Quinoline Triradicals: A Reactivity Study - American Chemical Society

Lucas Szalwinski, Duanchen Ding, John J. Nash, and Hilkka I. Kenttämaa*. Department of Chemistry, Purdue University, 560 Oval Drive,. West Lafayette ...
0 downloads 0 Views 352KB Size
Subscriber access provided by UNIV OF LOUISIANA

Article

Quinoline Triradicals: A Reactivity Study Raghavendhar R. Kotha, Ravikiran Yerabolu, Mohammad Sabir Aqueel, James S. Riedeman, Lucas Szalwinski, Duanchen Ding, John J. Nash, and Hilkka I. Kenttamaa J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b01740 • Publication Date (Web): 04 Apr 2019 Downloaded from http://pubs.acs.org on April 4, 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 26 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

Quinoline Triradicals: A Reactivity Study

Raghavendhar R. Kotha, Ravikiran Yerabolu, Mohammad Sabir Aqueel, James S. Riedeman, Lucas Szalwinski, Duanchen Ding, John J. Nash, and Hilkka I. Kenttämaa*

Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana, 47907

*Corresponding author: Professor Hilkka I. Kenttämaa, Email: [email protected], Telephone: +1 (765) 494-0882.

1 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

Abstract The gas-phase reactivities of several protonated quinoline-based σ-type (carbon-centered) mono-, bi- and triradicals towards dimethyl disulfide (DMDS) were studied by using a linear quadrupole ion trap mass spectrometer. The mono- and biradicals produce abundant thiomethyl abstraction products and small amounts of DMDS radical cation, as expected. Surprisingly, all triradicals produce very abundant DMDS radical cations. A single-step mechanism involving electron transfer from DMDS to the triradicals is highly unlikely because the (experimental) adiabatic ionization energy of DMDS is almost 3 eV greater than the (calculated) adiabatic electron affinities of the triradicals. The unexpected reactivity can be explained based on an unprecedented two-step mechanism wherein the protonated triradical first transfers a proton to DMDS, which is then followed by hydrogen atom abstraction from the protonated sulfur atom in DMDS by the radical site in the benzene ring of the deprotonated triradical to generate the conventional DMDS radical cation and a neutral biradical. Quantum chemical calculations as well as examination of deuterated and methylated triradicals provide support for this mechanism. The proton affinities of the neutral triradicals (and DMDS) influence the first step of the reaction while the vertical electron affinities and spin-spin coupling of the neutral triradicals influence the second step. The calculated total reaction exothermicities for the triradicals studied range from 27.6 up to 29.9 kcal mol-1.

2 ACS Paragon Plus Environment

Page 2 of 26

Page 3 of 26 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

Introduction Aromatic carbon-centered σ-type biradicals, such as the ortho-, meta- and para-benzynes, have applications in many areas, including organic synthesis and drug design.1–4 Hence, numerous theoretical and experimental studies have been carried out in order to improve the understanding of their reactivity.5-9 The chemical properties of such positively-charged biradicals in the gas phase are currently best understood. They have been found to be influenced by the vertical electron affinity,6 hydrogen bonding capability10 and the magnitude of singlet-triplet splitting6,11-14 of the charged biradical, as well as by the distortion energy6,14 for meta-benzynes. However, only limited experimental gas-phase data are available for related triradicals while condensed-phase data are nearly nonexistent due to difficulties in generating the triradicals in condensed phases.11,12,15-21 A few theoretical studies have been carried out to better understand their physical and chemical properties.22,23 As triradicals contain three electrons in nonbonding molecular orbitals that are similar in energy, these species contain several low-lying electronic states. Their complex electronic structures suggest interesting chemical properties. Better knowledge on the factors that control their reactions, as well as those of their neutral analogs, is critically important for their rational utilization in organic synthesis, development of new organic materials, and design of more efficient antitumor drugs. Furthermore, this is necessary for improving the fundamental understanding of organic reaction intermediates. In order to continue to improve our understanding on structure-reactivity relationships for aromatic carbon-centered σ-type triradicals, examination of the gas-phase reactivities of seven protonated quinoline-based σ-type radicals, including mono-, bi-, and four isomeric triradicals (Chart 1), towards dimethyl disulfide was carried out. Quantum chemical calculations were utilized to facilitate understanding of the experimental results.

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

5

Page 4 of 26

4

6

3

7

1N

8

H

4

N H

2

1

5

N H

2

N H

6

N H

N H

3

7

N H

N H

Chart 1. Numbering scheme and the mono-, bi- and tridehydroquinolinium cations studied. Experimental Section Most chemicals were obtained from commercial sources and used as received. The radical precursors 2-iodoquinoline, 4-iodoquinoline, 2,4-diiodoquinoline, 2,4,5-triiodoquinoline, 2,4,6triiodoquinoline, 2,4,7-triiodoquinoline, and 2,4,8-triiodoquinoline for radicals 1-7 have been synthesized previously.13 A linear quadrupole ion trap mass spectrometer (LQIT), equipped with a manifold for reagent introduction, was used for the gas-phase reactivity experiments as described in the literature.13,19 Solutions of all radical precursors were introduced into the LQIT and protonated by using atmospheric pressure chemical ionization (APCI). Collision-activated dissociation (CAD) of protonated radical precursors (using helium as a collision gas) was used to cleave C-I bonds and to generate radicals 1-7. Each CAD event generated one radical site; hence, the appropriate number of CAD events (i.e., one CAD for mono-, two for bi-, and three for triradicals) was used to generate the desired radicals. After generation, each radical was isolated by ejecting all unwanted ions from the ion trap and its reaction products and efficiencies upon interaction with DMDS were determined. The reaction efficiency is the fraction of ion-molecule collisions that leads to reaction and is given as kexp/kcoll, where kexp is the measured experimental rate constant and kcoll is the theoretical collision rate constant calculated by using a parameterized trajectory theory.10,13,19 4 ACS Paragon Plus Environment

Page 5 of 26 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

The N-deuterated 2,4,7-tridehydroquinolinium cation was generated as follows. 2,4,7Triiodoquinoline (1 mg) was dissolved in methanol-d4 (1 mL) and injected into the APCI source at a flow rate of 10 mL/min to generate the N-deuterated 2,4,7-triiodoquinolinium cation. Upon three ion trap CAD experiments, the N-deuterated 2,4,7-tridehydroquinolinium cation was generated. The N-methylated 2,4,7-tridehydroquinolinium cation was generated as follows. 2,4,7Triiodoquinoline (1 mg) was dissolved in 20/80 (v/v) iodomethane/carbon disulfide (1 mL) solution and injected into the APCI source at a flow rate of 10 mL/min to generate N-methylated 2,4,7-triiodoquinolinium cation. Upon three ion trap CAD experiments, the N-methylated 2,4,7tridehydroquinolinium cation was generated. Computational Methods Molecular geometries for all species were optimized at the density functional (DFT) level of theory by using the correlation-consistent polarized valence-triple- (cc-pVTZ) basis set.24 The DFT calculations used the gradient-corrected exchange functional of Becke,25 which was combined with the gradient-corrected correlation functional of Lee, Yang and Parr26 (B3LYP). Except for those structures in which the dehydrocarbon atom separation(s) (DAS) was held constant (see below), all DFT geometries were verified to be local minima by computation of analytic vibrational frequencies, and these (unscaled) frequencies were used to compute zero-point vibrational energies (ZPVE) and 298 K thermal contributions (H298 – E0). All DFT calculations for the mono-, bi-, and triradicals employed an unrestricted formalism. To improve the molecular orbital calculations, dynamic electron correlation was also accounted for by using multi-reference second-order perturbation theor27,28 (CASPT2) for multiconfigurational self-consistent field (MCSCF) reference wave functions; these calculations were carried out for the DFT optimized geometries. The MCSCF calculations were of the complete 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

active space (CASSCF) variety29 and included (in the active space) the full -space for each molecule and, for each of the mono-, bi-, and triradicals, the nonbonding  orbital(s). Some caution must be applied in interpreting the CASPT2 results since this level of theory is known to suffer from a systematic error proportional to the number of unpaired electrons.30 Thus, the electronic energies are of the CASPT2/CASSCF(m,n)/cc-pVTZ//UB3LYP/cc-pVTZ variety (where m is the number of active electrons and n is the number of active orbitals), and estimates of the thermodynamic quantities, E0 and H298, are derived by adding to these electronic energies ZPVE and the sum of ZPVE and (H298 – E0), respectively, where the latter are derived from the DFT calculations. In order to compute vertical electron affinities (EAv) for the mono-, bi-, and triradicals, single-point calculations (CASPT2/CASSCF(m,n)/cc-pVTZ), using the UB3LYP/cc-pVTZ optimized geometry for each radical, were also carried out for the states that are produced when a single electron is added to the nonbonding  orbital (or one of the two, or three such orbitals) of each molecule.31 Thus, for the monoradicals and triradicals (doublet ground states) these calculations were carried out for (zwitterionic) singlet states, whereas for the biradical (singlet ground state) calculations were carried out for (zwitterionic) doublet states.32 Adiabatic electron affinities (EAa) for the 2,4,5-, 2,4,6- and 2,4,8-tridehydroquinolinium cations were calculated in a similar manner except that the geometry of the (zwitterionic) singlet state was optimized (UB3LYP/cc-pVTZ) in each case. In addition, the calculated EAa values include the zero-point and thermal corrections obtained from the (unscaled) UB3LYP/cc-pVTZ frequencies. For the potential energy surfaces shown in Figures 1, 2 and 4, a series of DFT calculations was carried out in which the C-2/C-4 dehydrocarbon atom separation (DAS) was held constant and systematically varied in increments of 0.1 Å from 1.3 to 2.3 Å (all other geometric parameters

6 ACS Paragon Plus Environment

Page 6 of 26

Page 7 of 26 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

were optimized for each structure). For each structure, a single-point calculation (CASPT2/CASSCF(m,n)/cc-pVTZ) was then performed using the UB3LYP/cc-pVTZ optimized geometry. Because each of these geometries was obtained by using a fixed DAS, they are not true local minima, and, as a result, zero-point and thermal corrections were not applied to the computed CASPT2 energies. All CASPT2/MCSCF and DFT calculations were carried out with the MOLCAS 8.033 and Gaussian 0934 electronic structure program suites, respectively. Results and Discussion The mono- and biradicals studied here (1-3; Chart 1) react with dimethyl disulfide (DMDS) by predominant abstraction of one or two thiomethyl groups (Table 1). This was expected as previous studies have demonstrated that thiomethyl abstraction is a dominant reaction for many related charged mono- and biradicals (distonic ions).5,11,12,35 Only a small amount of DMDS radical cation (1-14% relative abundance; Table 1) was observed. In sharp contrast, the four isomeric triradicals studied (4-7; Table 1) produce the DMDS radical cation as the major product (83-94% relative abundance; Table 1). This finding was entirely unexpected as the experimentally determined36 adiabatic ionization energy (IEa) of DMDS (8.2 ± 0.2 eV) is 2.0-2.7 eV greater than the (calculated) adiabatic electron affinities (EAa) of the triradicals studied here (for example, the calculated EAa values of the 2,4,5-, 2,4,7- and 2,4,8-tridehydroquinolinium cations at the 5-, 7and 8-radical sites are 5.47, 5.54 and 6.17 eV; in comparison, the vertical EAs (EAv) are 5.19, 4.92 and 5.61 eV, respectively; Table 2). These electron transfer reactions are much too endothermic to be observed under the experimental conditions used here. More exothermic, dissociative electron transfer has been reported35 for some radical cations upon interaction with DMDS, as for example for ●CH2CH2C=O+ that generates abundant DMDS radical cation as well as ethylene and carbon 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

monoxide upon dissociative electron transfer, but this type of reaction cannot take place for the triradicals discussed here. The formation of the DMDS radical cation for the triradicals might be explained by a twostep mechanism wherein the protonated quinoline-based triradical (or mono- or biradical) first transfers a proton to DMDS, which is then followed by hydrogen atom abstraction from protonated DMDS by the neutral (deprotonated) radical (Scheme 1, for example). This mechanism was probed by examining the reactivity of an analogous N-methylated triradical, the N-methyl-2,4,7tridehydroquinolinium cation, toward DMDS. Only a minor DMDS radical cation (5% relative abundance; thiomethyl abstraction: 92%; addition: 3%; total reaction efficiency 53%) was generated in this reaction, in support of above mechanism. Above mechanism suggests that the proton affinities of the neutral triradical (or mono- or biradical) and DMDS influence the first step of the reaction whereas the EAv of the neutral triradical (or mono- or biradical) influences the second step.5,6 The proposed two-step mechanism (Scheme 1) was explored for all radicals (1-7) by using quantum chemical calculations. First, proton affinities (PA) for deprotonated 1-7 were computed by using an isodesmic reaction involving proton transfer to quinoline (Scheme 2, for example), which has a known PA.37 The calculated PAs, as well as several other (calculated) thermochemical properties, for 1-7 are given in Table 2.

8 ACS Paragon Plus Environment

Page 8 of 26

Page 9 of 26 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

Table 1. Reaction Efficienciesa and Product Branching Ratiosb,c for Reactions of Radicals 1-7 with Dimethyl Disulfide

1 branching Ratios

N H

2

SCH3 abs H abs e¯ abs

86% 7% 6%

N H

SCH3 abs e¯abs

3 99% 1%

N H

SCH3 abs 86% (2 o) SCH3 abs e¯ abs 14%

reaction efficiency

99%

80%

15%

% electron transferd

6%

1%

2%

4 branching Ratios reaction efficiency

e¯ abs SSCH3 abs SCH3 abs

5 94% 4% 2%

76%

% electron transferd

71%

7 e¯ abs branching ratios

N H

SCH3 abs

reaction efficiency

80%

% electron transferd

74%

N H

e¯ abs SCH3 abs (2 o) SCH3 abs SSCH3 abs 63% 54%

6 85% 13% 2%

N H

e¯ abs

83% SCH3 abs 16% (2 o) SCH3 abs SSCH3 abs 1% 69%

57%

N H 93% 7%

aReaction

efficiency (% of collisions leading to reaction) = kreaction/kcollision x 100; precision + 10%; accuracy + 50%. babs = abstraction. cSecondary products are listed below the primary product that produced them and are indicated by 2o. d% Electron transfer was determined by multiplying the reaction efficiency with the banching ratio for electron transfer.

9 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 10 of 26

proton transfer

5

N+ H

controlled by PA of neutral radical and DMDS S S

CH3

N H + CH3 S S H 3C

H 3C

/

controlled by EAv of neutral radical

hydrogen atom abstraction H

+

N

+ CH3 S S H 3C

DMDS radical cation (m/z 94)

Scheme 1. Proposed two-step mechanism for the formation of the DMDS radical cation (m/z 94) for triradical 5. Note that a similar mechanism would apply for radicals 1-4 and 6 and 7.

H298

+ 1

N + H

N

+ N

N + H

quinoline (PAexp = 227.8 kcal mol-1)

Scheme 2. Isodesmic reaction used to calculate the proton affinity of a dehydroquinoline. The calculated PAs for the deprotonated monoradicals 1 and 2 are 217.1 and 225.0 kcal mol-1, respectively (Table 2). These PAs are 22.2 and 30.1 kcal mol-1, respectively, higher than the PA of DMDS (PAexp = 194.9 kcal mol-1).37 Thus, ca. 22 and 30 kcal mol-1 of energy is required for DMDS to abstract a proton from monoradicals 1 and 2, respectively In general, ca. 15 – 25 kcal mol-1 of internal energy is available in gas-phase ion-molecule collision complexes due to 10 ACS Paragon Plus Environment

Page 11 of 26 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

solvation of the ion by the neutral reagent molecule (e.g., DMDS).38 Therefore, proton transfer from 2 to DMDS appears too endothermic to take place under the conditions employed here while proton transfer from 1 may be feasible. However, the following hydrogen atom transfer reaction is not as feasible for 1 as for 2. The rates of these reactions are controlled by the vertical electron affinities of the deprotonated monoradicals.5,6 The calculated vertical electron affinities (EAv) for the deprotonated monoradicals 1 and 2 are 0.13 and 0.96 eV, respectively. The higher EAv for deprotonated 2 (compared to deprotonated 1) suggests that deprotonated 2 should undergo hydrogen atom abstraction from protonated DMDS faster than deprotonated 1. However, monoradical 2 produces only 1% DMDS radical cation compared to 6% for monoradical 1 (Table 1). Thus, it appears that 1 (whose conjugate base has the lower PA) transfers a proton to DMDS to a greater extent than 2, even though the second step (i.e., hydrogen atom abstraction from protonated DMDS; Scheme 1) is slower for 1 due to its lower EAv. Deprotonated biradical 3 has a lower PA (214.9 kcal mol-1) than deprotonated monoradical 1 (PA: 217.1 kcal mol-1; Table 2) although it is still 20 kcal mol-1 greater than that of DMDS. Thus, based on the PAs of the deprotonated radicals, biradical 3 should transfer a proton to DMDS to a greater extent than monoradical 1. However, 3 produces only 2% DMDS radical cation compared to 6% for monoradical 1 (Table 1). The lower production of DMDS radical cation for 3 (compared to 1) must then be a result of a slower second step (i.e., hydrogen atom abstraction from protonated DMDS) for deprotonated 3. The EAv calculated for deprotonated 3 with a DAS of 2.30 Å (0.50 eV) is substantially greater than that for deprotonated monoradical 1 (0.13 eV).

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

Page 12 of 26

Table 2. Calculateda C-2/C-4 Dehydrocarbon Atom Separations At Minimum Energy Geometries (2,4-DAS), Singlet-Triplet Splittings (ΔES-T),b Doublet-Quartet Splittings (ΔED-Q),c Distortion Energies to DAS of 2.30 Å (ΔE2.30),d Proton Affinities (PA) of Their Conjugate Bases,e and Vertical Electron Affinities at Minimum Energy Geometries (EAv) and at 2.30 Å DAS (EAv,2.30) for Radicals 1-7

1

2,4-DAS, Å ΔES-T,f kcal mol-1 ΔED-Q,f kcal mol-1 ΔE2.30, kcal mol-1 PA,f kcal mol-1 EAv, eV EAv,2.30, eV

N+ H

217.1 6.30 -

. 2

.

.

N+ H

N+ H

225.0 5.59 -

3

.

.. 4

1.443 -26.4g 8.5 214.9 6.21

N+ H

1.440 -26.2h 8.7 211.3 5.19 6.35

aAll

. . . . . . . . . . 5

N+ H

1.443 -26.5h 8.7 212.2 4.71 6.35

6

N+ H

1.442 -27.1h 9.1 212.6 4.92 6.38

7

1.443 -26.9h 7.7 212.6 5.61 6.77

values calculated at the CASPT2/CASSCF(m,n)/cc-pVTZ//UB3LYP/cc-pVTZ level of theory. bFor the biradical only. cFor the triradicals only. dThe energy required to increase the C-2/C-4 dehydrocarbon atom separation from the minimum energy geometry (2,4-DAS) to 2.30 Å. ePAs calculated by using an isodesmic equation involving proton transfer to quinoline. f Corrected for zero-point vibrational energy differences at 298K by using the UB3LYP/cc-pVTZ frequencies. gSinglet ground state. hDoublet ground state.

12 ACS Paragon Plus Environment

N+ H

Page 13 of 26

The larger EAv for deprotonated 3 compared to deprotonated 1 should permit faster hydrogen atom abstraction by the former. The observation that 3 produces less DMDS radical cation than 1 suggests that the strong spin-spin coupling between the unpaired electrons (indicated by the relatively large ΔES-T of -26.7 kcal mol-1 for deprotonated 3) is responsible for the smaller yield of DMDS radical cation. This is likely due to the need to partially uncouple the electrons in the transition state of a radical reaction,9 which reduces the rate of the hydrogen atom abstraction reaction from protonated DMDS. There is an alternative view for biradical 3 that is based on the fact that this molecule has a relatively flat potential energy surface associated with the C-2/C-4 dehydrocarbon atom separation (DAS; Figure 1). The energy (“distortion energy”, or ΔE2.30) required for this molecule

relative energy, kcal mol-1

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

12.0 10.0 8.0 6.0

biradical 3

4.0

neutral biradical 3

2.0 0.0 1.3

1.5

1.7

1.9

2.1

2.3

dehydrocarbon atom separation, Å Figure 1. Relative energy versus dehydrocarbon atom separation (DAS) for biradical 3 and neutral (deprotonated) biradical 3 (calculated at the CASPT2/CASSCF(12,12)/cc-pVTZ//UB3LYP/ccpVTZ level of theory). The relative energy at a DAS of 2.3 Å corresponds to ΔE2.30. The (calculated) EAv at this geometry corresponds to EA2.30.

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

to distort from the minimum energy geometry (2,4-DAS: 1.443 Å; Table 2) to a DAS of 2.30 Å (a 2.30 Å DAS is characteristic of transition state geometries for hydrogen atom abstraction by charged meta-benzynes6) is only 8.5 kcal mol-1 (Table 2), which should be well within the range of available solvation energy in the ion-molecule collision complex (see above). In order to determine whether or not the DAS affects the PAs of the neutral (deprotonated) bi- and triradicals, PAs were also computed at a (fixed) DAS of 2.30 Å (PA2.30; Table 3) for the neutral (deprotonated) radicals 3-7. Interestingly, the PA2.30 calculated for deprotonated biradical 3 (206.9 kcal mol-1; Table 3) is 8.0 kcal mol-1 lower at a DAS of 2.30 Å than at its minimum energy geometry (2,4DAS: 1.443 Å; PA: 214.9 kcal mol-1; Table 2). The PAs calculated for deprotonated triradicals 4-7 (discussed in detail below) are also much lower at a DAS of 2.30 Å than at their minimum energy geometries (by 7.1 – 8.5 kcal mol-1). The solvation energy (ca. 15 – 25 kcal mol-1; see above) associated with the 3-DMDS ion-molecule collision complex is more than enough to permit 3 to achieve a geometry with a DAS of 2.30 Å and thereby a lower PA for its deprotonated form, which may lead to a greater extent of proton transfer to DMDS than for 3 in its minimum energy geometry (note that PA2.30 is still ca. 12 kcal mol-1 higher than that for DMDS). In addition, the calculated distortion energy (ΔE2.30) for deprotonated 3 is only 1.4 kcal mol-1 (Figure 1), which indicates that this molecule should be able to achieve the transition state geometry (i.e., DAS of 2.30 Å) for a radical reaction relatively easily. However, as described above, the relatively large ΔES-T (-26.7 kcal mol-1) for deprotonated 3 is still likely to slow substantially the hydrogen atom abstraction from protonated DMDS to produce DMDS radical cation. Next, the fast generation of DMDS radical cations by the isomeric triradicals (4-7) is discussed. The PAs calculated for deprotonated 4-7 (Table 2) are 16.4 – 17.7 kcal mol-1 greater than that for DMDS (194.9 kcal mol-1). Thus, proton transfer from the protonated triradicals to

14 ACS Paragon Plus Environment

Page 14 of 26

Page 15 of 26 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

Table 3. Calculateda Proton Affinities (PA2.30) for the Deprotonated Radicals 3-7 at a C2/C-4 Dehydrocarbon Atom Separation of 2.30 Å

. N 3

PA2.30, kcal

mol-1

206.9

. . .. . . . . . . . . . N

N

N

4

203.3

5

6

204.0

204.1

N

7

205.5

aCalculated

at the CASPT2/CASSCF(m,n)/cc-pVTZ//UB3LYP/cc-pVTZ level of theory. Proton affinities calculated by using an isodesmic equation involving proton transfer to quinoline.

DMDS may take place within a gas-phase ion-molecule collision complex due to solvation energy. Furthermore, the PAs calculated for deprotonated 4-7 at a DAS of 2.30 Å are substantially lower Table 3) and are only 8.4 – 10.6 kcal mol-1 greater than the PA of DMDS. Like biradical 3, all four protonated triradicals have relatively flat (and nearly identical) potential energy surfaces associated with the C-2/C-4 dehydrocarbon atom separation (DAS; Figure 2). The distortion energies (ΔE2.30) for the C-2/C-4 moieties in triradicals 4-7 are 8.7, 8.7, 9.1 and 7.7 kcal mol-1, respectively (Table 2). The fact that triradical 7 has the lowest calculated distortion energy can be explained by the relatively strong spin-spin coupling between the radical sites at C-4 and C-8, which reduces the spin-spin coupling between the radical sites at C-4 and C-2. All of these distortion energies lie well within the range of available solvation energy in the ion-molecule collision complexes (see above) for the triradicals and should therefore lead to a greater extent of proton transfer to DMDS compared to the related mono- and biradicals. The total endothermicities (including ΔE2.30) for proton transfer from the triradicals 4-7 with DAS of 2.30 Å to DMDS are 17.1, 17.8, 18.3 and 18.3 kcal mol-1, respectively. These values are similar to those determined for proton transfer directly from the minimum energy geometries for 4-7 (16.4 – 17.7 kcal mol-1). 15 ACS Paragon Plus Environment

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

relative energy, kcal mol-1

Journal of the American Chemical Society

Page 16 of 26

10.0 8.0 6.0

triradical 4

4.0

triradical 5

2.0

triradical 6

0.0

triradical 7

1.3

1.5

1.7

1.9

2.1

2.3

dehydrocarbon atom separation, Å Figure 2. Relative energy versus dehydrocarbon atom separation (2,4-DAS) for triradicals 4-7 (calculated at the CASPT2/CASSCF(13,13)/cc-pVTZ//UB3LYP/cc-pVTZ level of theory). The relative energy at a 2,4-DAS of 2.3 Å corresponds to ΔE2.30. The (calculated) EAv at this geometry corresponds to EA2.30.

Therefore, proton transfer may occur via either or both mechanisms. The subsequent hydrogen atom abstraction from protonated DMDS by the deprotonated triradicals (to produce the DMDS radical cation) could potentially occur at any one of the three radical sites. However, the spin-spin coupling between the unpaired electrons at C-2 and C-4 is strong (this is a meta-benzyne moiety; for deprotonated biradical 3, ΔES-T is -26.7 kcal mol-1), which should substantially reduce the rate of hydrogen atom abstraction at either one of these two radical sites. On the other hand, the third radical site in the benzene ring of all four triradicals is only weakly coupled to the radical sites at C-2 or C-4. For example, the calculated (CASPT2/CASSCF(12,12)/cc-pVDZ//CASSCF(12,12)/cc-pVDZ) ΔES-T values39 for the 2,5-, 2,6, 2,7- and 2,8-didehydroquinolinium cations are 0.3, -0.5, -2.1 and -0.4 kcal mol-1, respectively, and for the 4,5-, 4,6-, 4,7- and 4,8-didehydroquinolinium cations, the ΔES-T values39 are -0.4, -0.8, 16 ACS Paragon Plus Environment

Page 17 of 26 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

0.3 and -7.2 kcal mol-1, respectively. The corresponding ΔES-T values for the neutral (deprotonated) biradicals are similar (-1.3, -0.7, -1.6 and -1.8 for the 2,5-, 2,6-, 2,7- and 2,8-didehydroquinolines, respectively, and -1.6, -0.8, -0.2 and -7.9 for the 4,5-, 4,6-, 4,7- and 4,8-didehydroquinolines, respectively; calculated at the same level of theory as for the protonated species). Thus, the radical site in the benzene ring is likely to be the site that undergoes hydrogen atom abstraction fastest. The calculated EAv,2.30 values for the deprotonated triradicals 4-7 are: 1.03, 0.96, 1.03 and 1.39 eV, respectively. An examination of the highest occupied molecular orbitals for the states that are produced when a single electron is added to one of the three nonbonding  orbitals of each triradical (i.e., zwitterionic singlet states) shows that the electron density is largely localized on the radical site in the benzene ring. Thus, the EAv,2.30 values given above are for this particular radical site in each triradical (note that the EAv,2.30 for deprotonated 7 is ca. 0.4 eV greater than that for deprotonated 4-6 due to the proximity of the radical site at C-8 to the electronegative nitrogen atom). Calculated enthalpies of reaction for hydrogen atom abstraction from protonated DMDS (as shown in Scheme 1) by each radical site in each deprotonated triradical to produce a neutral biradical and DMDS radical cation (Figure 3) also suggest that the radical site in the benzene ring is the one involved in hydrogen atom abstraction. For all four deprotonated triradicals, the enthalpy of reaction is most exothermic for this particular radical site (Figure 3). Based on the calculated endothermicities of the proton transfer reactions (Table 2; PA of DMDS is 194.9 kcal mol-1) and the calculated exothermicities of the hydrogen atom abstraction reactions (Figure 3), the total reaction exothermicities for 4 – 7 are 29.9, 28.7, 27.6 and 28.0 kcal mol-1, respectively. A comparison of the calculated enthalpy changes and products for the direct electron transfer reaction and the proposed stepwise mechanism for 4 is shown in Scheme 3.

17 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

N+ H

+

S S

CH3

+62.3 kcal mol-1

Page 18 of 26

+ N+ H

H 3C

+ CH3 HS S H 3C

4

+ N+ H

S S

CH3

-29.9 kcal mol-1

H + CH3 S S

+ N

H 3C

H 3C

4

Scheme 3. Reaction products and calculated enthalphy changes for direct electron transfer (top) and the proposed stepwise reaction (bottom) for triradical 4. In order to test whether the ionic product of above reactions is the conventional radical cation of dimethyl disulfide, as proposed (Scheme 1), and not one of two possible distonic isomers, the reaction of the N-deuterated 2,4,7-tridehydroquinolinium cation with DMDS was examined. The observation of the formation of the undeuterated DMDS radical cation (72%; total reaction efficiency 68%) with only minor (4%) deuterated DMDS radical cation demonstrates that the conventional radical cation instead of a distonic ion was predominantly formed (a minor kinetic isotope effect was observed: the undeuterated cation yielded 83% of DMDS radical cations while the deuterated cation yielded only 76%). This is rationalized based on the calculated (UB3LYP/ccpVTZ//UB3LYP/cc-pVTZ level of theory) relative energies of the three isomeric DMDS radical cations as the conventional radical cation lies 21.1 kcal mol-1 lower in energy than its distonic isomer ●CH2S+(H)-SCH3 and 37.2 kcal mol-1 lower in energy than the second distonic isomer ●CH2S-S+(H)-CH3.

Finally, it is noteworthy that the calculated distortion energies (ΔE2.30) for all four deprotonated triradicals are very small and nearly identical (1.5 – 1.8 kcal mol-1; Figure 4) and similar in magnitude to that for deprotonated biradical 3 (1.4 kcal mol-1; Figure 1). Moreover, the DAS potential energy surfaces for the deprotonated triradicals 4-7 and biradical 3 are nearly 18 ACS Paragon Plus Environment

Page 19 of 26 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

indistinguishable (Figure 4), which indicates that the exact location of the radical site in the benzene ring (for the neutral triradicals) has virtually no influence on the 2,4-DAS potential energy profile.

. .

.. N

.. N

Hrxn: -27.1

.

4

Hrxn: -33.5

.

H

N Hrxn: -46.3 H

+ DMDSH+

. N

. .

.

N

Hrxn: -25.7

.

6

.

Hrxn: -32.1

Hrxn: -45.3

+ DMDSH+

H

N

H

. . . .

Hrxn: -26.1 Hrxn: -32.2

.

H

N

5

N Hrxn: -46.0

+ DMDSH+

.

.

H

N

+ DMDS

+ DMDS

.

.

N

H

N

. N

.

H

.

. .

.

Hrxn: -27.5 Hrxn: -30.9

N

7

+ DMDSH+

.

N

H

. .

H

H

. . . . N

Hrxn: -45.7

N H + DMDS

+ DMDS

Figure 3. Calculated (UB3LYP/cc-pVTZ//UB3LYP/cc-pVTZ) enthalpies of reaction (kcal mol-1) for hydrogen atom abstraction reactions by neutral (deprotonated) triradicals 4-7 from protonated DMDS (DMDSH+) to produce neutral biradicals and the DMDS radical cation.

19 ACS Paragon Plus Environment

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

relative energy, kcal mol-1

Journal of the American Chemical Society

Page 20 of 26

30.0 25.0 20.0

neutral biradical 3

15.0

neutral triradical 4

10.0

neutral triradical 5

5.0

neutral triradical 6

0.0

neutral triradical 7

1.3

1.5

1.7

1.9

2.1

2.3

dehydrocarbon atom separation, Å Figure 4. Relative energy versus dehydrocarbon atom separation (DAS) for deprotonated biradical 3 and deprotonated triradicals 4-7 (calculated at the CASPT2/CASSCF(13,13)/ccpVTZ//UB3LYP/cc-pVTZ level of theory). The relative energy at a DAS of 2.3 Å corresponds to ΔE2.30.

Conclusions The surprising discovery that four protonated isomeric quinoline triradicals (4-7) react with dimethyl disulfide (DMDS) to predominantly generate DMDS radical cation is rationalized based on a two-step mechanism evaluated by using both experimental methods and quantum chemical calculations. The data suggest that, in the first step, proton transfer occurs from the protonated triradical to DMDS. In the second step, the radical site in the benzene ring of the neutral triradical abstracts a hydrogen atom from protonated DMDS, which yields the DMDS radical cation. The proton transfer reaction is endothermic but the following hydrogen atom abstraction is highly exothermic, making the total reaction exothermic by 27.6 – 29.9 kcal mol-1. The data further 20 ACS Paragon Plus Environment

Page 21 of 26 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

suggest that the lower the (calculated) proton affinity (PA) and the higher the (calculated) vertical electron affinity (EAv) of the neutral triradical, the greater the yield of DMDS radical cation. The deprotonated triradicals 4-7 have lower PAs and higher EAv than either the deprotonated monoradicals or the deprotonated biradicals (1-3) and, thus, produce significantly greater amounts of DMDS radical cation upon interaction with DMDS. The exact location of the radical site in the benzene ring of the neutral triradicals was found to have virtually no influence on the potential energy profile for the meta-benzyne moiety (i.e., 2,4-dehydrocarbon atom separation). The same is true for the charged triradicals, with one exception. Triradical 7 with radical sites at C-2, C-4 and C-8 has a lower distortion energy than the other triradicals (4-6) due to a relative strong coupling between the radical sites at C-4 and C8.40 This finding is in general agreement with a previous report41 demonstrating that spin-spin coupling between two meta-benzyne moieties in a related charged tetraradical increases their reactivity while spin-spin coupling within each meta-benzyne moiety decreases their reactivity. Clearly, coupling between a meta-benzyne moiety (C-2/C-4) and a third unpaired electron in charged triradicals has a similar albeit smaller effect.

Acknowledgments This material is based upon work supported by the National Science Foundation under Grant Number CHE-1464712.

Supporting Information. Computational details.

21 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

References (1) "m-Benzyne and p-Benzyne", Sander, W. Acc. Chem. Res. 1999, 32, 669–676. (2) "Chemistry and Biology of the Enediyne Anticancer Antibiotics", Nicolaou, K. C.; Dai, W.-M. Angew. Chem. Int. Ed. Engl. 1991, 30, 1387–1416. (3) "Atom Transfer Radical Polymerization", Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921–2990. (4) "One Century of Aryne Chemistry", Wenk, H. H.; Winkler, M.; Sander, W. Angew. Chem. Int. Ed. 2003, 42, 502–528. (5) “Ion-molecule Reactions of Aromatic Carbon-centered Distonic Radical Cations in the Gas Phase”, Williams, P. E.; Jankiewicz, B. J.; Yang, L.; Kenttämaa, H. I. Chem. Rev. 2013, 113, 6949–6985. (6) “On the Factors that Control the Reactivity of meta-Benzynes”, Gao, J.; Jankiewicz, B. J.; Reece, J.; Sheng, H.; Cramer, C. J.; Nash, J. J.; Kenttämaa, H. I. Chem. Sci. 2014, 5, 2205–2215. (7) "A Multiconfigurational SCF and Correlation-consistent CI Study of the Structures, Stabilities, and Singlet-triplet Splittings of o-, m-, and p-Benzyne", Wierschke, S. G.; Nash, J. J.; Squires, R. R. J. Am. Chem. Soc. 1993, 115, 11958–11967. (8) "Ultraviolet Photoelectron Spectroscopy of the o-, m-, and p-Benzyne Negative Ions. Electron Affinities and Singlet−Triplet Splittings for o-, m-, and p-Benzyne", Wenthold, P. G.; Squires, R. R.; Lineberger, W. C. J. Am. Chem. Soc. 1998, 120, 5279–5290. (9) "9,10-Dehydroanthracene: p-Benzyne-Type Biradicals Abstract Hydrogen Unusually Slowly", Schottelius, M. J.; Chen, P. J. Am. Chem. Soc. 1996, 118, 4896–4903. (10) “Effects of Hydrogen Bonding on the Gas-phase Reactivity of Didehydroisoquinolinium Cation Isomers”, Vinueza, N. R.; Jankiewicz, B. J.; Gallardo, V. A.; Nash, J. J.; Kenttämaa, H. I. Phys. Chem. Chem. Phys. 2018, 20, 21567–21572. (11) “Reactivity of an Aromatic σ,σ,σ-Triradical: the 2,4,6-Tridehydropyridinium Cation”, Jankiewicz, B. J.; Adeuya, A.; Yurkovich, M. J.; Vinueza, N. R.; Gardner, S. J.; Zhou, M.; Nash, J. J.; Kenttämaa, H. I. Angew. Chem. Int. Ed. 2007, 46, 9198–9201.

22 ACS Paragon Plus Environment

Page 22 of 26

Page 23 of 26 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

(12) “Reactivity of the 3,4,5-Tridehydropyridinium Cation, an Aromatic σ,σ,σ-Triradical”, Jankiewicz, B. J.; Reece, J. N.; Vinueza, N. R.; Nash, J. J.; Kenttämaa, H. I. Angew. Chem. Int. Ed. 2008, 47, 9860–9865. (13) Kotha, R. R. "Gas-Phase Reactivity Studies of Charged Quinoline Polyradicals by Using Distonic Ion Approach and Linear Quadrupole Ion Trap Mass Spectrometry, and Mechanistic Studies on Metal-Promoted Self-Assembly of Collagen Mimetic Peptides." Ph.D. Dissertation, Purdue University, 2017. (14) “The Reactivity of the 4,5-Didehydroisoquinolinium Cation”, Vinueza, N. R.; Archibold, E. F.; Jankiewicz, B. J.; Gallardo, V. A.; Habicht, S. C.; Aqueel, M. S.; Nash, J. J.; Kenttämaa, H. I. Chem. Eur. J. 2012, 18, 8692–8698. (15) "1,2,3‐Tridehydrobenzene", Venkataramani, S.; Winkler, M.; Sander, W. Angew. Chem. Int. Ed. 2005, 44, 6306–6311. (16) "Is the 1,3,5-Tridehydrobenzene Triradical a Cyclopropenyl Radical Analogue?" Lardin, H. A. A.; Nash, J. J.; Wenthold, P. G. J. Am. Chem. Soc. 2002, 124, 12612–12618. (17) “A Reactivity Study on a 1,2,3,5-Tetradehydrobenzene: the 2,4,6-Tridehydropyridine Radical Cation”, Gallardo, V. A.; Jankiewicz, B. J.; Vinueza, N. R.; Nash, J. J.; Kenttämaa, H. I. J. Am. Chem. Soc. 2012, 134, 1926–1929. (18) “Reactivity Controlling Factors for an Aromatic Carbon-centered σ,σ,σ-Triradical: the 4,5,8Tridehydroisoquinolinium Cation”, Vinueza, N. R.; Jankiewicz, B. J.; Gallardo, V. A.; LaFavers, G. Z.; DeSutter, D.; Nash, J. J.; Kenttämaa, H. I. Chem. Eur. J. 2016, 22, 809–815. (19) “Reactivity of σ,σ,σ,σ,σ-Pentaradicals”, Max, J. P.; Ma, X.; Kotha, R. R.; Ding, D.; Milton, J.; Nash, J. J.; Kenttämaa, H. I. Int. J. Mass Spectrom. 2019, 435, 280–290. (20) “Effects of a Hydroxyl-Substituent on the Reactivity of the 2,4,6-Tridehydropyridinium Cation, an Aromatic σ,σ,σ-Triradical”, Jankiewicz, B. J.; Vinueza, N. R.; Reece, J. N.; Lee, Y. C.; Williams, P.; Nash, J. J.; Kenttӓmaa, H. I. Chem. Eur. J. 2012, 18, 969–974. (21) "Substituent Effects on the Reactivity of the 2,4,6-Tridehydropyridinium Cation, an Aromatic σ,σ,σ-Triradical”, Gao, J.; Jankiewicz, B. J.; Sheng, H.; Kirkpatrick, L.; Ma, X.; Nash, J. J.; Kenttämaa, H. I. Eur. J. Org. Chem. 2018, 6582–6589. (22) "Electronic Structure of the 1,3,5-Tridehydrobenzene Triradical in Its Ground and Excited States", Slipchenko, L. V.; Krylov, A. I. J. Chem. Phys. 2003, 118, 9614–9622. 23 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

(23) "Triradicals", Krylov, A. I. J. Phys. Chem. A 2005, 109, 10638–10645. (24) "Gaussian Basis Sets For Use in Correlated Molecular Calculations. I. The Atoms Boron Through Neon And Hydrogen", Dunning, T. H. J. Chem. Phys. 1989, 90, 1007–1023. (25) "Density-functional Exchange-energy Approximation With Correct Asymptotic Behavior", Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100. (26) "Development of the Colle-Salvetti Correlation-energy Formula Into a Functional of the Electron Density", Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B, 1988, 37, 785–789. (27) "Second-order Perturbation Theory With a CASSCF Reference Function", Andersson, K.; Malmqvist, P.-Å.; Roos, B. O.; Sadlej, A. J.; Wolinski, K. J. Phys. Chem. 1990, 94, 5483–5488. (28) "Different Forms of the Zeroth-order Hamiltonian in Second-order Perturbation Theory With a Complete Active Space Self-consistent Field Reference Function", Andersson, K. Theor. Chim. Acta 1995, 91, 31–46. (29) "A Complete Active Space SCF Method (CASSCF) Using a Density Matrix Formulated Super-CI Approach", Roos, B. O.; Taylor, P. R.; Siegbahn, P. E. M. Chem. Phys. 1980, 48, 157– 173. (30) "Multiconfigurational Second‐order Perturbation Theory: A Test of Geometries and Binding Energies", Andersson, K.; Roos, B. O. Int. J. Quantum Chem. 1993, 45, 591–607. (31) Note that, for these calculations, we are computing the electron affinity of the radical site not the electron affinity of the molecule. (32) Because the mono-, bi-, tri-, tetra- and pentaradicals studied here contain a formal positive charge on the nitrogen atom, the state that is produced when an electron is added to the nonbonding orbital of any one of these species is formally zwitterionic; that is, it contains localized positive () and negative () charges. (33) (a) MOLCAS 7.4: Aquilante, F.; De Vico, L.; Ferré, N.; Ghigo, G.; Malmqvist, P.-°A.; Neogrády; Pedersen, T. B.; Pitonak, M.; Reiher, M.; Roos, B. O.; Serrano-Andrés, L.; Urban, M.; Veryazov, V.; Lindh, R. J. Comp. Chem. 2010, 31, 224–247. (b) Code Development: Veryazov, V.; Widmark, P.-O.; Serrano- Andrés, L.; Lindh, R. Roos, B. O. Int. J. Quant. Chem. 2004, 100, 626–635. (c) MOLCAS 7: Karlstrőm, G.; Lindh, R.; Malmqvist, P.-°A.; Roos, B. O.; Ryde, U.; Veryazov, V.; Widmark, P.-O.; Cossi, M.; Schimmelpfennig, B.; Neogrády, P.; Seijo, L. Comp. Mat. Sci. 2003, 28, 222–239. (34) 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.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, 24 ACS Paragon Plus Environment

Page 24 of 26

Page 25 of 26 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

E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian, Inc., Wallingford CT, 2013. (35) "The Identification of Distonic Radical Cations on the Basis of a Reaction with Dimethyl Disulfide", Stirk, K. G.; Orlowski, J.; Leeck, D. T.; Kenttämaa, H. I. J. Am. Chem. Soc. 1992, 114, 8604–8606. (36) "The Heat of Formation of the Radical Cation of Dimethyl Disulfide", Leeck, D. T.; Kenttämaa, H. I. Org. Mass Spectrom. 1994, 29, 106–107. (37) "Evaluated Gas Phase Basicities and Proton Affinities of Molecules; Heats of Formation of Protonated Molecules", Lias, S. G.; Liebman, J. F.; Levin, R. D. J. Phys. Chem. Ref. Data 1984, 13, 695–808. (38) "Thermochemical Data on Gas‐phase Ion‐molecule Association and Clustering Reactions", Keesee, R. G.; Castleman, A. W. J. Phys. Chem. Ref. Data 1986, 15, 1011–1071. (39) “Quantum Chemical Characterization of the Structures, Thermochemical Properties, and Singlet-Triplet Splittings of Didehydroquinolinium and Didehydroisoquinolinium Ions”,Nash, J. J.; Kenttämaa, H. I.; Cramer, C. J. J. Phys Chem. A 2005, 109, 10348–10356. (40) “Benzynes, Dehydroconjugated Molecules, and the Interaction of Orbitals Separated by a Number of Intervening Sigma Bonds”, Hoffmann, R.; Imamura, A.; Hehre, W. J. J. Am. Chem. Soc. 1968, 90, 1499–1509. (41) “Spin-spin Coupling Between Two meta-Benzyne Moieties In a Quinolinium Tetraradical Cation Enhances Their Reactivities”, Kotha, R. R.; Yerabolu, R.; Ding, D.; Szalwinski, L.; Ma, X.; Wittrig, A.; Kong, J.; Nash, J. J.; Kenttämaa, H. I. Chem. Eur. J. published on line (DOI: 10.1002/chem. 201806096).

25 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

Content graphic The proposed mechanism for the production of dimethyl disulfide (DMDS) radical cation for selected charged triradicals upon reaction with DMDS. proton transfer

5

N+ H

controlled by PA of neutral radical and DMDS S S

CH3

N H + CH3 S S H 3C

H 3C controlled by EAv of neutral radical

hydrogen atom abstraction H

+

N

+ CH3 S S H 3C

DMDS radical cation (m/z 94)

26 ACS Paragon Plus Environment

Page 26 of 26