Mechanistic Investigations on the ... - ACS Publications

Aug 22, 2016 - M(CO)4(CS) (M = Cr, Mo, and W) complexes as model systems. This study ... pyramidal M(CS)(CO)4 complex that has an apical CS group (in...
0 downloads 4 Views 4MB Size
Download PDF
Article pubs.acs.org/IC

Mechanistic Investigations on the Photorearrangement Reactions of M(CO)4(CS) (M = Group 6 Metal) Zheng-Feng Zhang,† Hsu-Cheng Hua,† Shih-Hao Su,† and Ming-Der Su*,†,‡ †

Department of Applied Chemistry, National Chiayi University, Chiayi 60004, Taiwan Department of Medicinal and Applied Chemistry, Kaohsiung Medical University, Kaohsiung 80708, Taiwan

‡

S Supporting Information *

ABSTRACT: The mechanisms for the photoisomerization reactions are studied theoretically at the M06-2X/Def2-TZVPD level of theory, using the five-coordinated M(CO)4(CS) (M = Cr, Mo, and W) complexes as model systems. This study provides the first theoretical evidence for the mechanisms of these photorearrangements of the five coordinated metal complexes. That is, the photoisomerization process is primary the axial to basal movement of CS. The model study demonstrates that the preferred reaction route for the photorearrangement reactions is as follows: reactant → Franck−Condon region → minimum (triplet) → transition state (triplet) → triplet/singlet intersystem crossing → photoproduct. The theoretical results also show that the energy differences between the crucial points are quite small, which demonstrates that the CS group rotates easily to form the different conformations when the M(CO)4(CS) molecules have been photoirradiated. These photochemical mechanisms are consistent with the available experimental observations.

I. INTRODUCTION The five-coordinated group VIB transition metal carbonyl complexes, M(CO)5 (M = W, Cr, Mo), which are produced by the photodissociation of one carbonyl group from a parent hexacarbonyl complex M(CO)6 and its derivatives, play an essential role in the photochemistry of M(CO)6 complexes.1,2 It was experimentally reported by Poliakoff that the irradiation of W(CS)(CO)4 results in the exchange of apical and basal CS groups via a mechanism that does not involve photodissociation (see Scheme 1).3 However, since this discovery 40 years ago, Scheme 1

II. GENERAL CONSIDERATION Although the photorearrangement reactions for the fivecoordinated W(CS)(CO)4 complex that have been reported experimentally3 show that there are various types of reactions, there is a specific uniformity in these photoreactions that at least supports the following discussion. A schematic representation of the electronic behavior of the valence molecular orbitals (MOs) of the M(CS)(CO)4 complex4 that would lead to an intersystem crossing is given in Figure 1. Figure 1 shows that when a square pyramidal M(CS)(CO)4 complex that has an apical CS group (in C4v symmetry) absorbs a photon, one electron in the M(CS)(CO)4 complex is promoted to an excited state (a1) from the occupied orbitals (e + b2). This results in structural instability because there are Jahn−Teller distortions.5 As a result, M(CS)(CO)4 distorts to the trigonal bipyramidal structure, from which an excited electron falls to the occupied orbitals from where it originated. It may then be finally possible to obtain a

neither experimental nor theoretical mechanisms for the photochemical isomerization reactions of the five-coordinated M(CS) (CO)4 molecules have been reported, except for a simple qualitative theoretical analysis that was based on the extended Hückel method.4 The lack of theoretical studies may be because computational methods for spin−orbit coupling that involves transition metal elements have not been available until recently. This study gives an insight into the seemingly complicated photochemical rearrangement mechanisms for M(CS) (CO)4 (M = group VIB transition metal), using a more sophisticated theoretical method (see eqs 1 and 2). It is hoped that any subsequent study of the mechanisms for the photoreactions of these five-coordinated transition metal complexes will lead to the development of related synthesis and catalytic processes.(M = Cr, Mo, and W) © XXXX American Chemical Society

Received: June 23, 2016

A

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. Electronic behavior of the valence molecular orbitals of the M(CS)(CO)4 (M = Cr, Mo, and W) complex in C4v symmetry during the photochemical process.

M06-2X/Def2-TZVPD level of theory, to allow a comparison with the available experimental results. Figure 3a,b shows the relative energies of the key points with respect to the corresponding ground-state minima, Rea-A-S0-W and Rea−BS0-W, in which the CS group of the W(CS)(CO)4 complex is on the axial (A) and basal (B) positions, respectively. The important geometrical parameters of these points are shown in Figure 4. It has to be mentioned here that the present computations are necessarily accomplished in the gas phase. It was, however, experimentally reported by Poliakoff that the photochemical rearrangement reactions of W(CO)4(CS) proceeded in Ar and CH4 matrices at 20 K conditions,3 which are unusual nowadays. Moreover, the reviewer indicated that this is quite important because the thermal energy of the molecules is unusually so small that it can be ignored. This strongly implies that there is no thermal energy to allow solvent movement. In this work, as mentioned earlier, we thus performed the first theoretical computations on such photochemical rearrangement reactions, which were based on the gas phase. As a result, although the potential energy surfaces obtained in this work are still semiquantitative, they at least give us a clear photochemical reaction mechanism. Besides this, it has to be emphasized that both previous experimental work3 and the present theoretical study (vide infra) all demonstrate the photochemical isomerization reaction mechanisms for W(CO)4(CS) are indeed topologically equivalent to the Berry pseudorotation.2 First, the photochemical mechanisms for the tungsten element that are represented by eqs 1 and 2 are studied theoretically. The TD-DFT/M06-2X/Def2-TZVPD computational data indicates that the relative energies (kcal/mol) of the electronic states are as follows (see Supporting Information): S0 (0.0) < T1 (45.3) < T2 (62.3) < S1 (92.5) and S0 (0.0) < T1 (49.4) < S1 (60.3) < T2 (80.1), for W(CS)(CO)4 in A and B conformations with squarepyramidal geometry and with the CS group in the axial and basal positions, respectively. As shown in Figure 3, our computations demonstrate that the transition from the singlet ground state to the triplet excited state occurs to start the reaction from the triplet excited Franck−Condon region, which is a spin-forbidden excitation. In fact, it is well-established that spin-allowed absorption cross sections are mostly larger than those for spinforbidden excitations. Therefore, such a spin-forbidden excitation might be true only if the TD-DFT results exactly agree with the experimental observations. As a result, one

square pyramidal M(CS)(CO)4 complex, but one that features a basal CS substituent. The entire structural motion is considered to be a photochemical Berry process.2 The principal feature of the photochemical mechanisms for the five-coordinated M(CS)(CO)4 complexes is the location of the spin−orbit coupling in the excited- and ground-electronic states. In this study, the frontier MO model is used to search for the intersystem crossings for the photoisomerization of the M(CS)(CO)4 molecules, as outlined in Figure 1. Figure 2 shows the qualitative potential energy surfaces for the S0 and T1 states of the M(CS)(CO)4 complexes due to the bending of two basal CO ligands (i.e., bent at angle θ).6 By bending the ∠CMC angle in M(CS)(CO)4, the energy of its singlet state (S0) curve is raised in energy and the energy of its triplet curve (T1) is decreased because of the respective increase and decrease in antibonding interactions. Consequently, these states become degenerate at a geometry of approximately 157°, 152°, and 155°, for Cr(CS)(CO)4 (1), Mo(CS)(CO)4 (2), and W(CS)(CO)4 (3), respectively, as shown in Figure 2. In other words, this photoexcitation removes the barrier in order to bend around the former ∠CMC angle. The formation of this degenerate point may explain an enhanced intramolecular bend in this square pyramidal geometry, possibly indicating the existence of an intersystem crossing, where decay to the singlet ground state is entirely effective. These results are used in the following discussion to clarify the mechanisms for the photorearrangement reactions for the M(CS)(CO)4 complexes.

III. RESULTS AND DISCUSSION All density functional theory (DFT) calculations are performed using the GAUSSIAN 09 package of programs7 at the M06-2X/ Def2-TZVPD level of theory.8,9 Time-dependent density functional theory (TD-DFT) computations are also performed using this density functional theory10−12 and basis sets. The minimum energy crossing points between the singlet and triplet potential energy surfaces are also calculated using the GAUSSIAN 09 package, together with the code that was developed by Harvey et al.13 For more information see Supporting Information. III.1. Mechanisms for the Photoisomerization Reactions of W(CO)4(CS). Since the photochemical isomerization reactions of W(CO)4(CS) have been extensively studied experimentally,3 the reaction mechanisms are studied using the B

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 2. Minimum-energy pathway for a spin−orbit coupling in the M(CS)(CO)4 reactant [(a) 1, (b) 2, and (c) 3] along the bending angle (θ). The coordinates are optimized for the S0 and T1 states at the M062X/Def2-TZVPD level of theory. For more details, see the text.

reviewer indicated that the mechanism of initial excitation of a higher singlet state (such as S1 or S2) that relaxes to the lower triplet is possible. It has to be mentioned that our TD-DFT results show that the energies of two triplet states (T1 and T2) are lower than that of singlet first excited state (S1) as given in Figure 3. This means that even photoexcitation promotes W(CS)(CO)4 to a singlet excited state (such as S1 or S2); this species should subsequently relax, branching between the T1 and S0 states. As one can see, the discussions of intersystem crossings among the singlet and the triplet excited states could be too complicated to understand the photochemical mechanisms of W(CS)(CO)4. As a result, such discussions are beyond the scope of the present work and will not be offered in this study. The other reason that we focus on the triplet energy surfaces as well as the spin

Figure 3. Potential energy surfaces for the photochemical isomerization models for (a) A-S0-W and (b) B−S0-W. A and B, respectively, denote the positioning of the CS group of the W(CS)(CO)4 complex on the axial and basal positions. The abbreviations, FC, Min, TS, and Pro, respectively, denote Frank−Condon, intermediate, transition state, and product. The M06-2X/Def2-TZVPD optimized structures of the crucial points are shown in Figure 4. For more information see the text.

crossovers is due to the valence MOs’ analysis given in Figure 1, which was predicted to start on the triplet state. With the above reasons taken together and for simplicity, the present considerations concentrate only on the triplet state energy surface. C

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

demonstrate that the triplet Min-A-T1-W complex is 17.0 kcal/ mol above the corresponding singlet state species. A comparison of the geometrical structures of Min-A-T1-W with those of the corresponding singlet ground state, Rea-A-S0-W (Figure 4), shows that the triplet state exhibits significantly longer bonds and greater bond angles than the corresponding closed shell singlet state. The available experimental observations3 show that the photoisomerization of Rea-A-S0-W is nonadiabatic, which strongly implies that the photochemical reaction starts from the triplet surface and ultimately proceeds along the singlet ground-state pathway. Therefore, the intersystem crossing point for the T1 and S0 surfaces must play a decisive role in the mechanistic photorearrangement reactions of Rea-A-S0-W and Rea−B-S0-W (see Figure 3). From the Rea-A-S0-W perspective, the W(CS)(CO)4 complex can undergo two W−C−O axial bends via a transition state, TSA-T1-W. In terms of the geometrical structures that are shown in Figure 4, it is obvious that this transition state connects the triplet local minimum (Min-A-T1-W) to the corresponding intersystem crossing points (T1/S0-A-W). The M06-2X calculations demonstrate that the generation of T1/S0-A-W from Min-AT1-W progresses via a transition state, TS-A-T1-W, with a barrier of 6.8 kcal/mol. The computational data shows that the fivecoordinated W(CS)(CO)4 complex has sufficient internal energy (28.3 kcal/mol) to overcome the barrier height between the Min-A-T1-W minimum and the T1/S0-A-W intersection point after photoexcitation to the triplet state (T1). Figure 4 shows that the geometry of the T1/S0-A-W crossing point has a pseudo-trigonal-bipyramid structure, which is consistent with the earlier predictions that are shown in Figures 1 and 2.4 That is, the triplet (T1) and singlet (S0) surfaces intersect, and the five-coordinated W(CS) (CO)4 molecule that features a square pyramidal geometry distorts to a trigonal bipyramidal structure, in which the CS group lies on the trigonal plane from the former axial site.2 The M06-2X computations show that the T1/S0-A-W intersection point is located 4.7 kcal/ mol below the TS-A-T1-W points and 19.1 kcal/mol above the corresponding singlet ground-state reactants (Rea-A-S0-W). Finally, from this intersystem crossing point (T1/S0-A-W), the five-coordinated W(CS)(CO)4 compound returns to the singlet square-pyramidal geometry with the CS group in the basal site (i.e., Pro-A-S0-W or Rea−B-S0-W). On the basis of these theoretical studies, the photochemical isomerization reactions of Rea-A-S0-W (i.e., eq W-1) are summarized as follows: Rea‐A‐S0‐W + hν → FC‐A‐T1‐W → Min‐A‐T1‐W → TS‐A‐T1‐W → T1/S0‐A‐W → Pro‐A‐S0‐W (Rea‐B‐S0‐W)

(W-1)

The photochemical rearrangement process for eq W-2 is determined theoretically, as is the mechanistic analysis for eq W1. As shown in Figure 3b, these theoretical computations suggest that the reaction mechanism for Rea−B-S0-W proceeds as follows:

Figure 4. M06-2X/Def2-TZVPD geometries (in Å and deg) for W(CO)4(CS). This includes the singlet reactant (Rea-A-S0-W; Rea−BS0-W), the triplet minimum (Min-A-T1-W; Min-B-T1-W), the triplet transition state (TS-A-TS-W; TS-B-TS-W), and the intersystem crossing (A-T1/S0−W; B-T1/S0−W). For more information see Supporting Information.

Rea‐B‐S0‐W + hν → FC‐B‐T1‐W → Min‐B‐T1‐W → TS‐B‐T1‐W → T1/S0‐B‐W

As seen in Figure 3, the five-coordinated W(CS)(CO)4 (ReaA-S0-W) complex is initially irradiated to its lowest lying triplet excited state (FC-A-T1-W). It then relaxes to its local minimum (Min-A-T1-W) on the triplet surface (T1), which is close to the corresponding geometry on the ground state (S0). The M06-2X/ Def2-TZVPD computations that are shown in Figure 3

→ Pro‐B‐S0‐W (Rea‐A‐S0‐W)

(W-2)

When the W(CS)(CO)4 complex wherein the CS group is located on the basal position (Rea−B-S0-W) absorbs a photon, it is promoted to its excited triplet state by a vertical excitation (FCD

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

have a square-pyramidal geometry wherein the CS group is in either the axial (A) or the basal (B) positions. Figure 6 shows all of the potential energy surfaces of the reactants (Rea-A-S0-Cr or

B-T1-W). After the vertical excitation process, Rea−B-S0-W relaxes to the excited triplet (T1) surface, which is termed Min-BT1-W. The local minimum (Min-B-T1-W) is calculated to be 16.3 kcal/mol higher than the singlet Rea−B-S0-W. Using the local minimum, Min-B-T1-W, a transition state search for the CS group rearrangement is undertaken, using the model of the Min-B-T1-W conformation. The theoretical data shows that the transition state (TS-B-T1-W) lies 6.6 kcal/mol above the corresponding Min-B-T1-W point. The M06-2X computations also show that the TS-B-T1-W species is 4.6 kcal/ mol above the corresponding T1/S0−B-W. Since the energy release when FC-B-T1-W relaxes to Min-B-T1-W is computed to be 33.1 kcal/mol, the barrier from Min-B-T1-W to T1/S0−B-W can be readily overcome. Finally, similarly to the Rea-A-S0-W case study (eq W-1), relaxing through T1/S0−B-W easily results in a five-coordinated W(CS)(CO)4 (Pro−B-S0-W or Rea-A-S0W) complex wherein the CS group is located on the axial site. It is noteworthy that the spin density values for the triplet intermediates of W(CS)(CO)4 (1.92 for Min-A-T1-W and 1.86 for Min-B-T1-W) that are shown in Figure 5 indicate that both species have two unpaired electrons that are located on the tungsten center.

Figure 5. Spin density plots of the triplet states for the Min-A-T1-W and Min-A-T2-W complexes based on the M06-2X/Def2-TZVPD level of theory.

In short, these theoretical findings demonstrate that the M062X results for the photoisomerization reactions for two kinds of isomers of the five-coordinated W(CS)(CO)4 complex, in which the CS group is located at either the axial or the basal position in the square pyramidal geometry, provide a good theoretical explanation of the experimental observations.3 III.2. Mechanisms for the Photoisomerization Reactions of Cr(CO)4(CS). Bearing these theoretical analyses of the tungsten case in mind, the mechanisms for the photoisomerization reactions of an analogous complex, Cr(CO)4(CS), are studied. This chromium complex has two types of isomers that

Figure 6. Potential energy surfaces for the photochemical isomerization models for (a) A-S0-Cr and (b) B−S0-Cr. A and B, respectively, denote the respective positioning of the CS group of the Cr(CS)(CO)4 complex in the axial and basal position. The abbreviations, FC, Min, TS, and Pro, respectively, denote Frank−Condon, intermediate, transition state, and product. The M06-2X/Def2-TZVPD optimized structures for the crucial points are shown in Figure 7. For more information, see the text. E

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Rea−B-S0-Cr), the intermediates (Min-A-S0-Cr or Min−B-S0Cr), the transition states (TS-A-S0-Cr or TS−B-S0-Cr), and the intersystem crossings (T1/S0-A-Cr or T1/S0−B-Cr) that connect the reactants and the final photoproducts (Pro-A-S0Cr or Pro−B-S0-Cr). The specific optimized geometrical parameters for the critical points of eq Cr-1 and eq Cr-2, based on M06-2X/Def2-TZVPD calculations, are shown in Figure 7. These TD-DFT/M06-2X/Def2-TZVPD calculations for both isomers of Cr(CO)4(CS), Rea-A-S0-Cr and Rea−B-S0-Cr, show that their first excited state is the triplet state and the relative energy (kcal/mol) increases in the following order: S0 (0.0) < T1 (53.4) < T2 (81.9) < T3 (85.8) < S1 (98.1) and S0 (0.0) < T1 (53.1) < T2 (58.7) < T3 (60.6) < T4 (68.0) < S1 (98.7), respectively (Supporting Information). This explanation for the Cr(CO)4(CS) complexes with a square pyramidal geometry, which is based on the TD-DFT computations, is consistent with the earlier prediction using the extended-Hückel method, as shown in Figure 1.4 It is predicted that the intersystem crossing mechanism has a prominent role in the photorearrangement reactions of the five-coordinated Cr(CO)4(CS) molecules. Similar to the computational results for the W(CO)4(CS) case, as seen in Figure 6 a,b, the M06-2X studies suggest that the mechanisms for the photochemical rearrangement reactions for Cr(CO)4(CS) proceed as follows: Rea‐A‐S0‐Cr + hν → FC‐A‐T1‐Cr → Min‐A‐T1‐Cr → TS‐A‐T1‐Cr → T1/S0‐A‐Cr → Pro‐A‐S0‐Cr (Rea‐B‐S0‐Cr)

(Cr-1)

Rea‐B‐S0‐Cr + hν → FC‐B‐T1‐Cr → Min‐B‐T1‐Cr → TS‐B‐T1‐Cr → T1/S0‐B‐Cr → Pro‐B‐S0‐Cr (Rea‐A‐S0‐Cr)

(Cr-2)

Figure 6a (eq Cr-1) shows that because there is a large excess energy of 49.9 kcal/mol due to the decay from FC-A-T1-Cr to Min-A-T1-Cr, the barrier of 2.6 kcal/mol for Min-B-T1-Cr to TS-B-T1-Cr can be easily surmounted. Figure 6b (eq Cr-2) shows that because a large excess of energy (51.1 kcal/mol) is produced by the decay from FC-B-T1-Cr to Min-B-T1-Cr, this relaxation energy is sufficient to provoke the photorearrangement reaction. These theoretical observations show that when it is irradiated with visible light (λ > 420 nm), the five-coordinated Cr complex, Rea-A-S0-Cr, which features a square pyramidal geometry wherein the CS substituent is situated on the axial point, rearranges to form the isomer, Rea−B-S0-Cr, wherein the CS group is located on the basal position, and vice versa. Similarly to the case for W(CS)(CO)4 studied earlier, the spin densities for Min-A-T1-Cr and Min-B-T1-Cr are computed to be 2.93 and 2.88, respectively. See Figure 8. These values strongly indicate that two unpaired electrons are situated on the chromium center for each triplet minimum. No experimental results have been reported for the photochemical isomerization reactions of the Cr(CO)4(CS) complex so the above theoretical conclusions are predictions for future experimental studies. III.3. Mechanisms for the Photoisomerization Reactions of Mo(CO)4(CS). Finally, the same level of theory (M062X/Def2-TZVPD) is used to study the photoisomerization mechanisms for the five-coordinated molecule, Mo(CO)4(CS). Similar to the W and Cr complexes that have been studied, Figure 9a,b shows the respective photochemical reaction paths for eqs

Figure 7. M06-2X/Def2-TZVPD geometries (in Å and deg) of Cr(CO)4(CS). This includes the singlet reactant (Rea-A-S0-Cr; Rea− B-S0-Cr), the triplet minimum (Min-A-T1-Cr; Min-B-T1-Cr), the triplet transition state (TS-A-TS-Cr; TS-B-TS-Cr), and the intersystem crossing (A-T1/S0−Cr; B-T1/S0−Cr). For more information, see Supporting Information.

Mo-1 and Mo-2. The specific geometrical parameters are shown in Figure 10. F

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 8. Spin density plots of the triplet states for the Min-A-T1-Cr and Min-A-T2-Cr complexes based on the M06-2X/Def2-TZVPD level of theory.

The TD-DFT/M06-2X/Def2-TZVPD method is applied to both geometrical isomers of Mo(CO)4(CS), Rea-A-S0-Mo and Rea−B-S0-Mo, and the results are shown on the left-hand of Figure 9a,b, respectively. The computed order for the relative energy (kcal/mol) is S0 (0.0) < T1 (53.4) < T2 (81.9) < T3 (85.8) < S1 (98.1) and S0 (0.0) < T1 (53.1) < T2 (58.7) < T3 (60.6) < T4 (68.0) < S1 (98.7) for Rea-A-S0-Mo and Rea−B-S0-Mo, respectively. Similarly to the W(CO)4(CS) and Cr(CO)4(CS) systems that have been studied, the intersystem crossing mechanisms for the Mo(CO)4(CS) are the focus of this study. Again, the spin densities for Min-A-T1-Mo and Min-B-T1-Mo are calculated at the M06-2X/Def2-TZVPD level of theory. As seen in Figure 11, the spin densities are estimated to be 1.99 and 1.93, respectively. Therefore, these computations confirm that the two triplet intermediates each have two unpaired electrons in the molybdenum nucleus. It has to be noted that the spin densities for the chromium complexes shown in Figure 8 (ca. 2.9) were calculated to be larger than those for the molybdenum (ca. 2.0; Figure 11) or tungsten (ca. 1.9; Figure 5) complexes. However, it was already stated by several studies that the spin− orbit coupling for tungsten is larger than those for chromium and molybdenum, which can assist the direct population of triplet states.14 The computational results that are shown in Figure 9 show that the photoisomerization reactions of the five-coordinated molybdenum complexes should proceed as follows:

Figure 9. Potential energy surfaces for the photochemical isomerization models for (a) A-S0-Mo and (b) B−S0-Mo. A and B denote the respective positioning of the CS group of the Mo(CS)(CO)4 complex on the axial and basal positions. The abbreviations, FC, Min, T, and Pro, respectively, denote Frank−Condon, intermediate, transition state, and product. The M06-2X/Def2-TZVPD optimized structures for the crucial points are shown in Figure 8. For more information, see the text.

Rea‐B‐S0‐Mo + hν → FC‐B‐T1‐Mo → Min‐B‐T1‐Mo → TS‐B‐T1‐Mo → T1/S0‐B‐Mo

Rea‐A‐S0‐Mo + hν → FC‐A‐T1‐Mo

→ Pro‐B‐S0‐Mo (Rea‐A‐S0‐Mo)

→ Min‐A‐T1‐Mo → TS‐A‐T1‐Mo → T1/S0‐A‐Mo → Pro‐A‐S0‐Mo (Rea‐B‐S0‐Mo)

(Mo-2)

The M06-2X calculations show that the energies of the intersystem crossing points (T1/S0-A-Mo and T1/S0−B-Mo),

(Mo-1) G

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 11. Spin density plots of the triplet states for the Min-A-T1-Mo and Min-A-T2-Mo complexes based on the M06-2X/Def2-TZVPD level of theory.

kcal/mol for Min-B-T1-Mo → TS-B-T1-Mo. Because there is a large surplus of energy of 31.1 and 32.3 kcal/mol, from the FC points to the local minima, these smaller barriers that are shown in Figure 9a,b are easily overcome. Therefore, the photoisomerization reactions for both of the geometrical isomers of Mo(CO)4(CS) should progress without any difficulty. Again, to the best of our knowledge, until now no related experimental results have been found for the photochemical isomerization reactions of the Mo(CO)4(CS) complex. As a result, the above computational conclusions are predictions for future experimental studies.

IV. CONCLUSIONS These theoretical studies demonstrate that all of the fivecoordinated M(CS)(CO)4 (M = Cr, Mo and W) complexes, including both the Rea-A-S0-M and the Rea−B-S0-M conformations, follow the photochemical Berry routes to generate the final products (Pro-A-S0-M and Pro−B-S0-M) via a fast triplet/singlet intersystem crossing (T1/S0-A-M and T1/S0-B-M, respectively) along a radiation-less pathway. The M06-2X data that is shown in Figure 3 for both the Rea-A-S0-W and Rea−BS0-W cases is consistent with the available experimental results,3 which demonstrates that the photoisomerization of fivecoordinated W(CS)(CO)4 initially occurs on the triplet surface. Before relaxation to the singlet surface, W(CS)(CO)4 still has sufficient internal energy to overcome all of the energy barriers on the triplet surface. The M06-2X computational results also show that the differences in energy between the crucial points of the photorearrangement reactions of the Cr(CS)(CO)4 and Mo(CS)(CO)4 complexes are small, as shown in Figures 6 and 9. This strongly supports the theory that the photochemical isomerization reactions of the other five-coordinated (i.e., Cr(CS)(CO)4 and Mo(CS)(CO)4) molecules should occur without any difficulty, when these compounds are photoirradiated.

Figure 10. M06-2X/Def2-TZVPD geometries (in Å and deg) of Mo(CO)4(CS). This includes the singlet reactant (Rea-A-S0-Mo; Rea− B-S0-Mo), the triplet minimum (Min-A-T1-Mo; Min-B-T1-Mo), the triplet transition state (TS-A-TS-Mo; TS-B-TS-Mo), and the intersystem crossing (A-T1/S0−Mo; B-T1/S0−Mo). For more information, see Supporting Information.

relative to the ground-state minima (Rea-A-S0-Mo and Rea−BS0-Mo), are 23.7 and 18.4 kcal/mol. These values are, respectively, 29.4 and 30.2 kcal/mol lower than the corresponding FC points (FC-A-T1-Mo and FC-B-T1-Mo). From the local minima (Min-A-T1-Mo and Min-B-T1-Mo), the rearrangement reactions also encounter small barriers, which are computed to be about 8.3 kcal/mol for Min-A-T1-Mo → TS-A-T1-Mo and 11.0 H

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(7) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. GAUSSIAN 09; Gaussian, Inc.: Wallingford, CT, 2013. (8) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157. (9) For the Def2-TZVPD basis sets, see: (a) Andrae, D.; Haeussermann, U.; Stoll, M. H.; Preuss, H.; Dolg, M. Theor. Chim. Acta 1990, 77, 123. (b) Metz, B.; Stoll, H.; Dolg, M. J. Chem. Phys. 2000, 113, 2563. (c) Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. J. Chem. Phys. 2003, 119, 11113. (10) (a) Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J. J. Chem. Phys. 1998, 109, 8218. (b) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J. Chem. Phys. 1998, 108, 4439. (11) The reviewers suggested that the CASSCF level of theory should be a good choice to study the photochemical rearrangement mechanisms of eqs 1 and 2 . Since we considered the computer memories as well as the disk space utilized in this work and also wanted to demonstrate the spin states adopted in the present photochemical reactions, we thus chose a more convenient method (M06-2X) to compute the whole potential energy surfaces of photochemical reactions. Besides this, according to our previous computational experiences (ref 12) and the calculated data given in Supporting Information, compared with the available experimental works (ref 3), we found that the M06 functional should be reliable in the present study. (12) (a) Su, S.-H.; Su, M.-D. RSC Adv. 2016, 6, 50825. (b) Su, M.-D. ChemistrySelect 2016, 1, 1588. (c) Su, S.-H.; Su, M.-D. Phys. Chem. Chem. Phys. 2016, 18, 16396. (13) (a) Harvey, J. N.; Aschi, M.; Schwarz, H.; Koch, W. Theor. Chem. Acc. 1998, 99, 95. (b) Harvey, J. N.; Aschi, M. Phys. Chem. Chem. Phys. 1999, 1, 5555. (14) (a) Lees, A. J. Chem. Rev. 1987, 87, 711. (b) Wrighton, M.; Hammond, G. S.; Gray, H. B. J. Am. Chem. Soc. 1971, 93, 4336. (c) Cole, G. M.; Garrett, B. B. Inorg. Chem. 1970, 9, 1898. (15) One reviewer suggested that picosecond time-resolved infrared spectroscopy should be a good choice to confirm the predictions shown in this work.

The authors await experimental results that confirm these predictions.15

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b01500. Theoretical methods, references, and additional tables of data (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are grateful to the National Center for HighPerformance Computing of Taiwan for generous amounts of computing time, and the Ministry of Science and Technology of Taiwan for the financial support. One of the authors (M.-D.S.) also wishes to thank Professor Michael A. Robb, Dr. Michael J. Bearpark, Dr. S. Wilsey, (University of London, UK), and Professor Massimo Olivucci (Universita degli Studi di Siena, Italy) for their encouragement and support during his stay in London. Special thanks are also due to reviewers 1, 2, and 3 for very helpful suggestions and comments.

■

REFERENCES

(1) (a) Wrighton, M. Chem. Rev. 1974, 74, 401. (b) Vogler, A. In Concepts in Inorganic Photochemistry; Adamson, A. W., Fleischauer, P. D., Eds.; Welly: New York, 1975. (c) Kuendig, E. P.; Ozin, G. A. J. Am. Chem. Soc. 1974, 96, 3820. (d) Huber, H.; Kundig, E. P.; Ozin, G. A.; Poe, A. J. J. Am. Chem. Soc. 1975, 97, 308. (e) Perutz, R. N.; Turner, J. J. J. Am. Chem. Soc. 1975, 97, 4800. (f) Burdett, J. K.; Perutz, R. N.; Poliakoff, M.; Turner, J. J. J. Chem. Soc., Chem. Commun. 1975, 157. (g) Barnes, D. S.; Pilcher, G.; Pittam, D. A.; Skinner, H. A.; Todd, D. J. Less-Common Met. 1974, 38, 53. (h) Burdett, J. K.; Grzybowski, J. M.; Perutz, R. N.; Poliakoff, M.; Turner, J. J.; Turner, R. F. Inorg. Chem. 1978, 17, 147. (i) Simpson, M. B.; Poliakoff, M.; Turner, J. J.; Maier, W. B.; McLaughlin, J. G. J. Chem. Soc., Chem. Commun. 1983, 1355. (j) Maier, W. B.; Poliakoff, M.; Simpson, M. B.; Turner, J. J. J. Chem. Soc., Chem. Commun. 1980, 587. (k) Upmacis, K.; Gadd, G. E.; Poliakoff, M.; Simpson, M. B.; Turner, J. J.; Whyman, R.; Simpson, A. F. J. Chem. Soc., Chem. Commun. 1985, 27. (l) Shanoski, J. E.; Payne, C. K.; Kling, M. F.; Glascoe, E. A.; Harris, C. B. Organometallics 2005, 24, 1852. (m) Shanoski, J. E.; Glascoe, E. A.; Harris, C. B. J. Phys. Chem. B 2006, 110, 996. (2) Hay, P. J. J. Am. Chem. Soc. 1978, 100, 2411. (3) (a) Poliakoff, M. Inorg. Chem. 1976, 15, 2022. (b) Poliakoff, M. Inorg. Chem. 1976, 15, 2892. (4) Albright, T. A.; Burdett, J. K.; Whangbo, M. H. In Orbital Interaction in Chemistry; Wiley: New York, 1985; p 318−319. (5) Gillespie, R. J. In Molecular Geometry; Van Nostrand-Reinhold: London, 1972. (6) For comparisons, the M−C, C−O, and C−S bonds in the M(CS) (CO)4 complexes shown in Figure 2 are fixed to be 1.35, 1.30, 1.09, 1.30, and 1.30 Å, respectively. Also, the ∠CMC(apcial), ∠MCO, and ∠MCS bond angles are fixed to be 90°, 180°, and 180°, respectively. These bending angles were obtained without full optimizations of the reactants. Nevertheless, they at least give us a hint that degeneracy between the singlet and triplet can exist as a result of the bend of a ∠CMC angle. I

DOI: 10.1021/acs.inorgchem.6b01500 Inorg. Chem. XXXX, XXX, XXX−XXX