in acetonitrile with inefficient isomerization leading to a mixture of CTT and TCT isomers clearly indicates that the emission is from the twisted state and the S₀–S₁ gap is very low, which supports a non-radiative process. The appearance of the TCT isomer in only polar solvents and overall poor isomerization in these solvents seem to be connected to the lower fluorescence yield in these solvents than in the non-polar solvents. This certainly opens up a possibility that in the polar solvent, the low energy gap in the twisted ES_b state may lead to efficient S₁–S₀ internal conversion (IC) which populates the vibrationally excited level of the ground state and subsequently leads to the ground state isomerization paths of TTT by crossing the barrier to form the TCT isomer, which is more stable than the CTT in DPH systems. Therefore, this isomerization seems to originate from the excess vibrational energy in the ground state. Normally under a thermal process, the barrier for TTT–TCT is quite substantial and may be difficult to overcome in practice. However, this above-mentioned process will also lead to the TTT isomer and will certainly affect the overall isomerization efficiency. This type of ground state isomerization through an S₁–S₀ IC process originating from the twisted S₁ geometry is a well-known phenomenon observed in several systems. Alternatively, it may also happen that the centrally twisted CI_d-type conical intersection becomes slightly more accessible in polar solvents and leads to the TCT isomer. However, the first possibility seems to be more acceptable, as the vectors of this CI geometry (in polyene I and polyene II) have shown no possibility of TCT isomer formation.

4. Conclusion

This study has put forward a possible mechanism for the photo and thermal isomerization processes of push–pull hexatriene systems. Experimental studies on the photo-isomerizations on the all-trans hexatrienes with donor–acceptor substituted phenyl groups at the two ends were reported previously. These results had clearly indicated the dominance of the cis–trans–trans and trans–cis–trans isomers as the products. Our present study on terminally-substituted hexatriene systems with relatively smaller non-aryl donor–acceptor groups has also indicated the possibility of these two isomers due to their better stabilities. These photo-isomerization routes through conical intersections have been thoroughly explored along with their ground state thermal isomerization channels. One major important result of this work lies in the prediction of the mechanism for the isomerizations of the experimentally reported larger donor–acceptor-substituted hexatriene systems through computationally cost-effective smaller systems. This study can be also viewed from another important perspective; these proposed results can be of huge significance in revealing the photo-excitation processes of the non-aryl donor–acceptor-substituted polyenes. In spite of the huge number of NLO-related studies, the photo-isomerizations of these systems have not been reported, so far. Overall, the proposed non-radiative channels through conical intersections and thermal isomerization paths can have far-reaching consequences in terms of revealing the fate of photo-excitations of an important class of organic compounds and may give significant information to experimentalists to carry out further work on related systems in future.

Acknowledgements

We gratefully acknowledge the financial support provided to our department by the Department of Science and Technology (DST) under the DST-FIST program.

References

1. Y. Sheng, Y. Jiang and X.-C. Wang, J. Chem. Soc., Faraday Trans., 1998, 94, 47È52.
2. G. Chen and S. Mukamel, J. Chem. Phys., 1995, 103, 9355.
3. W. Rettig and M. Dekhtyar, Chem. Phys., 2003, 293, 75.
4. S. D. Bella, I. L. Fragala, M. A. Ratner and T. J. Marks, J. Am. Chem. Soc., 1993, 115, 683.
5. Y. Luo, H. Agren and S. Stafstriim, J. Phys. Chem., 1994, 98, 7782.
6. N. S. Labidi, Open Access Library Journal, 2014, 1, 1.
7. N. S. Labidi and A. Djebaili, Mater. Sci. Eng., B, 2010, 169, 28.
8. J. Saltiel, G. Krishnamoorthy, Z. Huang, D.-H. Ko and S. Wang, J. Phys. Chem. A, 2003, 107, 3178.
9. J. Saltiel, S. Wang, D.-H. Ko and D. A. Gormin, J. Phys. Chem. A, 1998, 102, 5383.
10. J. Saltiel, S. Wang, L. P. Watkins and D.-H. Ko, J. Phys. Chem. A, 2000, 104, 11443.
11. C. T. Lin, H. W. Guan, R. K. McCoy and C. W. Spangler, J. Phys. Chem., 1989, 93, 39.
12. Y. Sonoda and Y. Kawanishi, Chem. Lett., 2003, 32, 978.
13. Y. Sonoda, M. Goto, S. Tsuzuki and N. Tamaoki, J. Phys. Chem. A, 2006, 110, 13379.
14. M. Garavelli, P. Celani, F. Bernardi, M. A. Robb and M. Olivucci, J. Am. Chem. Soc., 1997, 119, 6891.
15. R. A. Mathies, C. H. B. Cruz, W. T. Pollard and C. V. Shank, Science, 1988, 240, 777.
16. J. Dobler, W. Zinth, W. Kaiser and D. Oesterhelt, Chem. Phys. Lett., 1988, 144, 215.
17. W. Zinth and D. Oesterhelt, Photobiology, ed. E. Ricklis, Plenum Press, New York, 1991.
18. T. Arlt, S. Schmidt, W. Zinth, U. Haupts and D. Oesterhelt, Chem. Phys. Lett., 1995, 241, 559.
19. D. Xu, C. Martin and K. Schulten, J. Biophys., 1996, 70, 453.
20. J. E. Norton and K. N. Houk, Mol. Phys., 2006, 104, 993.
21. M. Olivucci, F. Bernardi, P. Celani, I. Ragazos and M. A. Robb, J. Am. Chem. Soc., 1994, 116, 1077.
22. P. Celani, M. Garavelli, S. Ottani, F. Bernardi, M. A. Robb and M. Olivucci, J. Am. Chem. Soc., 1995, 117, 11584.
23. A. Cembran, F. Bernardi, M. Olivucci and M. Garavelli, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 6255.
24. D. S. Ruiz, A. Cembran, M. Garavelli, M. Olivucci and W. Fu, Photochem. Photobiol., 2002, 76, 622.
25. J. A. Dobado and M. Nonella, J. Phys. Chem., 1996, 100, 18282.
26. I. Conti, F. Bernardi, G. Orlandi and M. Garavelli, Mol. Phys., 2006, 104, 915.
27. I. Conti and M. Garavelli, J. Photochem. Photobiol., A, 2007, 190, 258.
28. P. Saini and A. Chattopadhyay, Chem. Phys. Lett., 2015, 633, 6.
29. P. Saini and A. Chattopadhyay, RSC Adv., 2014, 4, 20466.
30. D. Hegarty and M. A. Robb, Mol. Phys., 1979, 38, 1795.
31. R. H. A. Eade and M. A. Robb, Chem. Phys. Lett., 1981, 83, 362.
32. H. B. Schlegel and M. A. Robb, Chem. Phys. Lett., 1982, 93, 43.
33. F. Bernardi, A. Bottini, J. J. W. McDougall, M. A. Robb and H. B. Schlegel, Faraday Symp. Chem. Soc., 1984, 19, 137.
34. M. J. Frisch, I. N. Ragazos, M. A. Robb and H. B. Schlegel, Chem. Phys. Lett., 1992, 189, 524.
35. N. Yamamoto, T. Vreven, M. A. Robb, M. J. Frisch and H. B. Schlegel, Chem. Phys. Lett., 1996, 250, 373.
36. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, 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, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision D.01, Gaussian Inc., Wallingford CT, 2010.
37. X. Li and M. J. Frisch, J. Chem. Theory Comput., 2006, 2, 835.
38. C. R. Kemnitz, G. B. Ellison, W. L. Karney and W. T. Borden, J. Am. Chem. Soc., 2000, 122, 1098.
39. M.-X. Song, Z.-X. Zhao, F.-Q. Bai, Y.-J. Liu, H.-X. Zhang and C.-C. Sun, J. Phys. Chem. A, 2010, 114, 7173.
40. T. Shiozaki, C. Woywod and H.-J. Werner, Phys. Chem. Chem. Phys., 2013, 15, 262.
41. H.-J. Werner, P. J. Knowles, G. Knizia, F. R. Manby, M. Schütz, P. Celani, W. Györffy, D. Kats, T. Korona, R. Lindh, A. Mitrushenkov, G. Rauhut, K. R. Shamasundar, T. B. Adler, R. D. Amos, A. Bernhardsson, A. Berning, D. L. Cooper, M. J. O. Deegan, A. J. Dobbyn, F. Eckert, E. Goll, C. Hampel, A. Hesselmann, G. Hetzer, T. Hrenar, G. Jansen, C. Köppl, Y. Liu, A. W. Lloyd, R. A. Mata, A. J. May, S. J. McNicholas, W. Meyer, M. E. Mura, A. Nicklaß, D. P. O'Neill, P. Palmieri, D. Peng, K. Pflüger, R. Pitzer, M. Reiher, T. Shiozaki, H. Stoll, A. J. Stone, R. Tarroni, T. Thorsteinsson and M. Wang, MOLPRO, version 2012.1. A package of ab initio programs, Cardiff, UK, 2012.
42. B. Brooks and H. F. Schaefer, J. Chem. Phys., 1979, 70, 5092.
43. B. Brooks, W. Laidig, P. Saxe, N. Handy and H. F. Schaefer, Phys. Scr., 1980, 21, 312.
44. A. Chattopadhyay, J. Phys. B: At., Mol. Opt. Phys., 2012, 45, 165101.
45. A. Chattopadhyay, J. Chem. Sci., 2012, 124, 985.
46. F. Weinhold, J. Chem. Phys., 1970, 54, 1874.
47. C. W. Bauschlicher and S. R. Langhoff, Theor. Chim. Acta, 1991, 79, 93.
48. M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. J. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L. Windus, M. Dupuis and J. A. Montgomery, J. Comput. Chem., 1993, 14, 1347.
49. Website of Chemcraft software, https://www.chemcraftprog.com.
50. I. Schapiro, D. Roca-Sanjuán, R. Lindh and M. Olivucci, J. Comput. Chem., 2015, 36, 312.
51. L. D. Vico, M. Olivucci and R. Lindh, J. Chem. Theory Comput., 2005, 1, 1029.

---

Footnote

† Electronic supplementary information (ESI) available: Table S1: comparison of radiative transition properties of isomers of polyene I and II; Table S2: energies of first three low-lying states of some important points at CASPT2 level; Fig. S1: (10,10) active space orbitals; Fig. S2: geometries of some less stable isomers of polyene I; Fig. S3: other conical intersections; Fig. S4: TTT–CTT thermal isomerization. See DOI: 10.1039/c6ra16812d

---