Phototransformation of tetrazoline oxime ethers – part 2: theoretical investigation

Abstract

The competitive photoisomerization and photodegradation reactions of a tetrazoline oxime ether showing important fungicidal activity are investigated by Time Dependent-Density Functional Theory (TD-DFT) and Complete Active Space Self-Consistent Field (CASSCF) calculations of ground state and excited state (singlet, S and triplet, T) potential energy surfaces. Two key experimental results reported previously are explained in this study: (1) why only E isomers undergo N–O bond scission and photodegradation and (2) what is the mechanism for deactivation of the Z isomers. In addition, the reactive pathway that involves intersystem crossing is shown to be competitive with reactions in the singlet state. Our results demonstrate that the photoreactivity (photodegradation or photoisomerization) of the eight conformers studied here is clearly related to the differences between the respective electronic configurations of the lowest singlet excited state. It is shown that the addition of a bulky substituent on the pyridyl group prevents E isomers from being photodegraded.

---

I. Introduction

The development of plant protection species faces key challenges such as photoinduced degradation reactions which significantly impact their efficiency.^2,3 Thus, understanding the mechanism pertaining to photodegradation is important to prevent dose increase or repetitive treatments. Many pesticides contain an oxime moiety.⁴ In the dark, oxime ethers are stable but reactive under light exposure.³ In a recent study,¹ we have shown that upon irradiation, the (Z)-but-3-yn-1-yl(6(((((1-methyl-1H-tetrazol-5-yl)(phenyl)methylene)amino)oxy)methyl)pyridine-2-yl)carbamate species (named Z hereafter; see Fig. 1) underwent photoizomerisation (Z ⇌ E) and photodegradation by homolytic cleavage of the N–O bond. Both phototransformations are common photochemical processes.^3,5–15 This fungicide is investigated by Bayer Crop Science (WO2003/016303 DaiNippon) because it is very active biologically in green house while much less active in the field. E does not show fungicide activity. Thus, the formation of the E isomer inhibits the crop protecting properties of this oxime species.

Fig. 1 The model structures used in this work for the theoretical calculations. Only the Z isomers are represented in this scheme. E isomers are obtained through rotation about the C=N bond (δ). δ, δ₁, δ₂ and δ₃ are C_PhC–NO, C_Ph–CN, N*C_Tz–CN and OCC_PyN_Py dihedral angles, respectively. CASSCF calculations were performed on the smaller Z_m model molecule.

Our recent results have shown that (1) the lower absorptivity of E compared to Z favors its accumulation and (2) the photodegradation proceeds exclusively from E (Z undergoes only photoisomerization) and starts by the cleavage of the N–O bond into iminyl and alkoxyl radicals (see Fig. 2). These results explained the experimental loss of biological activity of this species. In this study, the photoisomerization and photodegradation mechanisms are discussed on the basis of theoretical calculations in the ground state and low-lying excited states.

Fig. 2 The photoreaction scheme for Z (Ph = phenyl, Tz = tetrazole, Py = pyridyl, R = NHCOOC₄H₅). Experimental¹ rate constants are k_Z→E = 0.13 min⁻¹, k_E→Z = 0.07 min⁻¹, k_{E→Photoproducts} = 0.011 min⁻¹.

The simpler model structure Z′ (see Fig. 1) was also studied in order to investigate the role of the pyridyl group (Py) in Z.

II. Computational results and discussion

The photodegradation mechanism

The cis/trans C=N photoisomerization and photodegradation mechanisms were investigated by quantum calculations.¹⁶ TD-DFT calculations were performed for Z and E at the B3LYP/6-31+G(d,p) level to obtain electronic absorption spectra and investigate the potential energy surfaces of the low lying states in order to understand the differences of photoreactivity seen experimentally between Z and E isomers. CASSCF calculations were also performed to explore the potential energy surfaces in the vicinity of conical intersections. Therse calculations intend to identify structures that cannot be obtained by TD-DFT calculations. The energies obtained at the CASSCF level are not discussed here. Details are provided in the appropriate sections and ESI.†

The photoisomerization¹⁷ of conjugated isomers generally proceeds through intersystem crossing (isc) via T₁ or adiabatic singlet deactivation (see Fig. 3). In the latter case, S₁ reaches the conical intersection (CI) between S₀ and S₁. Both isomers are obtained in this manner. The S₁ state may have n/π*¹⁸ or π/π* character.^19–23 In case of isc, the triplet state T₁ is reached from S₁. The geometry of the minimum energy structure in T₁ is one where δ ∼ 90 deg. (see Fig. 1 for definition of δ). Z and E isomers are obtained through further T₁ → S₀ isc deactivation.²⁴

Fig. 3 The photoisomerization mechanism. The conical intersection (S₀/S₁) is labeled CI. Dihedral values are given as indicative purposes. Arrows describe the singlet state mechanism. See Fig. 1 for definition of δ.

The photodegradation of oxime species may also compete with the photoisomerization mechanism. Experimental analyses^1,3,10–15 and theoretical studies¹⁸ of photodegradation products gave evidence of N–O bond dissociation as one of the main degradation pathway. The photodissociation mechanism of the N–O bond can proceed through conical intersections that arise from the diabatic crossing of excited singlet states that have significantly different electronic character (π/π* and σ_NO/).¹⁸ If a local minimum energy structure is found in the S₁ state, an activation energy barrier is expected for N–O bond dissociation. Photodissociation may also be obtained through intersystem crossing. Finally, photocyclization products were also reported in photoexcitation experiments of conjugated photoisomers.^23,25

Structural analysis

An extensive search for the most stable Z and E conformers was performed because our results showed that the photoreactivity of these species was dependent on conformation. These preliminary calculations were based on the systematic variation of dihedral angles. Four conformers of similar energies (within less than 2 kcal mol⁻¹) were found (see Table 1 for energies; see Fig. 4 and ESI† for geometries). The energy of the other conformers were larger than 3.50 kcal mol⁻¹ above the global minimum. The relative weight of isomer Z(i) may thus be calculated according to , where ΔE_i is the relative electronic energy of conformer Z(i) with respect to the global minimum in the Z configuration, here Z-b. The relative energies of isomers Z-a, Z-b, Z-c and Z-d were 0.61, 0.0, 0.62 and 0.06 kcal mol⁻¹, respectively. Thus a species with energy larger than 3.50 kcal mol⁻¹ above Z-b would represent less than 0.1% of the population of the Z isomers (e^−3.5/RT/(e^−0.61/RT + 1 + e^−0.62/RT + e^−0.06/RT + e^−3.50/RT) = 0.1%). Relative weights are given in Table 1. Similar results were found for the E isomer, E-a being the most stable species (see Table 1 for energies and relative weights; see Fig. 5 and ESI† for geometries). The isomers of lowest energy Z-b and E-a were separated by 0.29 kcal mol⁻¹ (E-a is the global minimum). Our results will show that these structures react differently upon irradiation. The relationship between photoreactivity and structure is investigated in this work. Note also that hydrogen bonding N*–H* (see Fig. 1 and 4a) was observed only in Z-a. It is shown below that this feature was key to explain the singular photoreactivity of Z-a. Among E isomers, only E-c is slightly constrained by hydrogen bonding (N_Py–H_Ph = 3.538 Å).

Table 1 Electronic properties of the ground state and excited states of the isomers investigated in this work at the B3LYP/6-31+G(d,p) level

	S0^a	S1(FC)^b	λ(S1)^c	f(S1)^d	λ(S2)^c	f(S2)^d	PhR^e		S0^a	S1(FC)^b
a Energies in kcal mol^−1. Reference is the most stable species (Z-b, E-a or Zm-b). The relative weight of each conformer is given in parentheses. See text for calculation method.b Energy in kcal mol^−1 of the S1 state for the Franck–Condon structure.c Energy level in nm of the S1 and S2 states for the Franck-Condon structure.d Oscillator strengths of S1 and S2.e Photoreactivity: I for photoisomerization and P for photodegradation.
Z-a	0.61 (13.7%)	87.80	328	0.0005	287	0.2658	I	Zm-a	1.93 (2.0%)	79.47
Z-b	0.0 (38.3%)	100.22	285	0.3813	277	0.0112	I	Zm-b	0.0 (52.1%)	94.18
Z-c	0.62 (13.4%)	96.76	297	0.0126	285	0.4390	P	Zm-c	1.15 (7.5%)	87.35
Z-d	0.06 (34.6%)	98.87	289	0.4713	276	0.0019	I	Zm-d	0.18 (38.4%)	93.99
E-a	0.0 (68.4%)	97.61	293	0.0657	284	0.1819	P
E-b	0.69 (21.4%)	97.16	296	0.1390	285	0.1357	P
E-c	1.51 (5.4%)	97.06	289	0.2087	283	0.0076	I
E-d	1.58 (4.8%)	101.55	286	0.1906	281	0.0040	I

---

Fig. 4 The low energy structures of the Z isomers in the ground state at the B3LYP/6-31+G(d,p) level. (b) is the most stable Z isomer (see Table 1). The N*–H* bond distance is 2.270 Å in structure (a) (see also Fig. 1). The CH₃–N_Py bond distances are 2.825 Å and 3.015 Å in (b) and (d), respectively.

Fig. 5 The low energy structures of E isomers in the ground state at the B3LYP/6-31+G(d,p) level. (a) is the most stable isomer (see Table 1). The N_Py–H_Ph bond distance in structure (c) is 3.538 Å.

Electronic absorption spectra

The electronic absorption spectra of Z and E isomers were computed from TD-DFT calculations at the B3LYP/6-31+G(d,p) level and compared to experimental spectra (Fig. 6). Theoretical spectra were calculated from the respective weights of the isomers (see Table 1), assuming Gaussian band shapes with UV-Vis peak half width at half height of 0.333 eV. The position of excited states and oscillator strengths of the most stable Z and E isomers, i.e. Z-b and E-a, are also represented in this figure (see also Table 1).

Fig. 6 Experimental and theoretical UV spectra of Z (top) and E (bottom) isomers, and position of electronic excited states of the most stable isomers, i.e. Z-b (top) and E-a (bottom). Oscillator strengths are represented by vertical lines. The positions of S₁ and S₂ are also indicated. Theoretical spectra are averaged over the isomers according to their respective weight (see Table 1).

Z-Isomers

For Z-a, the absorption band at 287 nm is the S₀ → S₂ transition (Fig. 3 and Table 1). It is of π/π* character localized on the phenyl (Ph) and oxime (Ox) parts of the molecule. The tetrazole (Tz) substituent does not participate here because, in the Z isomer, the Ph and Ox substituents are coplanar and the twist between Tz and Ph is significant (δ₁ = 170.0 deg. and δ₂ = −110.1 deg.; see Fig. 1 and 4a). In Z-a, the hydrogen bond between the amino hydrogen (H*) and one of the Tz nitrogen atoms (N*) stabilizes the twisted structure (see labels in Fig. 1). The low-lying S₁ state in Z-a results primarily from excitation of π electrons from the Py substituent to the Ph and Ox parts of the molecule (oscillator strength f(S₁) = 5 × 10⁻⁴ at 328 nm; see Table 1). The two states (S₁ and S₂) switch in Z-b and the electronic configuration of S₁(Z-b) is similar to that of S₂(Z-a). The position of the absorption bands are well represented by theory at this level (see Fig. 6, top). Note that strong solvent interactions (water in our experimental work¹) may prevent hydrogen bonding in Z-a, and the weight of this structure may be dependent on the surrounding medium (liquid or solid phase, and solvent).

Finally, it was found that the electronic configuration of S₁ was similar for Z-a and Z-c on the one hand, and Z-b and Z-d on the other. The phototransformation of all these conformers will be studied in this work.

E-Isomers

The Ox and Tz groups are coplanar in the E isomers, and the Ph substituent is twisted (δ₁ = 133.8 deg. and δ₂ = 162.8 deg. in E-a; see Fig. 1 and 5a). The absorption band in the experimental spectrum at 280 nm results mainly from the overlap of S₁ and S₂. The S₀ → S₁ transition in E-a and E-b involves excitation of π electrons from the Py substituent (and to a less extent Ph/Ox) to the Ph, Tz and Ox parts of the molecule (f(S₁) = 6.57 × 10⁻² for E-a at 293 nm; see Table 1). S₂ involves excitation of π electrons mainly from the Ph/Tz/Ox part of the molecule (and to a less extent, Py) However, the ordering of electronic configurations in S₁ and S₂ (diabatic states) is switched for the less abundant E isomers (E-c and E-d) and thus, the photoreactive state of E-a/E-b is different from that of E-c/E-d.

Thus, the eight isomers investigated in this work (Fig. 4 and 5) can be separated into two sets: one where the first excited singlet state arose from promotion of a Py electron to the Ph/Ox/Tz part of the molecule, thus an electronic transfer through the oxime linkage (Z-a, Z-c, E-a, E-b), and the second group (Z-b, Z-d, E-c, E-d) where S₁ was characterized by an electronic transition that remained on the Ph/Ox/Tz part of the species. This is illustrated by the oscillator strengths given in Table 1. Large f(S₁) and small f(S₂) characterize species that belong to the second group (Z-b, Z-d, E-c, E-d), whereas small f(S₁) and large f(S₂) characterize species that belong to the first set (Z-a, Z-c, E-a, E-b; note that in E-b both S₁ and S₂ involve excitation from orbitals that have Py and Ph/Ox/Tz character, thus large f(S₁) and large f(S₂)). Thus, different phototransformation mechanisms are expected for the two sets of species. Last, although S₁ was similar for Z-a and Z-c/E-a/E-b, it will be shown below that Z-a reacted singularly because of the hydrogen bonding between the Py and Tz parts of the molecule.

Finally, the model molecule Z′ was also studied to investigate the role of the Py substituent on the phototransformation mechanisms.

Assuming photodegradation is governed by the reactivity of the first excited state, the difference of photoreactivity observed between Z and E isomers is strongly dependent on S₁ electronic configuration, as discussed in the next section.

The phototransformation of Z isomers

Optimization of Z-b and Z-d in S₁ at the TD-DFT/6-31+G(d,p) level yielded a structure similar to the conical intersection in Fig. 3 (CI). However, the actual location of CI cannot be obtained with monoconfigurational methods. CASSCF calculations were thus performed to identify this structure. The smaller model molecule Z_m (see Fig. 1) and the 6-31G basis were used here to reduce the computational cost (details of the computing procedure are given in ESI†). Preliminary TD-DFT calculations were performed on Z_m: conformers of similar energies and geometries were found for Z and Z_m (see Table 1 and ESI†). Moreover, the excited states of both species exhibited similar electronic configurations, and S₁ optimizations yielded the same structures for all the conformers of Z and Z_m. Detailed analysis of the low lying excited state configurations showed that an active space with the 4 highest molecular orbitals and 3 lowest MOs of Z_m-b in S₀ included all the MOs of interest, thus a CAS(8,7) calculation. 50–50 state averaged optimization of Z_m-b yielded the conical intersection conformation (δ = 92.1 deg.; active space and full structure are given in the ESI†).

Optimization of Z-c in S₁ yielded a minimum-energy structure with significant increase of the N–O bond (R_N–O(S₀) = 1.390 Å and R_N–O(S₁) = 1.477 Å). The Minimum Energy Path (MEP) was followed for fixed N–O distances and interpolated in order to estimate the energy barrier for bond dissociation and thus photodegradation. The electronic energy barrier was 6.8 kcal mol⁻¹ at the TD-B3LYP/6-31+G(d,p) level (see ESI†). Thus, rapid photodegradation through N–O bond scission is expected for Z-c. This conformer amounts to 13.4% of the Z isomers (see Table 1).

Z-a reacted singularly in S₁ because of significant hydrogen bonding between the Py and Tz parts of the molecule (see Fig. 4a). Considering that the electronic configuration of S₁ was similar for Z-a and Z-c, Z-a was expected to give photodegradation products through N–O bond scission. However, optimization of Z-a in the S₁ state yielded a structure where the hydrogen atom H* in Fig. 1 was shifted to N* (Tz). S₀ and S₁ are close-lying in this region of the PES, and the DFT method is not suited to identify the respective stationary point. State averaged CASSCF calculations were performed on Z_m (active space molecular orbitals are given in ESI section†) to explore carefully this region of the PES. CAS(8,7)/6-31G optimization of Z_m-a yielded a biradical minimum energy structure where the N–O and C=N bonds slightly increased by resonance effects (see Fig. 7). Optimization of this structure in S₀ yielded the Z isomer Z_m-a (see geometries in ESI†). Thus, the photoexcitation of Z-a is formally unreactive.

Fig. 7 The photoreaction path of Z_m-a. Z_m-a(S₁) is the biradical minimum structure obtained at the CAS(8,7)/6-31G level in the S₁ state. Cartesian coordinates of Z_m-a(S₁) are given in ESI† section. Hydrogen bond length between N* and H* (see Z in Fig. 1) is also indicated. The N–O and C=N bond lengths increased by 0.022 Å and 0.089 Å in the S₁ state with respect to the ground state structure.

Last, optimization of the Z isomers in T₁ yielded the expected minimum energy structure (δ ∼ 90 deg.; see Fig. 3 and ESI† for geometries). Photoisomerization is thus expected in case of significant intersystem crossing.

In conclusion, only a minor part of the Z isomers (Z-c) undergoes photodegradation, as found in our recent experimental study.¹

The phototransformation of E isomers

S₁ optimization of E-a yielded a minimum excited state structure where the N–O bond was significantly extended (R_N–O(E-a) = 1.473 Å). A relaxed scan of this bond (minimum energy path) was performed and fit in the S₁ state to estimate the energy barrier for N–O dissociation in S₁. It was found that an activation energy of 4.6 kcal mol⁻¹ was necessary to break the N–O bond in the S₁ state (see ESI†).

Optimization of E-b in S₁ gave a minimum energy structure with a geometry similar to E-a. The structure of E-b is roughly obtained by 180 deg. rotation of the N_OxO_Ox–CC_Py dihedral angle in E-a. The NO bond length is large as well (R_N–O(E-b) = 1.491 Å). In addition, electronic configurations of the respective S₁ states were similar for E-a and E-b (i.e. Py → Ph/Ox/Tz electronic transition). Thus similar photoreactivity is expected for E-a and E-b.

Contrarily, excited state optimization of E-c and E-d yielded the CI geometry in Fig. 3. Thus, photoisomerization is the preferred pathway for E-c and E-d, as expected from examination of electronic configurations (Ph/Ox/Tz localized).

All these results provide a coherent picture for the photodegradation of the oxime species investigated in this work and may be summarized as follows: isomers react according to the specificity of their first excited electronic state. If S₁ is characteristic of an electronic transition that spans regions of the molecule that are separated by the oxime bond (here Py and Ph/Ox/Tz), photodegradation is observed. Contrarily, if it is localized on the Ph/Ox/Tz substituents, photoisomerization is the preferred pathway. According to our results, Z-a, Z-b and Z-d undergo photoisomerization, i.e. 86.6% of the Z isomers (see Table 1). Contrarily, E-a and E-b are photodegraded through N–O bond scission, thus 89.8% of the E isomers (see Table 1). This is in excellent agreement with our previously reported experimental results.¹

Full optimization and relaxed N–O scan calculations were also performed to investigate the reactivity in the T₁ state of the most abundant E isomer E-a. The initial geometry was that of the minimum energy structure found in S₁. T₁ state optimization yielded the CI structure (Fig. 3), thus the photoisomerization pathway. Note that T₂ and S₁ are nearly degenerate in the region of the minimum energy structure in S₁, and an intersystem crossing mechanism may involve T₂, i.e. S₁/T₂ isc and T₂/T₁ relaxation. Nevertheless, the relaxed N–O scan in the T₁ state in the direction of increased bond length showed that an excess energy of 3.4 kcal mol⁻¹ (from isc) was sufficient to overcome the N–O dissociation barrier. Thus, photodegradation from the triplet state may not be ruled out. The experimental ratio of the photoisomerization and photodegradation rate constants for the E isomer was k_E→Z/k_{E→Photoproducts} = 6.4 in continuous irradiation conditions.¹ Our theoretical results show that the preferred pathway in S₁ is photodegradation, while photoisomerization (through isc) is favored from T₁. The experimental rate constants indicate slow photodegradation in comparison to isc. However, although photodegradation is a slow process, reactants will inevitably undergo complete photodegradation in continuous irradiation conditions, in agreement with our calculations.

The relationship between structure and excited states energies

It was shown above that photoreactivity was directly related to the electronic configuration of the first singlet excited state. It is interesting to investigate more deeply this relationship in order to design innovant molecular species that are not photodegraded in the open field. Detailed analysis revealed that the relative position of the two highest occupied MOs are key to address this issue, i.e. the bonding π MO of the Py group (named π_Py hereafter) and the bonding π MO of the Ox/Tz/Bz group (named π_Ox hereafter). S₁ and S₂ arise mainly from π_Py → , π_Ox → and π_Py ± π_Ox → transitions. It was shown above that isomerization was favored if S₁ corresponded to the π_Ox → transition. Photodegradation was obtained otherwise. A rigid scan about the dihedral angle δ₃ (OCC_PyN_Py; see Fig. 1) was performed for all the isomers encountered in this work. The respective diabatic states are plotted in Fig. 8. The position of the minimum energy structure is shown in the mesh area. The states π_Py → and π_Ox → were obtained from (1) systematic examination of the key molecular orbitals and (2) the magnitude of the oscillator strength f(S_i=1,2). Small f(S_i) values are characteristic of π_Py → transitions, while large values were obtained for π_Ox → and π_Py ± π_Ox → transitions. Oscillator strengths are given for E-a for illustrative purposes (Fig. 8, top right). Multiple crossings are seen in this figure. Photoisomerization is expected when the diabatic state π_Ox → (blue line in Fig. 8) is highest (photodegradation otherwise). Three important effects can be understood from Fig. 8:

Fig. 8 Rigid scan (rotation of OCCN angle, δ₃) energies of diabatic states Py → Ph/Ox/Tz (red) and Ph/Ox/Tz → Ph/Ox/Tz (blue) for all the isomers investigated in this work (Z-a reacted singularly and is not represented here). Species on the left (resp. right) undergo photoisomerization (resp. photodegradation). The position of the minimum energy structure is represented by the mesh area. Oscillator strengths of S₁ and S₂ are given in the top right figure for E-a. The Newman representation are for Z-d (alignment between the C_Py–N_Py and N–O bonds is reached for δ₃ ∼ +60 deg.). The respective positions are represented by arrows on the graph (bottom left).

(1) the S₁ electronic configuration at the ground state minimum energy structure (mesh area in Fig. 8) is consistent with the above theoretical results for all the conformers: E-a, E-b and Z-c underwent photodegradation (i.e. N–O bond scission; red line is highest in Fig. 8 for the respective geometries), while E-c, E-d, Z-b and Z-d followed the isomerization pathway (blue line is highest).

(2) The π_Ox → diabatic state (blue line) exhibits small changes in comparison to the other state. Thus, it is interesting to understand the origin of the variations of the π_Py → state. First, in all cases, for δ₃ ∼ ±180 deg. (see Newman representations in Fig. 8, far-right and far-left), S₁ and S₂ belong to π_Py − π_Ox → and π_Py + π_Ox → transitions, respectively (π_Py − π_Ox and π_Py + π_Ox are antibonding and bonding, respectively, between Py and the CN bond of the oxime group). In this configuration, the Py plane bisects the N–O bond, thus allowing mixing of the π_Py and π_Ox orbitals. Destabilization of π_Py − π_Ox (S₁), is obtained in this case. A similar effect, although smaller, is also observed for δ₃ ∼ 0 deg. (N_Py pointing toward the N–O bond): the MO is more diffuse on C** (see Fig. 1) than on N_Py (see Fig. SI-8†), and the variation of the respective diabatic state in this region is not as important.

(3) strong destabilization of the other diabatic state is seen for δ₃ ∼ ±60 deg. (red line in Fig. 8). In this geometric configuration, the C_Py–N_Py and N–O bonds parallel each other, with N_Py pointing toward the oxime bond for δ₃ ∼ +60 deg. (see Newman representations in Fig. 8 for Z-d). The π_Ox MO is destabilized by the antibonding contribution of the lone pair on N_Py and the σ(CH₂–C_Py) bond on the one hand, and π_Ox on the other (this MO is shown in Fig. SI-9† for E-a).

In this study, the experimental photoreactivity of oxime species Z was explained on a theoretical basis. Structures where δ₃ ∼ ±180 deg. are bound to undergo photodegradation. This effect can be prevented by addition of a bulky substituent on C** (Py cycle, see Fig. 1) because electronic repulsion between this substituent and the Ox/Ph/Tz part of the molecule favors conformations where −120 < δ₃ < 120 deg. Moreover, conformations where C_Py–N_Py is parallel to the N–O bond (e.g. δ₃ = −69 deg. in E-c, or δ₃ = +66 deg. in E-d) significantly enhances the probability of photoisomerization. This effect is achieved by hydrogen bonding between (1) N_Py and the substituent borne by Tz for the E isomers (methyl in our case) or (2) N_Py and Ph for the Z isomers. In Z-c, the methyl substituent on Tz and the Py group are on opposite sides of the C=N–O–C plane. Contrarily, in Z-b, these substituents are on the same side of the C=N–O–C plane. Hydrogen bonding is stabilizing and Z-b is more stable than Z-c. The addition of a hydrogen bonding substituent on Ph is necessary if significant photodegradation is detected from the Z isomers.

Last, S₁ optimization of Z′ (Fig. 1) followed the photoisomerization pathway to the conical intersection structure CI in Fig. 3. This was expected because the Py group was not present in this species. Optimization of E′ in the S₁ state leaded to the conical intersection (CI in Fig. 3) and photoisomerization.

III. Conclusion

The photoisomerization and photodegradation mechanisms of an important fungicide were investigated in this work from a theoretical point of view in an attempt to explain previous experimental results. Eight conformers were selected from a structural analysis. It was shown that it is the electronic configuration of the first singlet excited state that governed the photoreactivity of this oxime species. Photodegradation was the major pathway for geometries that favored appropriate mixing between the pyridyl and phenyl/terazyl substituents (the respective activation energy in S₁ was smaller than 7 kcal mol⁻¹). Otherwise, photoisomerization was observed. It was shown that both the relative orientation of the pyridinyl group with respect to the N–O bond and the proximity of the nitrogen atom N_Py with the oxime oxygen atom played a key role in the photoreactions investigated here. The Z-a isomer reacted singularly because of intramolecular hydrogen bonding. CASSCF calculations showed that a biradical structure was obtained for Z-a in S₁. Optimization of this species in S₀ yielded the reactant isomer Z-a(S₀). Our calculations explained why the larger part of the Z isomers underwent photoisomerization, while E isomers were mainly photodegraded through N–O bond scission, in agreement with experimental observations. The substitution of a hydrogen atom by a bulky alkyl group at C** on Py (see Fig. 1) will prevent the photodegradation of E isomers.

Investigation of the photoreactivity in the triplet state showed that photodegradation of E isomers may not be ruled out if excess energy released from isc is available to cross the small activation barrier in T₁.

Acknowledgements

This research was supported by the French Ministère de l'Enseignement et de la Recherche and Bayer Crop Science.

References

1. M. Fréneau, P. de Sainte Claire, N. Hoffmann, J.-P. Vors, J. Geist, M. Euvrard and C. Richard, RSC Adv., 2016, 6, 5512–5522.
2. S. Monadjemi, M. El Roz, C. Richard and A. Ter Halle, Environ. Sci. Technol., 2011, 45, 9582–9589.
3. B. Sevilla-Morán, P. Sandín-España, M. J. Vicente-Arana, J. L. Alonso-Prados and J. M. García-Baudín, J. Photochem. Photobiol., A, 2008, 198, 162–168.
4. H. D. Burrows, L. M. Canle, J. A. Santaballa and S. Steenken, J. Photochem. Photobiol., B, 2002, 67, 71–108.
5. A. Padwa and F. Albrecht, J. Am. Chem. Soc., 1974, 96, 4849–4857.
6. A. Padwa, Chem. Rev., 1977, 77, 37–68.
7. A. C. Pratt, Chem. Soc. Rev., 1977, 6, 63–81.
8. H. Roth, in The Chemistry of Hydroxylamines, Oximes and Hydroxamic Acids, ed. Z. Rappoport and J. F. Liebman, John Wiley & Sons, Chichester, 2011, vol. 2, pp.1–63.
9. H. Suginome, in CRC Handbook of Organic Photochemistry and Photobiology, ed. W. Horspool and F. Lenci, CRC Press, Boca Raton, 2nd edn, 2004, pp. 94/1–94/55.
10. M. A. Brown, D. W. Gammon and J. E. Casida, J. Agric. Food Chem., 1983, 31, 1091–1096.
11. D. S. Bohle, Z. Chua, I. Perepichka and K. Rosadiuk, Chem.–Eur. J., 2013, 19, 4223–4229.
12. L. Lapinski, T. Stepanenko, M. F. Rode and M. J. Nowak, J. Mol. Struct., 2010, 976, 181–189.
13. T. Stepanenko, L. Lapinski, M. J. Nowak and L. Adamowicz, Vib. Spectrosc., 2001, 26, 65–82.
14. F. Amat-Guerri, R. Mallavia and R. Sastre, J. Photopolym. Sci. Technol., 1995, 8, 205–232.
15. C. Kolano, G. Bucher, D. Grote, O. Schade and W. Sander, Photochem. Photobiol., 2006, 82, 332–338.
16. 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 C.01,Gaussian, Inc., Wallingford CT, 2010.
17. B. G. Levine and T. J. Martinez, Annu. Rev. Phys. Chem., 2007, 58, 613–634.
18. F. Stunnenberg, R. Rexwinkel and H. Cerfontain, Recl. Trav. Chim. Pays-Bas, 1990, 109, 502–508.
19. J. Lalevée, X. Allonas, J. P. Fouassier, H. Tachi, A. Izumitani, M. Shirai and M. Tsunooka, J. J. Photochem. Photobiol., A, 2002, 151, 27–37.
20. A. Cembran, F. Bernardi, M. Garavelli, L. Gagliardi and G. Orlandi, J. Am. Chem. Soc., 2004, 126, 3234–3243.
21. W.-G. Han, T. Lovell, T. Liu and L. Noodleman, ChemPhysChem, 2002, 3, 167–178.
22. S. A. Kovalenko, A. L. Dobryakov, I. Ioffe and N. P. Ernsting, Chem. Phys. Lett., 2010, 493, 255–258.
23. D. H. Waldeck, Chem. Rev., 1991, 91, 415–436.
24. A. Dargelos, D. Liotard and M. Chaillet, Theor. Chim. Acta, 1975, 38, 79–82.
25. R. Alonso, A. Caballero, P. J. Campos, D. Sampedro and M. A. Rodríguez, Tetrahedron, 2010, 66, 4469–4473.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07712a

---