Electronic Structure and Reactivity - ACS Publications - American

The logarithms of the equilibrium constants K for. X = H, MeO, Br, logK, were plotted against a solvent parameter. ET(30)45. As shown in Figure 3c, al...
1 downloads 0 Views 1MB Size
Subscriber access provided by TULANE UNIVERSITY

Article

1,2-Diazacyclopentane-3,5-diyl Diradicals: Electronic Structure and Reactivity Shohei Yoshidomi, and Manabu Abe J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b12254 • Publication Date (Web): 15 Feb 2019 Downloaded from http://pubs.acs.org on February 15, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

1,2-Diazacyclopentane-3,5-diyl Diradicals: Electronic Structure and Reactivity Shohei Yoshidomi† and Manabu Abe*,†,‡,§ †Department

of Chemistry, Graduate School of Science, Hiroshima University (HIRODAI), 1-3-1 Kagamiyama, HigashiHiroshima, Hiroshima 739-8526, Japan. ‡Hiroshima University Research Center for Photo-Drug-Delivery Systems (HiU-P-DDS) §JST-CREST, K’s Gobancho 6F, 7, Gobancho, Chiyoda-ku, Tokyo 102-0075, Japan *E-mail: [email protected] ABSTRACT: Localized singlet diradicals are key intermediates in bond-homolyses. A thorough study of the reactive species is needed to clarify the mechanisms of the homolytic bond-cleavage and formation processes. In general, the singlet diradicals are quite shortlived due to the fast radical–radical coupling reactions. The short-lived characteristic has retarded the thorough study on bondhomolyses. In this study, a new series of long-lived singlet diradicals, viz. 1,2-diazacyclopentane-3,5-diyl, were identified, and their electronic structures and novel reactivities were thoroughly studied using laser-flash photolysis (LFP), product analysis, and computational studies. A direct observation of the thermal equilibration (fast process) between the singlet diradicals and the corresponding ring-closing compounds was undertaken at the sub-microsecond time scale. The solvent and substituent effect on the equilibration constant and rate constants for the ring-closing reaction and ring-opening reaction clarify the novel nitrogen-atom effect on the localized singlet 1,3-diyl diradicals. Two types of alkoxy-migrated compounds, 9 and 10, were isolated with high yields as final products. Crossover, spin-trapping, and LFP experiments for the formation of alkoxy-group migration products, i.e. 9 versus 10, revealed the unique temperature effect on the product ratio of the two types of alkoxy-migration products. The temperature-insensitive intersystem crossing process (slow process, millisecond time scale) was found to be a key step in the formation of 9, which is an entropy-controlled pathway. An intramolecular migration process was identified for the formation of 10 that was accelerated by a polar solvent, in an enthalpy-controlled process. This unique heteroatom effect has opened up a new series of localized singlet diradicals that are crucial intermediates in bond-homolysis.

INTRODUCTION Chemistry involves bond-breaking and bond-formation processes. Localized singlet diradicals are crucial intermediates in bond homolyses (Chart 1).1–5 Thorough studies of the structure and reactivity of localized singlet diradicals are essential for our understanding of the mechanism of bondhomolysis processes and the related chemistry, which could then be applied to the field of materials chemistry, organic synthesis, and biological chemistry. This has already been established for the cations and anions that are key intermediates in bondheterolysis.6–8 The lifetime of singlet diradicals is, however, quite short due to the fast radical–radical coupling reaction. Thus, it is usually assumed that the experimental elucidation of the structure and reactivity of such intermediates would be very difficult.9–11 Over the last decade, significant progress has been achieved by the application of appropriate molecular design to the generation of detectable localized singlet diradicals. Based on the findings, the direct observation of localized singlet diradicals has become possible, enabling the experimental investigation of carbon-carbon bond homolysis processes. Borden and ourselves used molecular strains that kinetically stabilize the singlet 1,3diradicals to understand the electronic structures and reactivity.12,13 The studies clarified that there are two types of localized singlet diradicals, 1: type-I and type-II.14–21 The disrotatory ring-closing mode is energetically favored for type-I

molecules having the electron-withdrawing group (X) substituents to give bicyclo[2.1.0]pentane derivatives. The typeII molecules need high energy for the bond formation reaction because the conrotatory ring-closing mode produces corresponding trans-fused bicyclic compounds, in which the electron-donating groups X are substituted at the C2 position. Thus, the type-I molecules are suitable targets for investigating the radical-radical coupling processes. Indeed, the singlet diradicals, 2, generated by clean photochemical denitrogenation reactions, have been directly observed in the sub-microsecond time scale to elucidate these features. The electron-donating group substituted diradical, 2 (Ar = p-MeOC6H4), was found to be longer-lived than 2 having electron-withdrawing substituents (Ar = p-CNC6H4),22,23 although the radical stabilization ability of the p-CN group is much higher than that of the p-MeO group.24 Thus, 2 has the same characteristics of zwitterion ZI due to hyperconjugation.22,25 The kinetic stabilization of 2 was achieved by using sterically hindered substituents and a macrocyclic system to afford extremely long-lived singlet diradicals with a lifetime up to ca. 15 μs at 293 K.26–29 Multiphoton absorption characteristics was found for the singlet diradicaloid structures.30 Such open-shell singlet molecules have attracted much attention as important building blocks for future materials in the fields of singlet fissions and non-linear optics.31

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chart 1. Localized Diradicals, Electronic Structure and Reactivity

Page 2 of 15

between S-7 and 8 (fast process). Furthermore, the mechanisms of the novel alkoxy-group migration reactions (slow process) are clarified by examining the crossover, spin-trapping, and temperature effect on the product distribution of 9 and 10. Scheme 1. Generation and Reactivity of S-7

RESULTS AND DISCUSSION Laser flash photolysis of 6a–j. Substituent Effect on Stability of S-7. The symmetrically and asymmetrically substituted azo-alkanes, 6a–j (λmax ≈ 416 nm, ε ≈ 455 L mol–1 cm–1), which were precursors to S-7a–j, were synthesized to investigate the characteristics of the singlet 4,4-dimethoxy-1,2diazacyclopentane-3,5-diyl diradicals, S-7, as well as their reactivity (Scheme S2). The transient absorption spectra in the photochemical denitrogenation of 6a–j were measured in dry toluene using a laser-flash photolysis (LFP) method based on a Nd: YAG-OPO laser (λexc = 430 nm, ca. 3 mJ, 15-ns width, Figures 1, 2). As shown in Figure 1a, the absorption maxima of all the diradical S-7 were observed at around 650 nm which is the typical absorption wavelength for singlet diradicals.41 The species was not quenched by molecular oxygen. The absorption maxima of the asymmetrically substituted singlet diradicals, S7h,j were slightly red-shifted from 650 to 675 nm (Figure 1a). The time profile of the decay traces of S-7 clarified that the singlet diradical has dual fall processes,39 i.e. fast and slow processes (Scheme 1, Figures 1b and 1c). The fast fall process of S-7, kfast = kCP + kDR, corresponds to the thermal equilibration with the ring-closed compound 8 (Scheme 1), which was proven in our previous study.39 From the equilibrium constant K = kCP/kDR (Figure 2) the respective rate constants of kCP and kDR were determined as shown in Tables S1,S2 (Eqs. 1, 2 in Scheme 1, Figures S65–S86). To understand the bond homolysis nature and the characteristics of S-7, the values of logK, logkCP, and logkDR were plotted against the Brown-Okamoto σp+ 42,43 parameters and Creary’s radical stability parameter σC· 24 (Figure 3).

Niecke et al. first isolated the type-II molecules using the unique effect of heteroatoms on the stabilization of the singlet state of the C2P2 molecules 3.32 Later, Yoshifuji and Ito reported on the isolation of analogous C2P2 molecules.33,34 Because of the negligible intramolecular bond-formation reactivity of the typeII molecules, several type-II singlet diradicals were identified using X-ray crystallographic analysis.35 On the other hand, Bertrand et al. succeeded in isolating the type-I singlet 1,3diradicals of B2P2 4 as a rare case and in detecting a B–B bond formation reaction.36,37 Thus, the heteroatom effect has been found to play an important role in controlling the electronic structure and reactivity of singlet 1,3-diradicals. Another significant transition in elucidating the bondhomolysis chemistry was achieved through the application of the notable nitrogen atom effect.38,39 The most stable electronic configuration of 1,2-diazacyclopentane-1,3-diyl diradicals, 5, is categorized as being the type-I singlet state (Chart 1). Surprisingly, the thermal equilibration of singlet diradicals, S-7a, with the corresponding ring-closing compound, 8a, i.e. K = 1.54 at 270 K in toluene, was directly observed using transient absorption and vibrational spectroscopic analyses in the photochemical denitrogenation of 6a.39 The final product, which was isolated with a ca. 70% yield from the photodenitrogenation of 6a in dry benzene, was the alkoxy-group migration species 9a (Scheme 1).40 In this paper, the nitrogen-atom effect on the electronic structure and chemical behavior of localized singlet 1,3diradicals is presented as part of a thorough study of the effects of substituents (X, Y) and solvents on the thermal equilibrium

2 ACS Paragon Plus Environment

Page 3 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society larger than that for S-7b–d, suggesting that the singlet diradicals, 7, are highly stabilized by the electron-withdrawing group substituents. The trend in the substituent effect is very different from that for 2, in which the singlet diradicals substituted with the electron-donating group were better stabilized than those having the electron-withdrawing groups.22,23 The equilibrium constant K in the polar acetonitrile solvent was found to be smaller than that of the non-polar toluene, suggesting that the singlet diradical has a small but significant zwitterionic characteristic which is stabilized by a polar solvent.

[8a]

[S-7a]

Figure 2. Transient decay traces of singlet diradicals, S-7a– g (λmon = 650 nm), generated by photolysis from azoalkanes, 6a– g, in toluene at 270 K; (1) p-CN-C6H4, (2) p-Br-C6H4, (3) p-ClC6H4, (4) p-OMe-C6H4, (5) p-Me-C6H4, (6) p-F-C6H4, and (7) pH-C6H4. The features of S-7 were further assessed by plotting logK against the radical stability parameter σC· (Figure 3b). As suggested by the Hammett plots, a good correlation was found for the diradicals substituted with electron-donating groups (S7a–d), ρ = –1.11 (R2 = 0.92). Meanwhile, the plots for the electron-withdrawing groups (X = Y = Cl, Br) were placed below the linear correlation observed for the case substituted by the electron-donating group, indicating that the benzylic carbon has an anionic characteristic as suggested by the reaction constant for the Hammett plot (Figure 3a). Thus, the singlet diradicals, S-7, possess a polar character that is visualized by the zwitterionic resonance structure ZI’ (Scheme 2).4

Time profile of slow process

Figure 1. (a) Transient absorption spectrum of S-7a–j (p-X, p-Y), generated from azoalkanes 6a–j (p-X, p-Y) by photolysis in toluene at 293 K. (b) Time profile of S-7a in toluene at 293 K (λmon = 650 nm) (c) Time profile of slow process of S-7a at 293 K (λmon = 650 nm). Substituent Effect on Equilibration Constant K. As shown in Figure 3a, the Hammett-type plots of logK, both in toluene and acetonitrile, exhibited reverse V-shaped lines, with the slope (= Hammett reaction constant, ρ) = 0.50 (R2 = 0.99) for S-7a–d and ρ = –4.23 (R2 = 0.99) for S-7a,e,f. The singlet diradicals are thermodynamically stabilized by introducing both the electron-donating (X = Y = OMe, Me, F) groups (S-7b–d) and the electron-withdrawing (X = Y = Cl, Br) groups (S-7e,f) at the para positions of the phenyl rings. The equilibration of S7g (X = Y = CN) with 8g was not observed (Figure 2-1), indicating that the singlet diradicals are well stabilized by the cyano group (X = Y = CN). The reverse V-shape of the Hammett plots is a typical example of the substituent effect on benzyl-type radicals.44 Interestingly, the slope (ρ) for S-7e,f substituted with electron-withdrawing groups (X = Y = Cl, Br) is negatively

3 ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 15

Figure 3. Hammett plot with σp+ (a, d, g) and σC· (b, e, h) for the equilibrium reaction between S-7 and 8; blue: toluene, red: acetonitrile. Correlation between (c) logK, (f) logkCP and (i) logkDR and the solvent parameter ET(30) for fast process (S-7 → 8), respectively; green: p-H-C6H4, red: p-OMe-C6H4, blue: p-Br-C6H4.

The asymmetrically substituted singlet diradical S-7h (X = OMe, Y = Br) was generated in the LFP of azoalkane 6h, in which the electron-donating (OMe) and electron-withdrawing (Br) substituents were introduced at the para position (Table S1: entries 3, 4, 11–14). In both toluene and acetonitrile, the equilibrium constants KOMe,Br were found to be ca. 0.41 in toluene and 0.19 in acetonitrile, which were nearly the same as the values obtained for X = Y = Br, KBr,Br ≈ 0.39, 0.13 at 270 K, respectively, but differed greatly from those for X = Y = OMe, KOMe,OMe ≈ 0.60, 0.36. The notable substituent effect evidently indicates that the contribution of ZI’ play an important role in determining the character. The benzyl cation part is highly stabilized by the nitrogen atom. Substituent Effect on Ring-Closing Process (kCP). To gain a further insight into the electronic characteristics of S-7 and its reactivity in the bond-formation process, the substituent effect on the ring-closing reaction (kCP) was examined using Hammetttype plots (Figures 3d–f and Figure S1b). Surprisingly, ringclosing reactions of substrates substituted with electrondonating groups (S-7b,c) were faster than that of S-7a (X = Y = H), although S-7 is energetically stabilized by introducing electron-donating substituents (Figures 3a,b). Indeed, the equilibration constant K for S-7b,c was smaller than that for S7a. In contrast to the interesting substituent effect on kCP, the

Effect of Solvent on Equilibration Constant K. To gain further information on the characteristics of the singlet diradicals, S-7, the solvent effect on the equilibrium constant K was examined in various solvents (Tables S1,   S2, Figure 3c and Figure S1a). The logarithms of the equilibrium constants K for X = H, MeO, Br, logK, were plotted against a solvent parameter ET(30)45. As shown in Figure 3c, although the solvent effect is small46, the slope is 0.020–0.036, with the solvent parameters being highly correlated with logK except in the DMF case. The equilibrium constant K decreases slightly as the solvent polarity increases. Thus, the solvent effect of the thermodynamic stability on the singlet diradicals was larger than that of the corresponding ring-closing compound, 8. The solvent effect is consistent with the correlation between the equilibrium constant K and the σp+, σC·, demonstrating that the singlet diradicals, S-7, exhibit the zwitterionic characteristic. The effect of the solvent on the electron-withdrawing groups (X = Y = Br) was slightly larger than that on the electron-donating groups (X = Y = OMe), as indicated by the slopes of the plots –0.036 (Br) < –0.027 (H) < –0.020 (OMe). This result supports that the electronwithdrawing substituents increase the zwitterionic character.

4 ACS Paragon Plus Environment

Page 5 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

electron-withdrawing group decelerated the ring-closing reaction (Table S1: entries 9–12). The negative reaction constants, –0.45 and –0.13, in the Hammett plots for both acetonitrile and toluene, logkCP versus σp+, respectively, clearly demonstrate that the ring-closing reaction is accelerated by the electron-donating substituents (Figure 3d), indicating that the benzylic carbon is more positively charged in the transition state than S-7 (Scheme 2, vide infra in Scheme 7). Indeed, the ringclosing reaction was highly dependent on the solvent polarity (Figure 3f). The rate constants in polar solvents such as acetonitrile were larger than that in a non-polar solvent such as toluene. On the other hand, the ring-closing reaction of S-7e,f with the electron-withdrawing group was prone to be slower in acetonitrile than in toluene (Figure 3f). A linear correlation for kCP versus σC· was observed for the electron-donating group into which S-7 was substituted, ρ = 1.11 (R2 = 0.95) (Figure 3e). The data for the case in which the electron-withdrawing group was substituted was found to deviate from this line.

groups accelerate the ring-closing reaction (Figures 3d–f). The transition state being more highly charged than S-7 explains the effect of the polar solvent on the acceleration of the ring-closing reaction for S-7, substituted by the electron-donating group (Scheme 3). In contrast to the rate acceleration induced by the electron donation, the deceleration was observed for the diradicals substituted by the electron-withdrawing group, S-7e,f (entries 9–12 in Table S1, Figures 3d–f). The notable substituent effects can be rationalized by the contribution of the zwitterionic resonance structure ZI’ in S-7, which is well stabilized by polar solvents such as acetonitrile, thus increasing the energy barrier for the formation of the ring-closing compound (Scheme 2, Figure 3d). During the bond formation, the negative charge on the benzylic carbon decreases to decrease the polarity in TS. Thus, the ring-closing rate constant in a polar solvent is smaller than that in a non-polar solvent (Figures 3d, f). Substituent Effect on Ring-Opening Reaction Process (kDR). Next, the substituent and solvent effects on the ringopening rate constant kDR were also examined by the Hammetttype plots using σp+ and σC· parameters to understand the homolytic bond-breaking process (Table S1, S2, Figures 3g–i and Figure S1c). First, the relation between σp+ and the ringopening rate constant logkDR was investigated (Figure 3g). Both the electron-donating and electron-withdrawing groups accelerated the ring-opening reaction (V-shape plot). Thus, the reaction constant ρ for S-7a,e,f, substituted by the electronwithdrawing group, was found to be 2.9 (R2 = 0.91) in toluene and 2.4 (R2 = 0.94) in acetonitrile. On the other hand, the reaction constant ρ for the S-7a–c, substituted by the electrondonating group was observed to be –0.63 (R2 = 0.97) in toluene and –0.79 (R2 = 0.98) in acetonitrile. These results clearly indicate that the radical character at the benzylic carbon gradually increases during the bond-breaking reaction as shown in CP_DR (Scheme 2). The radical stabilization effect that accelerates the ring-opening process was also proven by the good positive linear correlation of logkDR with σC·, the reaction constant ρ = 1.5 (R2 = 0.96) in toluene and ρ = 1.8 (R2 = 0.90) in acetonitrile (Figure 3h). Consistent with the radical character of the transition state, the solvent effect for kDR was found to be small (Figure 3i). Interestingly, the plotted data for the electronwithdrawing groups (X = Y = Cl, Br) were placed slightly above the linear correlation observed for the electron-donating groups (Figure 3h). The phenomenon is also explained by the zwitterionic characteristics of ZI2 in S-7e,f, in which the anionic characteristic exists in the benzylic carbon. To obtain further information on the ring-opening reaction, we investigated the correlation between logkDR and the solvent parameter; ET(30) (Tables S1, S2 and Figure 3i). Although the solvent effect was small, a good linear relation which gave a positive ρ value was obtained for log kDR with the solvent parameters.46 Alkoxy-Group Migration Reactions. As shown in Scheme 1, the major product isolated with a ca. 70% yield was the alkoxy-group-migrated compound 9a in the photochemical denitrogenation of the azoalkane 6a (λmax ≈ 416 nm, ε ≈ 455 L mol –1 cm –1 ) in a nonpolar solvent such as toluene or benzene. 39,40 To obtain more information about the reaction mechanism of the alkoxy-group migration reaction of singlet diradicals, S-7a, the photodenitrogenation of azoalkane, 6a, was conducted using 416 nm irradiation in

Scheme 2. Homolytic Ring-closing Reaction of S-7 and Homolytic Ring-opening Reaction of 8 H3CO OCH3 Ar

Ar N N O

N Ph

O

TS

Non-polar solvent

H3CO OCH3  Ar Ar 

H3CO OCH3 • • Ar Ar 

N N O

N N O

N Ph

O

DR_CP

O

N Ph

Polar solvent

CP_DR

Non-polar solvent

Non-polar solvent (a) Ar = electron-donating group

(b) Ar = electron-withdrawing group Polar solvent

Polar solvent H3CO OCH3 Ar

Ar

Ar

O

O

N Ph

Ar

Ar

S-7

N Ph

Ar N N

N N

N N O

H3CO OCH3

H3CO OCH3

O

O

ZI’

Ar

K = [8] / [S-7] at 270 K toluene, acetonitrile

C6H5 p-MeOC6H4

1.54, 0.72 0.60, 0.36

p-MeC6H4

1.03, 0.58

p-FC6H4

1.44, 0.59

p-ClC6H4

0.40, 0.15

p-BrC6H4

0.39, 0.13

N Ph

O

8

Based on the observed substituent and solvent effects on kCP, the energy profile for the ring-closing process from S-7 to 8 is shown in Scheme 2. The curved red-arrows in ZI’ describe the electron movement from S-7 to 8 via TS. The benzylic carbon gradually becomes positive during the bond formation reaction as shown in the structure DR_CP; thus, the electron-donating

5 ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

acetonitrile to determine the effect of the solvent (Table 1). The solvent was used after thorough dehydration.47 After the disappearance of the azoalkane, 6a, the photolysate was analyzed by 1H NMR (400 MHz) to identify the products and determine the chemical yields using triphenylmethane as an internal standard (entries 2, 4 in Table 1). Interestingly, although the major photoreaction product in toluene was 9aa (75% yield, entry 1), in polar acetonitrile, 10aa was isolated with a yield of 63%, together with 9aa (25% yield, entry 2). Thus, the alkoxygroup migration product (slow process in Figure 1b) was found to be highly dependent on the solvent polarity, as is clear when entries 1 and 3 are compared with entries 2 and 4. No concentration effect was, however, observed for the chemical yields or product ratios, as is clear when entries 1 and 2 are compared with entries 3 and 4. Crossover Experiments. To gain greater insights into the mechanism of the alkoxy-group migration reactions, azoalkanes 6k (OR = OC3H7) and 6l (OR = OC10H21) were prepared and crossover experiments in dry solvents were conducted for the photodenitrogenation of equimolar mixtures of 6a + 6k and 6a + 6l, respectively (Scheme 3). In both reactions, no crossover products were observed for migration compound 10, but for 9, 9ak (OCH3, OC3H7) and 9ka (OC3H7, OCH3), as well as 9al

Table 1. Product distribution for photodenitrogenation of azoalkane 6a in dried toluene and acetonitrile, respectively. a, b H3CO OCH3 O Ph Ph N N N N O N Ph

OCH3 h (416 ± 10 nm)

Ph

Ph N N

a toluene (Tol) a acetonitrile (ACN)

O

6a

Entry

+

O

N Ph

OCH3 OCH3

Ph

N N O

OCH3

9aa

Solv

b Yield

Conc./ mM

of

Ph

N Ph

O

10aa

b Yield

of

9aa

10aa

Ratios of 9/10

1

Tol

30

75

22

77/23

2

ACN

30

25

63

28/72

3

Tol

100

70

29

71/29

4

ACN

100

29

67

30/70

Page 6 of 15

The solvents were dried using 3-Å molecular sieves at 300 °C for 24 h in vacuo.b Product yields were determined by 1H NMR analysis. Ph3CH was used as an internal standard, ± 5% error. a

Scheme 3. Crossover Experiments Examining Formation of 9 and 10 in Photodenitrogenation of 6a,k,l 6a + 6l

6a + 6k h (416 ± 10 nm)

OCH3 Ph

O

Ph +

O

N Ph

(Tol) (ACN)

Ph N N

O

OCH3

9aa

h (416 ± 10 nm)

a acetonitrile

OCH3

Ph N N

a toluene

N Ph

Ph +

O

OC3H7

OCH3 OCH3

O

O

10aa

9ak

+

+ OC3H7 Ph

OC3H7

Ph N N

O

O

N Ph

Ph

Ph N N

+

OC3H7

O

9kk

N Ph

Ph

O

OCH3

9ka

OC3H7 OC3H7 Ph

O

O

O

N Ph

OC10H21

OC10H21

Ph O

N Ph

Ph +

O OC10H21

9ll

10kk

N Ph

10aa

+

O

9al

OC10H21 N N

Ph

N N

+

Ph N N

+

9aa O

N Ph

(Tol) (ACN)

OCH3 Ph

Ph

N N

a toluene

a acetonitrile

Ph N N

O

N Ph

OC10H21 OC10H21

Ph

N N

+

OCH3

O

N Ph

9la

Ph O

10ll

Table 2. Product distribution in photodenitrogenation of 1:1 mixture of 6a and 6k Entry

Solv.

9aa

9kk

9ak

9ka

10aa

10kk

c 9/10

1

Tol

21

21

21

21

7

9

84/16

2

ACN

8

7

5

5

37

38

24/76

Table 3. Product distribution in photodenitrogenation of 1:1 mixture of 6a and 6l Entry

Solv.

9aa

9ll

9al

9la

10aa

10ll

c 9/10

1

Tol

31

30

8

6

9

10

80/20

2

ACN

19

18

5

2

26

23

47/53

The solvents were dried using 3-Å molecular sieves at 300 °C for 24 h in vacuo. Product yields were determined by 1H NMR analysis. Ph3CH was used as an internal standard, ± 5% error. c 9 = 9aa + 9kk (or 9ll) + 9ak (or 9al) + 9ka (or 9la), 10 = 10aa + 10kk (or 10ll) a

b

6 ACS Paragon Plus Environment

Page 7 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society 8

(OCH3, OC10H21) and 9la (OC10H21, OCH3) were observed (Tables 2, 3). The results of the crossover experiments indicate that 9 is an intermolecular reaction product (out-of-cage product) while 10 is an intramolecular reaction product (in-cage product). The alkoxy-migrated products, 9aa, 9ak, 9kk, and 9ka, produced by the photolysis of a 1:1 mixture of azoalkanes 6a (OR = OCH3) and 6k (OR = OC3H7) were observed at the same product ratio of 21% in toluene and 5–7% in acetonitrile (Table 2, entries 1, 2). Meanwhile, in-cage products 9aa and 9ll were preferably formed by the photolysis of 6a and 6l (OR = OC10H21) both in toluene and acetonitrile (Table 3, entries 1, 2). The preferable formation of the in-cage products can be rationalized by the slower diffusion rate of the large decyloxy group (OC10H21) than the propoxy group.48 Based on the results of the crossover experiments and solvent effect on the product ratio of 9 versus 10, the mechanism for the alkoxy group migration reaction is proposed in Scheme 4. After the fast equilibration of S-7 with 8, the intramolecular 1,2-alkoxy migration occurs to produce 10 from S-7, with the process being accelerated in a polar solvent. The formation of out-of-cage product 9 may be rationalized by the triplet pair of radical species and OR·, derived from the C–O bond fission in the triplet diradical, T-7. After the intersystem crossing to the singlet state, bond formation selectively occurs for the carbonyl carbon to produce isomer, 9. If the triplet pair of radicals (RP) is formed, the radical species may be trapped by spin-trapping agents such as N-tert-butyl-α-phenylnitrone (PBN). Spin-trapping Experiments.49 Spin-trapping experiments were conducted to examine the photodenitrogenation (360–440 nm) of azoalkanes 6a (X = Y = H, σC·: 0.00), 6b (X = Y = OMe, 0.27) and 6f (X = Y = Br, 0.11) (ca. 17 mM) in the presence of N-tert-butyl-α-phenylnitrone (PBN) (ca. 102 mM) in dry benzene at room temperature (Figure 4, Figure S28). Although the in-situ EPR signals observed in the photolysis of 6a and 6f were relatively complex (Figure S28), the EPR signals of a nitroxide having the hyperfine coupling constants of A1N = 14.71 G, A1Hβ = 4.79 G were detected for the spin-trapping experiments in the photolysis of 6b (Figure 4). Atmospheric pressure chemical ionization (APCI) mass spectroscopic analysis of the in-situ EPR sample demonstrated the formation of molecular ion peaks of M+; m/z 633.27 (δ: 1.798 ppm) that correspond to the RPC spin adduct of PBN (Scheme 4, Figure S27). Thus, the outof-cage product 9 is proposed to be formed via the triplet radical pairs RP after the C-O bond homolytic cleavage from the triplet diradicals, T-7. The EPR signals did not conform to the reported MeO· radical trapped nitoroxide (AN = 13.73 G and AHβ = 1.93 G) with PBN.50–53 The EPR signals of the MeO radical trapped nitroxide were observed at higher concentration of PBN (Figure S29). The existence of the ion pair (IP) was proven by the selective formation of 9aa from the thermal reaction of 10aa with methanol (Scheme 4).

6

h –N2

O

Ar

Ar ISC

N N O

N Ph S-7

Ar

Ar

O

O

N Ph

N N

T-7

Ar

N Ph ZI’

O

O N cage Ph radical pair (RP)

OCH3 Ar

triplet reaction

Ar N N

OCH3

Ar

N N O

H3CO OCH3

O

OCH3

H3CO OCH3

H3CO OCH3 Ar

Ar N N

singlet reaction

O

+ O

N Ph RPC

OCH3

ISC ET H3CO Ar

OCH3 Ar

N N O

N Ph TS9

O

OCH3

10

Ar

Ar N N

9 O

N Ph C

+

OCH3

O ion pair (IP)

Figure 4. Spin-trapping experiments using EPR measurements. EPR spectrum obtained from photoreaction of 6b (X = Y = OMe) and PBN.

Regioselectivity of Alkoxy Migration Reaction. To obtain further information on the mechanism of 10 formation, the product selectivity was investigated in the photodenitrogenation of asymmetrically substituted azoalkanes6i (X = OMe, Y = H) and 6j (X = CN, Y = H) at 293 K using Xenon lamp (λexc = 416 ± 10 nm) (Table 4). In the photodenitrogenation of 6i (entries 1, 2), the four expected products, 9ii, 9i’i’, 10ii, and 10i’i’, were observed at a ratio of 37:38:19:6 in toluene and 9:9:55:27 in acetonitrile with > 90% total yields (entries 1, 2 in Table 4), which were analyzed by 1H NMR (400-MHz) spectroscopy. No regioselectivity was observed for 9, either in toluene or acetonitrile. The selective formation of 10 was, however, observed at a ratio of ca. 70:30 of 10ii:10i’I’ (entries 1, 2 in Table 4). Methoxy migration selectively occurred for the pMeOC6H4-substituted carbon. In contrast to the substituent effect on the regioselectivity, the selective formation of the 10j’j’ isomer was observed at a ratio of ca. 30:70 of 10jj:10j’j’ in the reaction of 6j (entries 3, 4). No regioselectivity in 9 was also observed for the reaction of 6j (entry 3).

Scheme 4. Mechanism for Formation of 9 and 10

7 ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Effect of Temperature on Regioselectivity of Alkoxygroup Migration. To obtain information on the potential energy profile of the alkoxy-group migration reactions of S-7, the temperature effect on the product selectivity, 9 versus 10, was investigated in the photochemical denitrogenation of 6a,b,e in dry toluene and acetonitrile over a temperatures range of 273– 313 K (Figure 5 and Figures S43, S46, S52 and S56). At 273 K, 9aa was mainly observed in dry toluene (9aa/10aa = 94/6) (Figure 5). The amount of 10aa product increased with the reaction temperature (9aa/10aa = 83/17 at 293 K, 65/35 at 303 K, and 62/38 at 308 K (Figure 5)). Surprisingly, the yield of 10aa suddenly increased at temperatures above ca. 310 K (= 37 °C). At around this temperature, the product ratio became inverted from 9aa/10aa = 62/38 to 9/91. Both 9aa and 10aa were stable under the reaction conditions; thus, the temperature effect clearly suggests that the activation entropy plays a key role in the product selectivity. Only 9aa was observed at temperatures < 263 K. At temperatures > 318 K, however, only 10aa was detected using 1H NMR spectroscopic analysis (Figure 6a, Figure S43). The temperature effect on the product selectivity suggests that the rate-determining step changes around 273–313 K. In acetonitrile, the temperature effect is essentially negligible for product ratios of 9aa/10aa = 14/86 to 9/91 (Figure 5). To obtain information on the effects of the temperature and solvent, as well as the mechanism, on the product selectivity of 9aa/10aa, the effect of temperature on the slow decay rate process of S-6a was investigated in both toluene and acetonitrile (Scheme 1, Figure 1b, Figures 6a,b). The Eyring plot, ln(kslow/T) versus 1/T, is shown in Figure 6. The linearity of the Eyring plot was not obtained for a toluene solution over a temperature range of 273–313 K (ii). At this temperature both 9aa and 10aa were formed (Figure 5). Two different linear correlations were observed in those regions with temperatures > 313 K (40 °C) (i) and < 263 K (–10 °C) (iii) (Figure 6a). 10aa was exclusively formed in temperature range (i). In temperature range (iii), only 9aa was detected. The obtained activation parameters in temperature regions (i) and (iii) are summarized in Table 5. In temperature region (i), the enthalpy factor dominates the Gibbs activation energy, ΔH‡ = 66.8 kJ mol–1, ΔS‡ = –7.1 J K–1 mol–1, ΔG‡ = 68.6 kJ mol–1 at 253 K, and ΔG‡ = 69.3 kJ mol–1 at 353 K. On the other hand, the entropy factor greatly affects the rate constant kslow in temperature region (iii). In this region, the activation parameters were obtained as ΔH‡ = 21.4 kJ mol–1, ΔS‡ = –145.3 J K–1 mol–1, ΔG‡ = 58.2 kJ mol–1 at 253 K, and ΔG‡ = 72.7 kJ mol–1 at 353 K. The large entropy contribution in the low-temperature region (iii) is reasonable, because 9 is formed from triplet diradical T-7 via the intersystem crossing process (ISC) from S-7 (Scheme 4). The spin-forbidden ISC processes are known to provide entropy control, whereby, in general, large negative entropy and small logA values are observed.11,54 In the high-temperature region (i), the Gibbs energy for the ISC process becomes larger than the energy for another reaction pathway from S-7a to 9aa. Thus, a dramatic temperature effect was observed for the product selectivity in the photodenitrogenation of 6a. The logA value of ca. 12 supports the spin-allowed reaction process.

Table 4. Product distributions for photodenitrogenation of azoalkanes 6i (entries 1, 2) and 6j (entries 3, 4) in dried toluene and acetonitrile at 293 K, respectively. a

Entr y

6 (X)

Solv.

9:9':10:10'

1

6i (OMe)

Tol

37:38:19:6

76/24

2

6i (OMe)

ACN

9:9:55:27

67/33

3

6j (CN)

Tol

40:40:5:15

38/62

4

6j (CN)

ACN

0:0:28:72

28/72

10/10’

b

a The solvents were dried with 3-Å molecular sieves that were dried at 300 °C for 24 h in vacuo.

Scheme 5. Resonance Structure of Donor–Acceptor Substituted S-7

EWD

EWD

EDG

N Ph S-7

EDG

EWD

O

O

EDG N N

N N

N N O

H3CO OCH3

H3CO OCH3

H3CO OCH3

O

O

N Ph ZI’

EDG

OCH3 EWD

N N O

N Ph C

O

ZI’’

OCH3 EWD

N Ph

EDG N N

O

O OCH3

N Ph C’

Page 8 of 15

O

The selective formation of 10ii and 10j’j’ is rationalized by the charge distribution on the singlet diradicals S-7i and S-7j (Scheme 5). The contribution of ZI’, EDG = p-MeOC6H4 and EWG = Ph for S-7i, EDG = Ph and EWG = p-CNC6H4 for S-7j, should be larger than ZI’’, because the electron-withdrawing group stabilizes the anion. Thus, the methoxy group selectively migrates to the EDG group substituted carbon. In the intermediate C that produces 9 (Scheme 4), an equal charge distribution would be expected because the positive charge is well stabilized by the nitrogen atoms. Thus, the methoxide equally attacks the carbonyl carbon to give a 1:1 mixture of 9 and 9’.

8 ACS Paragon Plus Environment

Page 9 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society 1

Tol

7a → 10aa

314–358

2

Tol

7a → 9aa

225–260

3

ACN

7a → 10aa

251–322

66.8 (69.6)

–7.1 (12.9) –145.3 (5.5) –12.1 (12.6)

21.4 (23.4) 56.2 (58.6)

a Activation enthalpy (activation energy) in kJ mol-1 for the migrated

reaction using Eqs. (1)–(3) (Scheme 1). bActivation entropy (logA) in J K-1 mol-1 for the migrated reaction using Eqs. (1)–(3) (Scheme 1)

The notable temperature effect on the product selectivity is visualized in Scheme 6b. In the low temperature region (iii) < ~260 K (blue color), the ISC process dominates the fate of S-7 to give the triplet state T-7, from which the C–O bond scission occurs to finally afford 9. In the higher temperature region (i) > ~310 K (red color), the energy barrier of the methoxy-migration step becomes smaller than the ISC process to produce 10 predominantly. In acetonitrile, the linear Eyring plot was obtained in the temperature region addressed in the present study (Figure 6b). The ratio of 9aa:10aa was constant at ca. 10:90 (Figure 5). The energy barrier was determined to be smaller than that in toluene. Thus, the concerted alkoxy-migration is accelerated by the polar solvent. Computational Study on Reactivity of 7. As mentioned above, the reactivity of S-7 was well-understood by the (1) substituent and solvent effects on K, kCP, and kDR; (2) crossover experiments; (3) spin-trapping experiments; (4) regioselectivity of alkoxy-migration; and (5) temperature effect on the product distribution. The complete active space SCF (CASSCF) computational calculation of the substituent and solvent effects on the reactivity of S-7 is challenging. First, the thermal equilibration constant K (= [8a]/[S-7a], X = Y = H) was computed by the broken-symmetry (BS)55 method using several types of density-functional theory (DFT), B3LYP,56 CAMB3LYP,57 M06-2x,58 and ωB98XD59/6-31G(d)60 levels of theory (Table 6), whose values were compared with the experimentally obtained K value (K = 1.74 and 0.95 at 298 K in toluene and acetonitrile, respectively). The four conformers of the open-shell singlet state of S-7a, i.e. S-7a-ii, S-7a-ig, S-7a-gg, and S-7a-gi, were found as equilibrium structures (Scheme 6a). The endo isomer 8a_endo was computed to be lower in energy than 8a_exo (Scheme 6b). The energy difference and Kcalc values were calculated between S-7-ii and 8_endo. The experimentally obtained K = 1.74 at 298 K in toluene (Tol) (entry 7) was fairly reproduced by the CAMB3LYP/6-31G(d) level of theory, which resulted in Kcalc = 1.35 at 298.15 K (entry 2). The

Figure 5. Temperature dependence of alkoxy-migrated species ratio (9aa vs. 10aa).

Table 6. Comparison of Computed and Experimental Gibbs energy differences (ΔG) and equilibrium constants (K) between S-7 and 8.

Figure 6. Eyring plot for alkoxy-migrated reaction (6a → 9aa + 10aa) in dried (a) toluene and (b) acetonitrile.

H3CO p-X-C6H4

H3CO OCH3

OCH3 p-Y-C6H4

p-X-C6H4

N N

Table 5. Activation parameters Δ H ‡ , Δ S ‡ , Ea, and logA for the alkoxy migrated reaction. Ent ry

Solv.

Reaction

Temperat ure range/K

ΔH‡ a (Ea)

O

ΔS‡ b (logA)

Entry

X,Y

9 ACS Paragon Plus Environment

N Ph

K

O

O

S-7-ii DFT methoda

p-Y-C6H4 N N

solventb

N Ph

O

ΔG in kJ mol–1

8endo Kcalc

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

B3LYP

2

CAMB3LYP

–14.43

0.003

0.75

1.35

M06-2x

28.06

82439

ωB98XD

17.51

1168

Tol

1.4

1.76

ACN

–0.07

0.97

Tol

1.38

1.74c

ACN

–0.13

0.95 c

CAMB3LYP

Tol

–1.23

0.61

b

Exp.

Tol

–0.98

0.67 c

OMe,OMe

CAMB3LYP

ACN

–2.12

0.43

12

Exp.

ACN

–1.76

0.49 c

13

CAMB3LYP

Tol

–2.08

0.43

e

Exp.

Tol

–1.79

0.49 c

Cl,Cl

CAMB3LYP

ACN

–3.79

0.22

16

Exp.

ACN

–3.66

0.23 c

17

CAMB3LYP

Tol

–8.67

0.03

18

g

Exp.

Tol

nd

~0.00 c

CN,CN

CAMB3LYP

ACN

–10.73

0.01

20

Exp.

ACN

nd

~0.00 c

21

CAMB3LYP

Tol

–3.08

0.29

h

Exp.

Tol

–1.73

0.50 c

OMe,Br

CAMB3LYP

ACN

–4.44

0.17

Exp.

ACN

–3.13

0.28 c

3 4 5

a H,H

6 7

Exp.

8 9 10 11

14 15

19

22 23 24

CAMB3LYP

Gas

d

d

Page 10 of 15

at the BS-(U)CAM-B3LYP/6-31G(d) level of theory to obtain the information on the reactivity of S-7a (Scheme 6b(1)). Interestingly, the transition state TSaendo was found to produce the endo conformer 8a_endo. In accordance with the computed result, 8a_endo was calculated to be more stable in terms of Gibbs energy than the exo isomer 8a_exo by 37.3 kJ mol–1 at the CAM-B3LYP/6-31G(d) level of theory. The energetic preference of the endo isomer is rationalized by the electron delocalization of the lone-pair of electrons on the nitrogen atoms to the C–C σ* of the cyclopropane ring. Indeed, the N–N distance in CP4a_endo (1.44 Å) was shorter than that in 8a_exo (1.45 Å). The C–C bond distance in 8a_endo (1.588 Å) was found to be longer than that in 8a_exo (1.568 Å). The N–C bond distance in 8a_endo (1.48 Å) was found to be shorter than that in 8a_exo (1.50 Å). The irc analysis revealed that the transition state TSaendo was connected with S-7a-ig. At this level of theory, the Gibbs energy barrier from S-7a-ig to 8a_endo was calculated to be 41.1 kJ mol–1 in the gas phase at 298 K. Thus, the DFT computations well-reproduced the Gibbs energy of activation (ΔG298‡ = 38.4 kJ mol–1, Ea = 39.8 kJ mol–1) in the ring-closing reaction (Scheme 6b(1), Table S1). The B3LYP method was prone to overestimate the stability of S-7a to give the Gibbs activation energy of 53.1 kJ mol–1 (Scheme 6b).

aThe 6-31G(d) level of theory was used for the computations. b The Polarizable continuum model (PCM) was used for estimating the solvent effect. c Experimental values at 298 K. d The values could not be determined by the experiments (Figure 2(1)).

B3LYP method overestimated the stability of S-7a (entry 1), while the other methods underestimated it (entries 3,4). Based on these results, it can be deduced that the CAM-B3LYP method would be appropriate for analyzing the reactivity of S-7a. The effect of solvent on K was computed using the polarizable continuum model (PCM)61 (entries 5 and 6). The solvation model well-reproduced the experimentally obtained K values, Kexp = 1.74 and 0.95 in toluene (Tol) and acetonitrile (ACN), respectively, at 298 K (entries 7 and 8). The computed K values (Kcalc) at the CAM-B3LYP level were also consistent with the experimentally obtained K values for other substrates (compare entries 9 and 11 with 10 and 12 for b (X = Y = OMe), entries 13 and 15 with 14 and 16 for c (X = Y = Cl), entries 17 and 19 with 18 and 20 for g (X = Y = CN), and entries 21 and 23 with 22 and 24 for h (X = OMe, Y = Br), respectively). Scheme 6b summarizes the experimentally observed reactivity of S-7a. The singlet ground state of S-7a-ii was confirmed by computing the triplet state, which was 16.6 (21.3) kJ mol–1 higher in Gibbs energy at 298 K than S-7a at the BSUCAM-B3LYP (UB3LYP)/6-31G(d) level of theory (Scheme 6b). Next, the ring-closing reaction of S-7a to 8a was computed

10 ACS Paragon Plus Environment

Page 11 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Scheme 6. (a) Equilibrated structures in S-7a and their relative energies at the BS-UB3LYP/6-31G(d) level; (b) Temperature dependency of product selectivity for S-7a.

The methoxy migration step was computed at the same level of theories (Scheme 6b(2), 6b(3)). A Gibbs energy of 105.9 kJ mol–1 at 298 K was computed for the formation of 10a at the CAM-UB3LYP/6-31G(d) level of theory (Scheme 6b(2)), which was higher than the experimental value of 68.9 kJ mol–1. The Gibbs energy of 73.0 kJ mol–1 was rather consistent with that obtained at the B3LYP/6-31G(d) level of theory. The endomethoxy group was found to preferably migrate to the benzyl carbon to afford 10a. The C–O bond scission in T-7a (Scheme 6b(3)) was also computed to estimate the energy barrier. The Gibbs energy barrier was found to be 81.4 kJ mol–1 and 58.5 kJ mol–1 at the CAM-UB3LYP and UB3LYP levels of theory, respectively, which were smaller than those computed for the migration step from S-7a to 10a. The substituent effect on the rate constant of the ring-closing reaction of S-7 to 8 clearly suggested that the positive charge at the benzylic position increases during the radical-radical

coupling reaction (Figure 3, Scheme 2). The charge movement was first computed for the parent system from S-7H to 8H via TSH using the atomic polar tensor (APT) charges at the radical carbon (Scheme 7). The APT charge of the radical carbon increased from 0.069 to 0.18 during the bond formation reaction. Thus, the charge movement was consistent with the experimental observation of the substituent effect. A similar increase in positive charge was also computed for the reaction of S-7a to 8a_endo (Scheme 7). For the reaction of S-7a to 8a, the total charge of the benzyl Scheme 7. Charge movement during ring-closing reaction from S-7 to 8 via TSH.

11 ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

HO OH H 0.069 O

HO OH H

H 0.18

N N O

N H

O

H3CO OCH3 Ph

O

N H

O

H3CO OCH3

Ph N N

0.25

O

TSH

Ph

O

O

S-7a

N H

8H OCH3

Ph

N N

0.29

O

N H

H3CO

Ph

dialkoxy-1,2-diazacyclopentane-3,5-diyl diradicals, S-7, were thoroughly investigated by transient absorption spectroscopic analysis in the photochemical denitrogenation of azoalkanes 6 by applying LFP experiments. In the transient absorption spectroscopic analysis, dual-decay processes were found for all the electron-donating and electron-withdrawing groups, substituted with S-7, except S-7g. The fast decay process was assigned to the thermal equilibration with the corresponding ring-closing compounds, 8. The slow process was found to be the alkoxy-group migration to produce 9 and/or 10. The characteristics of S-7 were revealed by the solvent and the substituent effect on the equilibration constant with their corresponding ring-closed compounds, 8, K = kCP/kDR, as a sensitive probe. The zwitterionic resonance structure ZI’ explains well the behavior of S-7.65 The ring-closing process (kCP) of S-7 and the openingprocess (kDR) of 8 were investigated in detail using their Hammett-type plots of σp+ and σC·. The results revealed the superimposed characteristics of a diradical and zwitterion in S7, which play an important role in determining the reactivity of the localized singlet diradicals. The crossover, spin-trapping, and LFP experiments for the slow decay process demonstrated the reaction mechanism of the formation of alkoxy-group migration products 9 and 10, which are the slow decay process of S-7. The unique effect of the temperature on the product ratio clarified the individual reaction mechanism for the formation of 9 and 10. The ISC process was found to be a key step in the formation of 9. The intramolecular migration process for the formation of 10 was found to be accelerated by a polar solvent.

OH

H H 0.22 N N

N N N H

S-7H

HO H

Ph

0.35 N N O

O

N H

TSaendo

O 8aendo

moiety was compared in the optimized structures of S-7a, TSaendo, and 8a_endo. The substituent effect on the Gibbs energy of TSendo was computed at the BS-UCAM-B3LYP/6-31G(d) level of theory at 298 K in the gas phase (Scheme 8).62-64 The electron-donating OMe substituent (7b: Ar = p-OMeC6H4) was found to slightly lower the energy (ΔG‡) of TSendo for 7a (Ar = p-HC6H4). The energy difference (ΔΔG‡rel) was found to be –0.3 kJ mol–1, which was consistent with the experimental energy difference in toluene (ΔΔG‡rel = –0.2 kJ mol–1). In contrast, the energy increased on introducing the electron-withdrawing Cl substituent (7e: Ar = p-ClC6H4) (ΔΔG‡rel = +2.0 kJ mol–1), as observed in the experiment. The computed substituent effect on the energy for the ring-closing reaction of S-7 to 8 was qualitatively consistent with that obtained from the experiments (Figure 3d). In polar acetonitrile using the PCM method, the computed Gibbs energy increased for the ring-closing reaction of 7a (ΔΔG‡rel = +1.3 kJ mol–1) and 7e (ΔΔG‡rel = +3.5 kJ mol– 1). However, the energy decreased in acetonitrile for the reaction of 7b (ΔΔG‡rel = –0.6 kJ mol–1). The solvent effect on the ringclosing rate constant, kCP, for 7b was consistent with that observed in the experiment, although the computed energy was not consistent for 7a.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Experimental procedures and characterization data for all compounds, including copies of 1H NMR, 13C NMR, the absolute values of kCP and kDR, computational results (PDF)

Scheme 8. Substituent effect on Gibbs energy of cyclization reaction from S-7 to 8. H3CO Ar

OCH3 Ar

H3CO Ar

N N O

N Ph

OCH3 Ar

O S-7

O

N Ph

AUTHOR INFORMATION

H3CO OCH3

Corresponding Author

Ar

*[email protected]

N N

Ar N N

O

O

N Ph

TSendo

ORCID O

Manabu Abe: 0000-0002-2013-4394

8endo

Author Contributions The manuscript was written through contributions of all authors. / All authors have given approval to the final version of the manuscript.

G‡rel at 298 K in kJ mol–1 7 (Ar)

at (U)CAM-B3LYP/6-31G(d)

gas phase

acetonitrile (PCM)

a (p-HC6H4)

0.0

+1.3

b (p-OMeC6H4)

–0.3

–0.6

e (p-ClC6H4)

+2.0

+3.5

Page 12 of 15

ACKNOWLEDGMENT M.A. gratefully acknowledges financial support by JSPS KAKENHI (Grant No. JP17H03022).

REFERENCES

G‡rel

: relative to Gibbs energy difference (G‡) beteen 7a and TSaendo

(1) Chambers, T. S.; Kistiakowsky, G. B. Kinetics of the Thermal Isomerization of Cyclopropane. J. Am. Chem. Soc. 1934, 56, 399–405. (2) (a) Berson, J. A.; Pedersen, L. D.; Carpenter B. K. Thermal Stereomutation of Cyclopropanes, J. Am. Chem. Soc. 1976, 98, 122–143. (b) Berson, J. A. A New Class of Non-Kekulé Molecules with Tunable Singlet−Triplet Energy Spacings. Acc. Chem. Res. 1997, 30, 238-244.

CONCLUSIONS In the present study, the characteristics and reactivity of a new series of localized singlet diradical, i.e. singlet 4,4-

12 ACS Paragon Plus Environment

Page 13 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

(3) Hoffmann, R. Trimethylene and the addition of methylene to ethylene. J. Am. Chem. Soc. 1968, 90, 1475–1485. (4) Salem, L.; Rowland, C. The Electronic Properties of Diradicals. Angew. Chem., Int. Ed. Engl. 1972, 11, 92–111. (5) Borden, W. T. In Diradicals; Borden, W. T., Eds.; WileyInterscience: New York, 1982; pp 1. (6) (a) Abe, M. Diradicals. Chem. Rev. 2013, 113, 7011–7088. (b)McClelland, R. A. In Reactive Intermediate Chemistry; Moss, R. A., Platz, M. S., Jones Jr. M., Eds; Wiley & Sons: Hoboken, New Jersey, 2004; pp 3–40. (7) (a) Amyes, T. L.; Toteva, M. M.; Richard, J. P. In Reactive Intermediate Chemistry; Moss, R. A., Platz, M. S., Jones Jr. M., Eds; Wiley & Sons: Hoboken, New Jersey, 2004; pp 41–68. (b) Gronert, S. In Reactive Intermediate Chemistry; Moss, R. A., Platz, M. S., Jones Jr. M., Eds; Wiley & Sons: Hoboken, New Jersey, 2004; pp 69–120. (8) (a) Mayr, H.; Patz, M. Scales of Nucleophilicity and Electrophilicity : A System for Ordering Polar Organic and Organometallic Reactions. Angew. Chem., Int. Ed. Engl. 1994, 33, 938−957. (b) For a comprehensive database of nucleophilicity parameters N and sN as well as electrophilicity parameters E, see: http://www.cup.lmu.de/oc/mayr/DBintro.html. (9) (a) Hoffmann, R. Trimethylene and the Addition of Methylene to Ethylene. J. Am. Chem. Soc. 1968, 90, 1475–1485. (b) Hoffmann, R. Interaction of Orbitals Through Space and Through Bonds. Acc. Chem. Res. 1971, 4, 1–9. (c) Getty, S. J.; Hrovat, D. A.; Borden, W. T. Ab Initio Calculations on the Stereomutation of 1,1-Difluorocyclopropane. Prediction of a Substantial Preference for Coupled Disrotation of the Methylene Groups. J. Am. Chem. Soc. 1994, 116, 1521–1527. (10) Adam, W.; Fragale, G.; Klapstein, D.; Nau, W. M.; Wirz, J. Phosphorescence and Transient Absorption of Azoalkane Triplet States. J. Am. Chem. Soc. 1995, 117, 12578–12592. (11) Kita, F.; Adam, W.; Jordan, P.; Nau, W. M.; Wirz, J. 1,3Cyclopentanediyl Diradicals:  Substituent and Temperature Dependence of Triplet−Singlet Intersystem Crossing. J. Am. Chem. Soc. 1999, 121, 9265–9275. (12) Lineberger, W. C.; Borden, W. T. The synergy between qualitative theory, quantitative calculations, and direct experiments in understanding, calculating, and measuring the energy differences between the lowest singlet and triplet states of organic diradicals . Phys. Chem. Chem. Phys. 2011, 13, 11792–11813. (13) Abe, M.; Ishihara, C.; Nojima, M. DFT Prediction of GroundState Spin Multiplicity of Cyclobutane-1,3-diyls:  Notable Effects of Two Sets of Through-Bond Interactions. J. Org. Chem. 2003, 68, 1618– 1621. (14) Abe, M.; Kawanami, S.; Ishihara, C.; Nojima, M. 2-Silyl Group Effect on the Reactivity of Cyclopentane-1,3-diyls. Intramolecular Ring-Closure versus Silyl Migration. J. Org. Chem. 2004, 69, 5622– 5626. (15) Abe, M.; Je, J.; Mishima, M. The chemistry of localized singlet 1,3-diradicals (biradicals): from putative intermediates to persistent species and unusual molecules with a π-single bonded character. Chem. Soc. Rev. 2012, 41, 3808–3820. (16) Abe, M.; Akisaka, R. Is π-Single Bonding (C–π–C) Possible? A Challenge in Organic Chemistry. Chem. Lett. 2017, 46, 1586–1592. (17) Ishihara, C.; Abe, M. Notable Substituent Effects on the Rate Constant of Thermal Denitrogenation of Cyclic Azoalkanes: Strong Evidence for a Stepwise Denitrogenation Mechanism. Aust. J. Chem. 2010, 63, 1615–1618. (18) Abe, M.; Adam, W.; Nau, W. M. Photochemical Generation and Methanol Trapping of Localized 1,3 and 1,4 Singlet Diradicals Derived from a Spiroepoxy-Substituted Cyclopentane-1,3-diyl. J. Am. Chem. Soc. 1998, 120, 11304–11310. (19) Abe, M.; Adam, W.; Heidenfelder, T.; Nau, W. M.; Zhang, X. Intramolecular and Intermolecular Reactivity of Localized Singlet Diradicals: The Exceedingly Long-Lived 2,2-Diethoxy-1,3diphenylcyclopentane-1,3-diyl. J. Am. Chem. Soc. 2000, 122, 2019– 2026. (20) Ye, J.; Hatano, S.; Abe, M.; Kishi, R.; Murata, Y.; Nakano, M.; Adam, W. A Puckered Singlet Cyclopentane‐1,3‐diyl: Detection of the Third Isomer in Homolysis. Chem. Eur. J. 2016, 22, 2299–2306. (21) Abe, M.; Tada, S.; Mizuno, T.; Yamasaki, K. Impact of Diradical Spin State (Singlet vs Triplet) and Structure (Puckered vs Planar) on the

Photodenitrogenation Stereoselectivity of 2,3Diazabicyclo[2.2.1]heptanes. J. Phys. Chem. B, 2016, 120, 7217–7226. (22) Abe, M.; Adam, W.; Hara, M.; Hattori, M.; Majima, T.; Nojima, M.; Tachibana, K.; Tojo, S. On the Electronic Character of Localized Singlet 2,2-Dimethoxycyclopentane-1,3-diyl Diradicals:  Substituent Effects on the Lifetime. J. Am. Chem. Soc. 2002, 124, 6540–6541. (23) Abe, M.; Hattori, M.; Takegami, A.; Masuyama, A.; Hayashi, T.; Seki, S.; Tagawa, S. Experimental Probe for Hyperconjugative Resonance Contribution in Stabilizing the Singlet State of 2,2Dialkoxy-1,3-diyls:  Regioselective 1,2-Oxygen Migration. J. Am. Chem. Soc. 2006, 128, 8008–8014. (24) Creary, X. Super Radical Stabilizers. Acc. Chem. Res. 2006, 39, 761–771. (25) Alabugin, I. V.; Gomes, G. dos. P.; Abdo, M. A. Hyperconjugation. WIREs Comput Mol Sci. 2018, e1389. (26) Nakagaki, T.; Sakai, T.; Mizuta, T.; Fujiwara, Y.; Abe, M. Kinetic Stabilization and Reactivity of π Single‐Bonded Species: Effect of the Alkoxy Group on the Lifetime of Singlet 2,2‐Dialkoxy‐1,3‐diphenyloctahydropentalene‐1,3‐diyls. Chem. Eur. J. 2013, 19, 10395–10404. (27) Ye, J.; Fujiwara, Y.; Abe, M. Substituent effect on the energy barrier for σ-bond formation from π-single-bonded species, singlet 2,2dialkoxycyclopentane-1,3-diyls. Beilstein. J. Org. Chem. 2013, 9, 925– 933. (28) Abe, M.; Furunaga, H.; Ma, D.; Gagliardi, L.; Bodwll, G. J. Stretch Effects Induced by Molecular Strain on Weakening σ-Bonds: Molecular Design of Long-Lived Diradicals (Biradicals). J. Org. Chem. 2012, 77, 7612–7619. (29) Harada, Y.; Wang, Z.; Kumashiro, S.; Hatano, S.; Abe, M. Extremely Long Lived Localized Singlet Diradicals in a Macrocyclic Structure: A Case Study on the Stretch Effect. Chem. Eur. J. 2018, 24, 14808–14815. (30) (a) Kamada, K.; Ohta, K.; Kubo, T.; Shimizu, A.; Morita, Y.; Nakasuji, K.; Kishi, R.; Ohta, S.; Furukawa, S.; Takahashi, H.; Nakano, M. Strong Two-Photon Absorption of Singlet Diradical Hydrocarbons. Angew. Chem., Int. Ed. 2007, 46, 3544–3546. (b) Kishi, R.; Murata, Y.; Saito, M.; Morita, K.; Abe, M.; Nakano, M. Theoretical study on diradical characters and nonlinear optical properties of 1, 3-diradical compounds. J. Phys. Chem. A, 2014, 118, 10837–10848. (31) (a) Nakano, M. Excitation Energies and Properties of Open-Shell Singlet Molecules, Springer, 2014. (b) Sun, Z.; Zeng, Z.; Wu, J. Acc. Chem. Res. 2014, 47, 2582. (c) Zeng, Z.; Shi, X.; Chi, C.; López Navarrete, J. T.; Casado, J.; Wu, J. Chem. Soc. Rev. 2015, 44, 6578. (d) Kubo, T. Chem. Lett., 2015, 44, 111. (e) Michl, J. Chem. Rev. 2010, 110, 6891–6936. (f) Hansmann, M. M.; Melaimi, M.; Munz, D.; Bertrand, G. J. Am. Chem. Soc. 2018, 140, 2546–2554. (g) Dressler, J. J.; Teraoka, M.; Espejo, G. L.; Kishi, R.; Takamuku, S.; Gómez-Garcia, C. J.; Zakharov, L. N.; Nakano, M.; Casado, J.; Haley, M. M. Nature Chem. 2018, 10, 1134–1140. (32) Niecke, E.; Fuchs, A.; Baumeister, F.; Nieger, M.; Schoeller, W. W. A P2C2 Four‐Membered Ring with Unusual Bonding—Synthesis, Structure, and Ring Opening of a 1,3‐Diphosphacyclobutane‐2,4‐diyl. Angew. Chem., Int. Ed. 1995, 34, 555–557. (33) Sugiyama, H.; Ito, S.; Yoshifuji, M. Synthesis of a 1,3‐Diphosphacyclobutane‐2,4‐diyl from Mes*C ≡ P. Angew. Chem., Int. Ed. 2003, 42, 3802–3804. (34) (a) Ito, S.; Ueda, Y.; Hgot, T. T. T.; Kobayashi, M.; Hashizume, D.; Nishida, J.-I.; Yamashita, Y.; Mikami, K. Direct Arylations for Study of the Air-Stable P-Heterocyclic Biradical: From Wide Electronic Tuning to Characterization of the Localized Radicalic Electrons. J. Am. Chem. Soc. 2013, 135, 17610–17616. (b) Ito, S.; Ueta, Y.; Koshino, K.; Kojima, K. M.; Mckenzie, I.; Mikami, K. Observation of a Metastable P‐Heterocyclic Radical by Muonium Addition to a 1,3‐Diphosphacyclobutane‐2,4‐diyl. Angew. Chem., Int. Ed. 2018, 57, 8608–8613. (35) (a) Schoeller, W. W.; Rozhenko, A.; Bourissou, D.; Bertrand, G. On the Electronic Structures of the 1,3‐Diboracyclobutane‐1,3‐diyls and Their Valence Isomers with a B2E2 Skeleton (E=N, P, As). Chem. Eur. J. 2003, 9, 3611–3617. (b) Soleilhavoup, M.; Bertrand, G. Let’s Play with Valence Isomers: The Influence of Different Main Group Elements on Their Relative Stability. Bull. Chem. Soc. Jpn. 2007, 80, 1241–1252.

13 ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(c) Breher, F. Stretching Bonds in Main Group Element Compounds— Borderlines between Biradicals and Closed-shell Species. Coord. Chem. Rev. 2007, 251, 1007–1043. (d) Wang, X.; Peng, P.; Olmstead, M. N.; Fettinger, J. C.; Power, P. P. An Unsymmetric Oxo/Imido-Bridged Germanium-Centered Singlet Diradicaloid. J. Am. Chem. Soc. 2009, 131, 14164–14165. (e) Cox, H.; Hitchcock, P. B.; Lappert, M. F.; Pirssens, L. J.-M. A 1,3-Diaza-2,4-distannacyclobutanediide: Synthesis, Structure, and Bonding. Angew. Chem. Int. Ed. 2004, 43, 4500–4504. (f) Demeshko, S.; Godemann, C.; Kuzora, R.; Schulz, A.; Villinger, A. An Arsenic–Nitrogen Biradicaloid: Synthesis, Properties, and Reactivity. Angew. Chem. Int. Ed. 2013, 52, 2105–2108. (g) Henke, P.; Pankewitz, T.; Klopper, W.; Breher, F.; Schnçckel, H. Angew. Chem. Int. Ed. 2009, 48, 8141–8145. (h) Doddi, A.; Gemel, C.; Winter, M.; Fischer, R. A.; Goedecke, C.; Rzepa, H. S.; Frenking, G. Angew. Chem. Int. Ed. 2013, 52, 450–454. (i) Beweries, T.; Kuzora, R.; Rosenthal, E.; Schulz, A.; Villinger, A. [P(m-NTer)]2: A Biradicaloid That Is Stable at High Temperature. Angew. Chem. Int. Ed. 2011, 50, 8974–8978. (j) Takeuchi, K.; Ichinohe, M.; Sekiguchi, A. Access to a Stable Si2N2 Four-Membered Ring with Non-Kekule Singlet Biradical Character from a Disilyne. J. Am. Chem. Soc. 2011, 133, 12478–12481. (36) Scheschkewitz, D.; Amii, H.; Gornitzka, H.; Schoeller, W. W.; Bourissou, D.; Bertrand, G. Singlet Diradicals: from Transition States to Crystalline Compounds. Science 2002, 295, 1880–1881. (37) (a) Amii, H.; Vranicar, L.; Gornitzka, H.; Bourissou, D.; Bertrand, G. Radical-Type Reactivity of the 1,3-Dibora-2,4Diphosphoniocyclobutane-1,3-diyl. J. Am. Chem. Soc. 2004, 126, 1344–1345. (b) Scheschkewitz, D.; Amii, H.; Gornitzizka, H.; Schoeller, W. W.; Bourissou, D.; Bertrand, G. Sigma-bond Stretching: A Static Approach for a Dynamic Process. Angew Chem., Int. Ed. 2004, 43, 585–587. (38) Nakamura, T.; Gagliardi, L.; Abe, M. Computational study of the cooperative effects of nitrogen and silicon atoms on the singlet– triplet energy spacing in 1,3-diradicals and the reactivity of their singlet states. J. Phys. Org. Chem. 2010, 23, 300–307. (39) Yoshidomi, S.; Mishima, M.; Seyama, S.; Abe, M.; Fujiwara, Y.; Ishibashi, T. Direct Detection of a Chemical Equilibrium between a Localized Singlet Diradical and Its σ-Bonded Species by TimeResolved UV/Vis and IR Spectroscopy. Angew. Chem., Int. Ed. 2017, 56, 2984–2988. (40) Abe, M.; Kubo, E.; Nozaki, K.; Matsuo, T.; Hayashi, T. An Extremely Long-Lived Singlet 4,4-Dimethoxy-3,5diphenylpyrazolidine-3,5-diyl Derivative: A Notable Nitrogen-Atom Effect on Intra- and Intermolecular Reactivity. Angew. Chem., Int. Ed. 2006, 45, 7828–7831; Corrigendum, idem, ibid, 2012, 51, 11911. (41) Abe, M.; Adam, W.; Borden, W. T.; Hattori, M.; Hrovat, D. A.; Nojima, M.; Nozaki, K.; Wirz, J. Effects of Spiroconjugation on the Calculated Singlet−Triplet Energy Gap in 2,2-Dialkoxycyclopentane1,3-diyls and on the Experimental Electronic Absorption Spectra of Singlet 1,3-Diphenyl Derivatives. Assignment of the Lowest-Energy Electronic Transition of Singlet Cyclopentane-1,3-diyls. J. Am. Chem. Soc. 2004, 126, 574–582. (42) Brown, H. C.; Okamoto, Y. Electrophilic Substituent Constants. J. Am. Chem. Soc. 1958, 80, 4979–4987. (43) Jeffe, H. H. A Reëxamination of the Hammett Equation. Chem. Rev. 1953, 53, 191–261. (44) Tanner, D. D.; Koppula, S.; Kandanarachchi, Pramod. Mechanism of the Solution-Phase Reaction of Alkyl Sulfides with Atomic Hydrogen. Reduction via a 9-S-3 Radical Intermediate. J. Org. Chem. 1997, 62, 4210–4215. (45) Reichardt, C. Empirical Parameters of Solvent Polarity as Linear Free‐Energy Relationships. Angew. Chem., Int. Ed. 1965, 4, 29–40. (46) Creary, X.; Mehrsheikh-Mohammadi, M. E.; McDonald, S. A comparison of the radical-stabilizing ability of aromatic groups. γ ・ values for aromatic groups. J. Org. Chem. 1989, 54, 2904–2910. (47) Bradley, D.; Williams, G.; Lawton, M. Drying of Organic Solvents: Quantitative Evaluation of the Efficiency of Several Desiccants. J. Org. Chem. 2010, 75, 8351–8354. (48) Einstein, A. Über die von der molekularkinetischen Theorie der Wärme geforderte Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen. Annalen. der. Physik, 1905, 17, 549–560.

Page 14 of 15

(49) Perkins, M. J.; Spin Trapping. In Advances in Physical Organic Chemistry; Gold, V.; Bethell, D., Eds.; Academic Press: New York, 1980; Vol. 17, pp 1–64. (50) Mao, S. W.; Kevan, L. Electron paramagnetic resonance studies of spin trapping of the primary neutral radicals formed in γ-irradiated methanol. Chem. Phys. Lett. 1974, 24, 505–507. (51) Schlick, S.; Kevan, L. Spin trapping of radicals formed in gamma-irradiated methanol: Effect of the irradiation temperature from 77 K to 300 K. Chem. Phys. Lett. 1976, 38, 505–509. (52) Halpern, A. Spin trapping of radicals in tritiated methanol. Chem. Phys. Lett. 1984, 103, 523–525. (53) Acken, B. J.; Warshaw, J. A.; Gallis, D. E.; Crist, D. R. Acyclic α-Alkoxynitrones. A New Class of Spin-Trapping Agents. J. Org. Chem. 1989. 54, 1743–1745. (54) Mizuno, T.; Abe, M.; Ikeda, N. Simultaneous Observation of Triplet and Singlet Cyclopentane-1,3-diyl Diradicals in the Intersystem Crossing Process. Aust. J. Chem. 2015, 68, 1700–1706. (55) Yamaguchi, K.; Jensen, F.; Dorigo, A.; Houk, K. N. A spin correction procedure for unrestricted Hartree-Fock and Møller-Plesset wavefunctions for singlet diradicals and polyradicals. Chem. Phys. Lett. 1988, 149, 537. (56) (a) Becke, A.D. Densityfunctional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648–5652. (b) Lee, C.; Yang, W. ; Parr, R. Development of the Colle-Salvetti correlationenergy formula into a functional of the electron density. G. Phys. Rev. B 1988, 37, 785–789. (57) Yanai, T.; Tew, D. P.; Handy, N. C. A new hybrid exchange– correlation functional using the Coulomb-attenuating method (CAMB3LYP). Chem. Phys. Lett, 2004, 393, 51–57. (58) Zhao, Y.; Truhlar, D. The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 2008, 120, 215−241. (59) Chai, J.-D.; Head-Gordon, M. Systematic optimization of longrange corrected hybrid density functionals. J. Chem. Phys. 2008, 128, 084106. (60) Ditchfield, R.; Hehre, W. J.; Pople, J. A. Self-Consistent Molecular-Orbital Methods. IX. An Extended Gaussian-Type Basis for Molecular-Orbital Studies of Organic Molecules. J. Chem. Phys. 1971, 54, 724–728. (61) (a) Miertuš, S.; Scrocco, E.; Tomasi, J. Electrostatic interaction of a solute with a continuum. A direct utilizaion of AB initio molecular potentials for the prevision of solvent effects. Chem, Phys. 1981, 55, 117–129. (b) Tomasi, J.; Mennucci, B.; Cammi, Roberto. Quantum Mechanical Continuum Solvation Models. Chem. Rev., 2005, 105, 2999–3094. (62) The molecular geometries for the transition states were first estimated by Reaction plus software package [63], based on the nudged elastic band (NEB) method [64], and were subsequently re-optimized using AAA software package. (63) Software to optimize reaction paths along the user's expected ones, HPC Systems Inc., http://www.hpc.co.jp/chem/react2.html (written in Japanese). (64) Jónsson, H.; Mills, G.; Jacobsen, K. W. Nudged Elastic Band Method for Finding Minimum Energy Paths of Transitions, in Classical and Quantum Dynamics in Condensed Phase Simulations, Ed. Berne, B. J.; Ciccotti G.; Coker, D. F., 385 (World Scientific, 1998); Henkelman G.; Jónsson, H. Improved tangent estimate in the nudged elastic band method for finding minimum energy paths and saddle points, J. Chem. Phys. 2000, 113, 9978. (65) Zhang, D. Y.; Hrovat, D. A.; Abe, M.; Borden, W. T. DFT Calculations on the Effects of Para Substituents on the Energy Differences between Singlet and Triplet States of 2,2-Difluoro-1,3diphenylcyclopentane-1,3-diyls. J. Am. Chem. Soc. 2003, 125, 12823.

14 ACS Paragon Plus Environment

Page 15 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Authors are required to submit a graphic entry for the Table of Contents (TOC) that, in conjunction with the manuscript title, should give the reader a representative idea of one of the following: A key structure, reaction, equation, concept, or theorem, etc., that is discussed in the manuscript. Consult the journal’s Instructions for Authors for TOC graphic specifications. Entropy Control

Enthalpy Control

Triplet

Singlet

ACS Paragon Plus Environment

15