Isomer-specific vibronic structure of the 9-, 1-, and 2-anthracenyl radicals via slow photoelectron velocity-map imaging

Marissa L Weichman; Jessalyn A DeVine; Daniel S Levine; Jongjin B Kim; Daniel M Neumark

doi:10.1073/pnas.1520862113

. 2016 Jan 20;113(7):1698–1705. doi: 10.1073/pnas.1520862113

Isomer-specific vibronic structure of the 9-, 1-, and 2-anthracenyl radicals via slow photoelectron velocity-map imaging

Marissa L Weichman ^a, Jessalyn A DeVine ^a, Daniel S Levine ^a, Jongjin B Kim ^a,¹, Daniel M Neumark ^a,^b,²

PMCID: PMC4763764 PMID: 26792521

Significance

Polycyclic aromatic hydrocarbons (PAHs) are involved in soot nucleation following inefficient fuel combustion and are considered mutagens and environmental pollutants. They are also suspected to exist in the interstellar medium, although mechanisms for their formation in space are speculative. It is of great interest in these diverse fields to better characterize PAHs, including their dehydrogenated and charged derivatives, which are harder to isolate and probe. We use high-resolution anion photoelectron spectroscopy and quantum chemistry calculations to study the energetics, electronic states, and vibrational frequencies of the three dehydrogenated radical isomers of anthracene. These results provide signatures of these species for potential identification in space and illuminate subtle isomer-specific properties relevant to modeling their behavior in combustion and interstellar environments.

Keywords: polycyclic aromatic hydrocarbons, anion photoelectron spectroscopy, velocity-map imaging, vibronic structure

Abstract

Polycyclic aromatic hydrocarbons, in various charge and protonation states, are key compounds relevant to combustion chemistry and astrochemistry. Here, we probe the vibrational and electronic spectroscopy of gas-phase 9-, 1-, and 2-anthracenyl radicals (C₁₄H₉) by photodetachment of the corresponding cryogenically cooled anions via slow photoelectron velocity-map imaging (cryo-SEVI). The use of a newly designed velocity-map imaging lens in combination with ion cooling yields photoelectron spectra with <2 cm⁻¹ resolution. Isomer selection of the anions is achieved using gas-phase synthesis techniques, resulting in observation and interpretation of detailed vibronic structure of the ground and lowest excited states for the three anthracenyl radical isomers. The ground-state bands yield electron affinities and vibrational frequencies for several Franck–Condon active modes of the 9-, 1-, and 2-anthracenyl radicals; term energies of the first excited states of these species are also measured. Spectra are interpreted through comparison with ab initio quantum chemistry calculations, Franck–Condon simulations, and calculations of threshold photodetachment cross sections and anisotropies. Experimental measures of the subtle differences in energetics and relative stabilities of these radical isomers are of interest from the perspective of fundamental physical organic chemistry and aid in understanding their behavior and reactivity in interstellar and combustion environments. Additionally, spectroscopic characterization of these species in the laboratory is essential for their potential identification in astrochemical data.

Polycyclic aromatic hydrocarbons (PAHs) are an important class of species in many areas of chemistry. They are major components in coal (1) and in soot formed from combustion of organic matter (2, 3). PAHs are therefore common environmental pollutants and have well-documented mutagenic and carcinogenic biological activity (4, 5). PAHs are also believed to be abundant in the interstellar medium (6) and may be carriers of the anomalous IR emission bands (7–9). Recent molecular beam studies indicate that PAH growth can proceed through cold collisions of smaller hydrocarbons under interstellar conditions (10, 11). Individual PAH molecules can subsequently provide nucleation sites for amorphous graphitic grains (9). Interstellar PAHs and their clusters therefore bridge the gap between small carbonaceous molecules and larger particles, analogous to their role in soot condensation in combustion environments (12).

In space, PAH species are likely to exist as an equilibrium of neutral and ionic charge states, with varying degrees of hydrogenation and dehydrogenation (13–15). Models of dense interstellar clouds find that anionic PAHs are the major carriers of negative charge, rather than free electrons (16). Closed-shell, singly deprotonated PAH carbanions have large electron affinities compared with radical anionic parent species and may therefore be reasonably robust in the interstellar medium (6, 13). In this article, we investigate the three dehydrogenated isomers of anthracene, the 9-, 1-, and 2-anthracenyl radicals, via slow photoelectron velocity-map imaging (cryo-SEVI) of the corresponding cryogenically cooled, deprotonated anions. Structures of the $C_{14} H_{9}^{-}$ isomers are shown in Fig. 1. This cryo-SEVI technique previously yielded highly vibrationally resolved spectra of α- and β-naphthyl (17). We demonstrate here that even larger PAH anions are accessible to detailed characterization with cryo-SEVI and that the three anthracenyl isomers demonstrate strikingly distinct energetics and spectroscopic signatures.

Fig. 1. — (A–C) SEVI spectra of photodetachment to (A) the ${\tilde{X}}^{2} A_{1}$ ground state of 9-anthracenyl and the ${\tilde{X}}^{2} A'$ ground states of (B) 1-anthracenyl and (C) 2-anthracenyl, with low-resolution overview scans in blue and high-resolution traces in black. Features in the spectrum of 2-anthracenyl appearing due to 1-anthracenyl contamination are plotted in gray. FC simulations are shown in red.

The anthracene molecule has been well characterized experimentally. Its infrared spectrum has been measured in a rare gas matrix (18, 19) and in the gas phase (20), and its vibrational structure has been largely assigned (21). The $S_{1} \leftarrow S_{0}$ electronic transition in anthracene has been investigated with cavity ring-down spectroscopy (22). The high-temperature oxidation reactions of anthracene with O₂ and OH have been studied experimentally (23, 24). In both cases, H atom abstraction to form the anthracenyl radical intermediate competes with direct oxidation of anthracene. These reaction pathways govern the balance between efficient combustion and soot formation; detailed understanding of the intermediates involved is essential for accurately modeling combustion.

The anthracenyl radical and anion isomers are not nearly as well characterized as the anthracene parent. The 9- and 1-anthracenyl radicals were examined with ESR spectroscopy (25), which suggested that the unpaired electron resides in a σ orbital localized at the site of dehydrogenation. The reactivity of the 9-anthracenyl radical with naphthalene and toluene has also been measured experimentally (26). A recent multiple-photon electron detachment study yielded the infrared action spectrum of the 9-anthracenyl anion (6). Anionic anthracene derivatives in various protonation states are also used in electron transfer dissociation mass spectrometry as an electron source to induce fragmentation of peptide backbones (27). A fair amount of additional theoretical work has been reported detailing the energetics, electronic structure, geometries, vibrational frequencies, and reactivities of the anthracenyl radicals and anions (1, 6, 28–32). The three C−H bond dissociation energies of anthracene are predicted to be very similar, leading to three anthracenyl radical isomers nearly degenerate in energy (1, 28, 29, 31). The anionic isomers are calculated to be more separated in energy, with the 9-anthracenyl anion being the most stable and the 1- and 2-anthracenyl anions lying 0.14 eV and 0.18–0.21 eV higher in energy, respectively (6, 29).

Anion photoelectron spectroscopy (PES) is a technique well suited for probing the vibronic structure of neutral radicals through photodetachment of a closed-shell anion (33, 34). The cryo-SEVI technique used here is a high-resolution variant of PES and yields spectra of complex anions with sub-meV resolution (17, 35, 36). Specific anthracenyl anion isomers are prepared using trimethylsilyl-anthracene precursors and cooled to cryogenic temperatures (∼10 K) before photodetachment, eliminating hot spectral features and narrowing the rotational profiles of the observed peaks. Cooling is essential for obtaining interpretable spectra of large molecular species with many low-frequency vibrational modes. We also report the use of a newly designed velocity-map imaging (VMI) electrostatic lens with improved energy resolution. This development allows us to observe peaks with 2–3 cm⁻¹ FWHM and resolve splittings as small as 3 cm⁻¹ for the anthracenyl system. Focusing is also improved at higher electron kinetic energy (eKE), yielding narrower features farther from threshold and thus facilitating studies of anions with poor threshold photodetachment cross sections.

We present highly vibrationally resolved, isomer-specific spectra of transitions to the ground and first excited states of the 9-, 1-, and 2-anthracenyl radicals, providing much spectroscopic information and demonstrating the improved capabilities of the cryo-SEVI instrument. We obtain precise electron affinities for three anthracenyl radical isomers and term energies for their excited states, and measure Franck–Condon active fundamental vibrational frequencies. Our results illuminate the distinct spectroscopy, energetics, and potential reactivity of these isomers, with far-reaching applications in interpretation of astrochemical data and modeling of combustion chemistry.

Results and Discussion

Experimental Photodetachment Spectra.

The cryo-SEVI spectra of 9-, 1-, and 2-anthracenyl presented in Figs. 1 and 2 comprise two well-separated electronic bands, which for each isomer are assigned as transitions from the anion ground state to the ground electronic state ( $\tilde{X}$ ) and the first excited state ( $\tilde{A}$ ) of the neutral radical. Franck–Condon (FC) simulations were carried out for all states and are plotted as red stick spectra. Experimental details are described in Materials and Methods, Experimental, and ab initio calculations are described in Materials and Methods, Electronic Structure Calculations. Experimental energetics and vibrational frequencies for all anthracenyl isomers are summarized in Table 1 and compared with calculated values. Positions and assignments for the peaks labeled in Figs. 1 and 2 are summarized in Tables S1–S3.

Fig. 2. — (A and B) SEVI spectra of photodetachment to (A) the ${\tilde{A}}^{2} B_{1}$ excited state of 9-anthracenyl and (B) the ${\tilde{A}}^{2} A^{″}$ state of 1-anthracenyl, with colored traces taken at progressively lower photon energies and FC simulations in red.

Table 1.

Experimental and calculated electron affinities, excited-state term energies, and vibrational frequencies for the 9-, 1-, and 2-anthracenyl radicals

State	Parameter	Experimental	Calculated
$9, \tilde{X}$	EA,^* eV	1.7155 (2)^†	1.6749
	ν₁₅, cm⁻¹	1,147 (1)	1,168
	ν₁₇, cm⁻¹	896 (2)	907
	ν₁₈, cm⁻¹	752 (4)	759
	ν₁₉, cm⁻¹	648 (2)	658
	ν₂₀, cm⁻¹	621 (2)	638
	ν₂₁, cm⁻¹	390 (2)	398
	ν₂₂, cm⁻¹	228 (2)	233
$9, \tilde{A}$	T₀,^‡ eV	1.205 (6)	1.0382
$1, \tilde{X}$	EA, eV	1.5436 (2)	1.5287
	ν_30, cm⁻¹	1,091 (4)	1,113
	ν_34, cm⁻¹	891 (2)	902
	ν_36, cm⁻¹	753 (2)	762
	ν_38, cm⁻¹	621 (2)	639
	ν_39, cm⁻¹	599 (2)	618
	ν_40, cm⁻¹	514 (2)	526
	ν_42, cm⁻¹	388 (2)	395
	ν_43, cm⁻¹	232 (2)	236
$1, \tilde{A}$	T₀, eV	1.515 (4)	1.3277
	ν_56, cm⁻¹	496 (36)	504
	ν_60, cm⁻¹	255 (33)	256
$2, \tilde{X}$	EA, eV	1.4671 (2)	1.4734
	ν₃₁, cm⁻¹	1,013 (2)	1,039
	ν₃₄, cm⁻¹	882 (2)	894
	ν₃₇, cm⁻¹	648 (1)	658
	ν₃₉, cm⁻¹	591 (1)	609
	ν₄₀, cm⁻¹	522 (2)	533
	ν₄₁, cm⁻¹	389 (1)	398
	ν₄₂, cm⁻¹	392 (1)	396
	ν₄₃, cm⁻¹	234 (1)	238
$2, \tilde{A}$	T₀, eV	1.755 (8)	1.5580

Open in a new tab

Electron affinity.

^{^†}

Uncertainties represent 1 SD of a Gaussian fit to the experimentally observed peak.

^{^‡}

Term energy.

Table S1.

Peak positions (cm⁻¹), offsets from the origin (cm⁻¹), and assignments for the SEVI spectra of 9-anthracenyl given in Figs. 1A and 2A

Peak	Position	Offset	Assignment	Band
A	13,837	0	$0_{0}^{0}$	${\tilde{X}}^{2} A_{1} \leftarrow {\tilde{X}}^{1} A_{1}$
B	14,064	228	$22_{0}^{1}$
C	14,227	390	$21_{0}^{1}$
D	14,457	621	$20_{0}^{1}$
E	14,485	648	$19_{0}^{1}$
F	14,588	752	$18_{0}^{1}$
G	14,687	850	$20_{0}^{1} 22_{0}^{1}$
H	14,732	896	$17_{0}^{1}$
I	14,847	1,011	$20_{0}^{1} 21_{0}^{1}$
J	14,875	1,039	$19_{0}^{1} 21_{0}^{1}$
K	14,978	1,142	$18_{0}^{1} 21_{0}^{1}$
L	14,984	1,147	$15_{0}^{1}$
M	15,077	1,241	$20_{0}^{2}$
N	15,104	1,268	$19_{0}^{1} 20_{0}^{1}$
O	15,121	1,284	$17_{0}^{1} 21_{0}^{1}$
P	15,207	1,370	$18_{0}^{1} 20_{0}^{1}$
Q	15,237	1,400	$18_{0}^{1} 19_{0}^{1}$
R	15,352	1,515	$17_{0}^{1} 20_{0}^{1}$
a	23,555	0	$0_{0}^{0}$	${\tilde{A}}^{2} B_{1} \leftarrow {\tilde{X}}^{1} A_{1}$

Open in a new tab

Table S3.

Peak positions (cm⁻¹), offsets from the origin (cm⁻¹), and assignments for the SEVI spectrum of 2-anthracenyl given in Fig. 1C

Peak	Position	Offset	Assignment	Band
A	11,833	0	$0_{0}^{0}$	${\tilde{X}}^{2} A' \leftarrow {\tilde{X}}^{1} A'$
B	12,067	234	$43_{0}^{1}$
C	12,222	389	$42_{0}^{1}$
D	12,224	392	$41_{0}^{1}$
E	12,354	522	$40_{0}^{1}$
F	12,424	591	$39_{0}^{1}$
G	12,432	599	$59_{0}^{1} 61_{0}^{1}$
H	12,481	648	$37_{0}^{1}$
I	12,588	756	$40_{0}^{1} 43_{0}^{1}$
J	12,614	781	$41_{0}^{2}, 42_{0}^{2}$
K	12,715	882	$34_{0}^{1}, 37_{0}^{1} 43_{0}^{1}$
L	12,746	913	$40_{0}^{1} 41_{0}^{1}, 40_{0}^{1} 42_{0}^{1}$
M	12,845	1,013	$31_{0}^{1}$
N	12,871	1,038	$37_{0}^{1} 41_{0}^{1}, 37_{0}^{1} 42_{0}^{1}$
O	12,876	1,044	$40_{0}^{2}$
P	12,949	1,116	$34_{0}^{1} 43_{0}^{1}, 39_{0}^{1} 40_{0}^{1}$
Q	12,977	1,145	$40_{0}^{1} 41_{0}^{1} 43_{0}^{1}, 40_{0}^{1} 42_{0}^{1} 43_{0}^{1}$
R	13,003	1,170	$37_{0}^{1} 40_{0}^{1}$
S	13,106	1,273	$34_{0}^{1} 41_{0}^{1}, 34_{0}^{1} 42_{0}^{1}$
T	13,129	1,296	$37_{0}^{2}$
U	13,237	1,404	$34_{0}^{1} 40_{0}^{1}, 37_{0}^{1} 40_{0}^{1} 43_{0}^{1}$

Open in a new tab

Cryo-SEVI spectra of the 9-, 1-, and 2-anthracenyl $\tilde{X}$ bands are shown in Fig. 1 as a function of electron binding energy (eBE). With SEVI, we first measure a low-resolution overview spectrum at a photon energy well above the eBE of the band of interest, plotted in Fig. 1 in blue. We then obtain high-resolution spectra at discrete photon energies tuned above the spectral features of interest and splice them together to create a composite high-resolution spectrum of the full region, plotted in black in Fig. 1. The high-resolution trace for a given feature is then scaled to the intensity of that feature in the overview spectrum to minimize any threshold effects that distort relative peak intensities.

The ground-state spectra of all isomers show congested but well-resolved structure that is accurately predicted by FC simulations. High-resolution peak widths are typically 4–5 cm⁻¹ FWHM for the ground states of all three anthracenyl radicals and as narrow as 2–3 cm⁻¹ FWHM. This constitutes a notable improvement in instrumental resolution over previous cryo-SEVI studies of aromatic molecules, where the narrowest features were more often 7–8 cm⁻¹ FWHM (17, 36). Peaks indicating some contamination of the 1-anthracenyl isomer in the 2-anthracenyl spectrum are plotted in gray in Fig. 1C. This contamination is likely due to decomposition of either the 2-trimethylsilyl-anthracene precursor or the 2-anthracenyl anions after formation in the ion source. The contamination worsens with increasing temperature of the ion source and over time with use of the same precursor sample.

Spectra of the $\tilde{A}$ bands of 9- and 1-anthracenyl are shown in Fig. 2 A and B. Due to poor threshold cross sections for these bands, we do not plot composite high-resolution spectra, but rather the full traces at progressively lower photon energies. Although we report its term energy in Table 1, the 2-anthracenyl $\tilde{A}$ -state spectrum is not presented due to contamination from 1-anthracenyl structure made worse by relative cross-section effects. The $\tilde{A}$ bands of all isomers have strong vibrational origins and weak vibrationally excited structure. The profiles of these bands are in qualitative agreement with FC simulations, although more vibrational activity is seen in the spectra than in the simulations.

Compared with the $\tilde{X}$ bands, the $\tilde{A}$ -state vibrational origins are very weak at low eKE, preventing measurements close to threshold where SEVI resolution is best. Hence, peak widths for the $\tilde{A}$ -band origins are limited to ∼100 cm⁻¹ FWHM, even with the improved resolution afforded by the redesigned VMI lens. The remaining features in the $\tilde{A}$ bands largely follow this trend of vanishing intensity close to threshold. However, the intensities of some peaks (most notably b and c) in the 1-anthracenyl spectrum increase dramatically at certain photon energies, evidenced by the purple trace plotted in Fig. 2B.

SEVI also provides information about the anisotropy of photodetachment transitions. For a photodetachment process with one photon of linearly polarized light, the photoelectron angular distribution (PAD) is given by (37)

I (θ) = \frac{σ_{tot}}{4 π} [1 + β P_{2} (\cos θ)],

[1]

where θ is the angle of electron signal relative to the polarization axis of the laser, P₂ is the second-order Legendre polynomial, and β is the anisotropy parameter, which varies between −1 for a PAD aligned perpendicular to the laser polarization and +2 for a PAD parallel to the laser polarization.

Experimental values of β for the vibrational origins of the $\tilde{X}$ bands of all isomers are plotted in Fig. 3 as a function of eKE, along with PAD simulations for the $\tilde{X}$ and $\tilde{A}$ bands. We do not report quantitative anisotropy parameters for the excited states, because high-eKE photoelectrons from detachment to the ground states create a background that distorts the excited-state anisotropies during image reconstruction. However, it is qualitatively clear that all radical isomers have $\tilde{A}$ -state vibrational origins with slightly perpendicularly polarized PADs ( $β < 0$ ). Intriguingly, in the 1-anthracenyl $\tilde{A}$ -state spectrum plotted in purple in Fig. 2B, peaks b and c demonstrate positive β values, in contrast to the vibrational origin a.

Assignment of Electronic Structure.

Cryogenic cooling ensures, in principle, that we photodetach from the ground vibrational and electronic state of each anthracenyl anion. Our electronic structure calculations indicate that the ground state of each neutral radical is accessed by removing an electron from an in-plane σ molecular orbital (MO) with s-p character localized on the deprotonated site of the closed-shell anion, whereas the first excited state of each radical is accessed by removal of an electron from a delocalized π MO of the anion. Calculated Dyson orbitals (38) for the photodetachment transitions for each isomer are shown in Fig. 4.

Fig. 4. — (*A–C*) Calculated cross sections and visualized Dyson orbitals for photodetachment to the $\tilde{X}$ and $\tilde{A}$ states of (A) 9-anthracenyl, (B) 1-anthracenyl, and (C) 2-anthracenyl as functions of eKE.

We assign the low binding energy bands in Fig. 1 to the ${\tilde{X}}^{2} A_{1} \leftarrow {\tilde{X}}^{1} A_{1}$ detachment transition of 9-anthracenyl and the ${\tilde{X}}^{2} A' \leftarrow {\tilde{X}}^{1} A'$ transitions of 1- and 2-anthracenyl. The peaks labeled A in Fig. 1 represent the vibrational origin of each $\tilde{X} \leftarrow \tilde{X}$ band, yielding precise experimental electron affinities (EAs) for the 9-, 1-, and 2-anthracenyl radicals that are in good agreement with the calculations carried out in the present work (Table 1) as well as in the literature (29). The decrease in EA as the site of deprotonation is moved from C₉ to C₁ to C₂ reflects the energetic ordering of the anions. The isomer energetics calculated in the present work (Table 2) are in good agreement with the literature (1, 6, 28, 29, 31); we find that the 9-anthracenyl anion is lowest in energy with the 1-anthracenyl and 2-anthracenyl anions lying 0.13 eV and 0.18 eV higher in energy, respectively. By contrast, the 9-, 1-, and 2-anthracenyl radicals fall within 0.02 eV of one another.

Table 2.

Calculated energies (eV) for all species at the B3LYP/6-311+G* level of theory, given relative to the 9-anthracenyl anion ground state and corrected for vibrational zero-point energy

State	9-Anthracenyl	1-Anthracenyl	2-Anthracenyl
Anion	0.0000	0.1309	0.1842
Ground	1.6749	1.6596	1.6576
Excited	2.7130	2.9873	3.2156

Open in a new tab

The higher eBE bands shown in Fig. 2 are assigned to the ${\tilde{A}}^{2} B_{1} \leftarrow {\tilde{X}}^{1} A_{1}$ band of 9-anthracenyl and the ${\tilde{A}}^{2} A^{″} \leftarrow {\tilde{X}}^{1} A'$ transition of 1-anthracenyl; the peaks labeled a are the vibrational origins of these two bands. Experimental and calculated energetics compare favorably (Table 1) and reproduce the trend of increasing $\tilde{A}$ -state term energy from 9- to 1- to 2-anthracenyl. Because the radical ground states are predicted to be nearly degenerate, the differences in T₀ are due to destabilization of the 1- and 2-anthracenyl $\tilde{A}$ states relative to that of 9-anthracenyl. The calculated results in Table 2 place the 1-anthracenyl and 2-anthracenyl $\tilde{A}$ states respectively 0.27 eV and 0.50 eV above the 9-anthracenyl $\tilde{A}$ state. Rationalizations for the energetic ordering of the anion and neutral isomers are discussed in further detail below.

Radical Ground States.

The $\tilde{X}$ bands of the anthracenyl radicals show extensive FC activity, indicative of a large change in geometry upon photodetachment. This FC structure is consistent with detachment of an electron from a highly localized MO (Fig. 4) and with geometry optimization calculations (Tables S4–S6), which indicate that the C−C−C interior bond angle at the deprotonated site widens by ∼14° in all three systems upon detachment to the radical ground state.

Table S4.

Optimized Cartesian coordinates (Å) of the anion and neutral 9-anthracenyl states calculated with B3LYP/6-311+G*

State	Atom	x	y	z
Anion $\tilde{X}$	C₁	0.0000000000	−2.4620600000	−1.4138759140
	C₂	0.0000000000	−3.6566900000	−0.7343559140
	C₃	0.0000000000	−3.6579300000	0.6888240860
	C₄	0.0000000000	−2.4744600000	1.3814940860
	C_4a	0.0000000000	−1.2150500000	0.7017240860
	C₅	0.0000000000	2.4744600000	1.3814940860
	C₆	0.0000000000	3.6579300000	0.6888240860
	C₇	0.0000000000	3.6566900000	−0.7343559140
	C₈	0.0000000000	2.4620600000	−1.4138759140
	C_8a	0.0000000000	1.1935700000	−0.7524959140
	C₉	0.0000000000	0.0000000000	−1.5284859140
	C_9a	0.0000000000	−1.1935700000	−0.7524959140
	C₁₀	0.0000000000	0.0000000000	1.3964840860
	C_10a	0.0000000000	1.2150500000	0.7017240860
	H₁	0.0000000000	−2.4333600000	−2.5001359140
	H₂	0.0000000000	−4.6010500000	−1.2740559140
	H₃	0.0000000000	−4.6035600000	1.2276340860
	H₄	0.0000000000	−2.4796100000	2.4714040860
	H₅	0.0000000000	2.4796100000	2.4714040860
	H₆	0.0000000000	4.6035600000	1.2276340860
	H₇	0.0000000000	4.6010500000	−1.2740559140
	H₈	0.0000000000	2.4333600000	−2.5001359140
	H₁₀	0.0000000000	0.0000000000	2.4865440860
Neutral $\tilde{X}$	C₁	0.0000000000	−2.4903245048	−1.4431914527
	C₂	0.0000000000	−3.6653826971	−0.7440876676
	C₃	0.0000000000	−3.6617710720	0.6808843118
	C₄	0.0000000000	−2.4837968675	1.3760293326
	C_4a	0.0000000000	−1.2264868379	0.6974902475
	C₅	0.0000000000	2.4837968675	1.3760293326
	C₆	0.0000000000	3.6617710720	0.6808843118
	C₇	0.0000000000	3.6653826971	−0.7440876676
	C₈	0.0000000000	2.4903245048	−1.4431914527
	C_8a	0.0000000000	1.2375428734	−0.7555250468
	C₉	0.0000000000	0.0000000000	−1.3641524660
	C_9a	0.0000000000	−1.2375428734	−0.7555250468
	C₁₀	0.0000000000	0.0000000000	1.3726166406
	C_10a	0.0000000000	1.2264868379	0.6974902475
	H₁	0.0000000000	−2.4863326037	−2.5278436424
	H₂	0.0000000000	−4.6124758595	−1.2738202670
	H₃	0.0000000000	−4.6066754977	1.2147008623
	H₄	0.0000000000	−2.4869416435	2.4623671874
	H₅	0.0000000000	2.4869416435	2.4623671874
	H₆	0.0000000000	4.6066754977	1.2147008623
	H₇	0.0000000000	4.6124758595	−1.2738202670
	H₈	0.0000000000	2.4863326037	−2.5278436424
	H₁₀	0.0000000000	0.0000000000	2.4592099737
Neutral $\tilde{A}$	C₁	0.0000000000	−2.4487622124	−1.4143301322
	C₂	0.0000000000	−3.6553967787	−0.7216686326
	C₃	0.0000000000	−3.6602471316	0.6817314098
	C₄	0.0000000000	−2.4656450631	1.3832762666
	C_4a	0.0000000000	−1.2294882229	0.6953851477
	C₅	0.0000000000	2.4656450631	1.3832762666
	C₆	0.0000000000	3.6602471316	0.6817314098
	C₇	0.0000000000	3.6553967787	−0.7216686326
	C₈	0.0000000000	2.4487622124	−1.4143301322
	C_8a	0.0000000000	1.2089808498	−0.7451534252
	C₉	0.0000000000	0.0000000000	−1.5357106792
	C_9a	0.0000000000	−1.2089808498	−0.7451534252
	C₁₀	0.0000000000	0.0000000000	1.3888802971
	C_10a	0.0000000000	1.2294882229	0.6953851477
	H₁	0.0000000000	−2.4251695029	−2.4986786282
	H₂	0.0000000000	−4.5953886979	−1.2639732170
	H₃	0.0000000000	−4.6031186222	1.2194076697
	H₄	0.0000000000	−2.4728777445	2.4701378826
	H₅	0.0000000000	2.4728777445	2.4701378826
	H₆	0.0000000000	4.6031186222	1.2194076697
	H₇	0.0000000000	4.5953886979	−1.2639732170
	H₈	0.0000000000	2.4251695029	−2.4986786282
	H₁₀	0.0000000000	0.0000000000	2.4763072681

Open in a new tab

Table S6.

Optimized Cartesian coordinates (Å) of the anion and neutral 2-anthracenyl states calculated with B3LYP/6-311+G*

State	Atom	x	y	z
Anion $\tilde{X}$	C₁	0.0000000000	−2.5390426031	−1.4146391679
	C₂	0.0000000000	−3.7894927876	−0.8089171589
	C₃	0.0000000000	−3.7009158762	0.6380151805
	C₄	0.0000000000	−2.5364171089	1.3705494893
	C_4a	0.0000000000	−1.2699623073	0.7147051971
	C₅	0.0000000000	2.4352519339	1.4071578975
	C₆	0.0000000000	3.6239600080	0.7231272470
	C₇	0.0000000000	3.6266995741	−0.7021615720
	C₈	0.0000000000	2.4433233896	−1.3940772055
	C_8a	0.0000000000	1.1820315649	−0.7212925638
	C₉	0.0000000000	−0.0416025576	−1.4000290948
	C_9a	0.0000000000	−1.2770859969	−0.7347903407
	C₁₀	0.0000000000	−0.0489194016	1.3965074735
	C_10a	0.0000000000	1.1810241040	0.7259112829
	H₁	0.0000000000	−2.4678815105	−2.5093837544
	H₃	0.0000000000	−4.6346919462	1.2101730402
	H₄	0.0000000000	−2.5546055100	2.4628442961
	H₅	0.0000000000	2.4314121832	2.4952976751
	H₆	0.0000000000	4.5670515688	1.2633021126
	H₇	0.0000000000	4.5728461730	−1.2372786910
	H₈	0.0000000000	2.4466826109	−2.4819545493
	H₉	0.0000000000	−0.0394575085	−2.4894290231
	H₁₀	0.0000000000	−0.0544676730	2.4860289086
Neutral $\tilde{X}$	C₁	0.0000000000	−2.5114993943	−1.4754104580
	C₂	0.0000000000	−3.6428478473	−0.7432778976
	C₃	0.0000000000	−3.7314770110	0.6578790150
	C₄	0.0000000000	−2.5485753590	1.3563569489
	C_4a	0.0000000000	−1.2875621493	0.6800978332
	C₅	0.0000000000	2.4004296428	1.4282576018
	C₆	0.0000000000	3.5919144931	0.7570201941
	C₇	0.0000000000	3.6167987285	−0.6674121456
	C₈	0.0000000000	2.4503861554	−1.3816230510
	C_8a	0.0000000000	1.1835384403	−0.7206757801
	C₉	0.0000000000	−0.0265386973	−1.4236809512
	C_9a	0.0000000000	−1.2588433658	−0.7649877359
	C₁₀	0.0000000000	−0.0759347866	1.3796309054
	C_10a	0.0000000000	1.1583493081	0.7216210243
	H₁	0.0000000000	−2.5121306369	−2.5617833902
	H₃	0.0000000000	−4.6879259645	1.1698856044
	H₄	0.0000000000	−2.5580205360	2.4430911174
	H₅	0.0000000000	2.3809836289	2.5143159621
	H₆	0.0000000000	4.5274684804	1.3070158211
	H₇	0.0000000000	4.5713755483	−1.1838081325
	H₈	0.0000000000	2.4708363176	−2.4675974322
	H₉	0.0000000000	−0.0070752651	−2.5103461993
	H₁₀	0.0000000000	−0.0943405185	2.4664536287
Neutral $\tilde{A}$	C₁	0.0000000000	−2.5095689598	−1.4174032825
	C₂	0.0000000000	−3.8000183519	−0.7990375270
	C₃	0.0000000000	−3.7160067887	0.6337238796
	C₄	0.0000000000	−2.5415951047	1.3591702723
	C_4a	0.0000000000	−1.2659293461	0.7051654263
	C₅	0.0000000000	2.4144903506	1.4128848006
	C₆	0.0000000000	3.6140216751	0.7271641391
	C₇	0.0000000000	3.6296198742	−0.6822813899
	C₈	0.0000000000	2.4451570997	−1.3898609328
	C_8a	0.0000000000	1.1996215176	−0.7161784601
	C₉	0.0000000000	−0.0343353224	−1.4087123384
	C_9a	0.0000000000	−1.2590341299	−0.7372913468
	C₁₀	0.0000000000	−0.0663905049	1.3926320603
	C_10a	0.0000000000	1.1852072249	0.7176179614
	H₁	0.0000000000	−2.4569659549	−2.5082578358
	H₃	0.0000000000	−4.6462809361	1.2020196645
	H₄	0.0000000000	−2.5574226270	2.4484436722
	H₅	0.0000000000	2.4069286016	2.4986228030
	H₆	0.0000000000	4.5507378963	1.2744362714
	H₇	0.0000000000	4.5777957212	−1.2092426425
	H₈	0.0000000000	2.4559588721	−2.4755033776
	H₉	0.0000000000	−0.0302145058	−2.4957404249
	H₁₀	0.0000000000	−0.0719724693	2.4796622968

Open in a new tab

Table S5.

Optimized Cartesian coordinates (Å) of the anion and neutral 1-anthracenyl states calculated with B3LYP/6-311+G*

State	Atom	x	y	z
Anion $\tilde{X}$	C₁	0.0000000000	−2.4943770380	−1.5558017103
	C₂	0.0000000000	−3.6490852413	−0.7800318218
	C₃	0.0000000000	−3.6717525894	0.6463767173
	C₄	0.0000000000	−2.5100022904	1.3802319334
	C_4a	0.0000000000	−1.2585184396	0.6995525986
	C₅	0.0000000000	2.4499581246	1.4009853529
	C₆	0.0000000000	3.6354050113	0.7125020607
	C₇	0.0000000000	3.6385144267	−0.7130688619
	C₈	0.0000000000	2.4521233203	−1.4013998958
	C_8a	0.0000000000	1.1945359591	−0.7253430613
	C₉	0.0000000000	−0.0311998399	−1.4056707726
	C_9a	0.0000000000	−1.2767856386	−0.7598773381
	C₁₀	0.0000000000	−0.0360322969	1.3882773251
	C_10a	0.0000000000	1.1924905998	0.7207667181
	H₂	0.0000000000	−4.6274321081	−1.2726378490
	H₃	0.0000000000	−4.6310147345	1.1698771345
	H₄	0.0000000000	−2.5270849287	2.4689885717
	H₅	0.0000000000	2.4502265439	2.4891581871
	H₆	0.0000000000	4.5794372432	1.2517814573
	H₇	0.0000000000	4.5839205009	−1.2492740271
	H₈	0.0000000000	2.4501790759	−2.4890275581
	H₉	0.0000000000	−0.0478040871	−2.4929299123
	H₁₀	0.0000000000	−0.0420719110	2.4790685300
Neutral $\tilde{X}$	C₁	0.0000000000	−2.4856990474	−1.4068400777
	C₂	0.0000000000	−3.6918630013	−0.8046329559
	C₃	0.0000000000	−3.7022732806	0.6312731707
	C₄	0.0000000000	−2.5295052520	1.3352887214
	C_4a	0.0000000000	−1.2592106553	0.6765946911
	C₅	0.0000000000	2.4298332963	1.4118262902
	C₆	0.0000000000	3.6205248108	0.7391505939
	C₇	0.0000000000	3.6451357645	−0.6852698904
	C₈	0.0000000000	2.4780823738	−1.3979720508
	C_8a	0.0000000000	1.2116799880	−0.7357574379
	C₉	0.0000000000	0.0033544121	−1.4402905048
	C_9a	0.0000000000	−1.2263560011	−0.7751692729
	C₁₀	0.0000000000	−0.0455470245	1.3705046400
	C_10a	0.0000000000	1.1865888059	0.7071916419
	H₂	0.0000000000	−4.6245104726	−1.3600280410
	H₃	0.0000000000	−4.6554109662	1.1513880625
	H₄	0.0000000000	−2.5477190154	2.4209926838
	H₅	0.0000000000	2.4120956322	2.4978443975
	H₆	0.0000000000	4.5564103967	1.2886653350
	H₇	0.0000000000	4.5993715096	−1.2021848008
	H₈	0.0000000000	2.4968021445	−2.4838854669
	H₉	0.0000000000	0.0146046518	−2.5257188760
	H₁₀	0.0000000000	−0.0601150162	2.4575413530
Neutral $\tilde{A}$	C₁	0.0000000000	−2.4673217352	−1.5595502229
	C₂	0.0000000000	−3.6591841922	−0.7622291087
	C₃	0.0000000000	−3.6870046125	0.6371366983
	C₄	0.0000000000	−2.5026285133	1.3693410966
	C_4a	0.0000000000	−1.2522948619	0.6954115189
	C₅	0.0000000000	2.4369962807	1.4031965331
	C₆	0.0000000000	3.6296635726	0.7091843285
	C₇	0.0000000000	3.6349778117	−0.7011176195
	C₈	0.0000000000	2.4446372189	−1.4006834368
	C_8a	0.0000000000	1.2055708784	−0.7197826387
	C₉	0.0000000000	−0.0347208380	−1.4095870125
	C_9a	0.0000000000	−1.2641227547	−0.7562614441
	C₁₀	0.0000000000	−0.0436378592	1.3899007519
	C_10a	0.0000000000	1.2004013594	0.7139272229
	H₂	0.0000000000	−4.6202629425	−1.2758419489
	H₃	0.0000000000	−4.6370039095	1.1671767112
	H₄	0.0000000000	−2.5246368702	2.4555817145
	H₅	0.0000000000	2.4369011141	2.4890739406
	H₆	0.0000000000	4.5706034973	1.2494746332
	H₇	0.0000000000	4.5792772513	−1.2351459936
	H₈	0.0000000000	2.4468838067	−2.4862486176
	H₉	0.0000000000	−0.0521726426	−2.4952359560
	H₁₀	0.0000000000	−0.0475798342	2.4778455153

Open in a new tab

Several fundamental frequencies of each radical ground state are observed (Table 1). The FC active vibrational modes should be totally symmetric and are therefore attributed to a₁ modes for the C_2v-symmetric 9-anthracenyl radical and aʹ modes for the C_s-symmetric 1- and 2-anthracenyl radicals. Schematics for normal mode displacements of all FC active vibrations are given in Fig. S1. The most highly FC active modes for each radical involve significant distortion at the deprotonated site, where there is the largest change in geometry upon photodetachment.

Fig. S1. — (*A–C*) Franck–Condon active vibrational modes for the ground-state photoelectron spectra of (A) 9-anthracenyl, (B) 1-anthracenyl, and (C) 2-anthracenyl.

For the 9-anthracenyl radical ground state (Fig. 1A), all spectral features can be assigned to fundamentals, progressions, and combination bands of the vibrational frequencies listed in Table 1. Full peak assignments are detailed in Table S1. The FC stick spectra are in very good agreement with the experimentally observed features. Only peak B does not appear in the simulation, but it can be assigned unambiguously to ν₂₂ based on comparison with the ab initio calculated frequencies of FC allowed a₁ modes.

The experimental 1-anthracenyl radical $\tilde{X}$ band also largely agrees with the FC simulation (Fig. 1B). Peak B again does not appear in the simulation, but it can be assigned to the ν₄₃ fundamental as the only FC allowed aʹ mode with a sufficiently low calculated frequency. Additionally, we observe a strong feature (peak F) not predicted by the FC simulation, appearing only 8 cm⁻¹ above ν₃₉ (peak E). Elsewhere in the spectrum, similar doublets (e.g., peaks N/O and V/W) appear where the FC simulation predicts only one peak involving excitation of the ν₃₉ mode. Nearly all other peaks in the spectrum can be assigned to progressions and combination bands of the observed fundamentals and peak F (Table S2).

Table S2.

Peak positions (cm⁻¹), offsets from the origin (cm⁻¹), and assignments for the SEVI spectra of 1-anthracenyl given in Figs. 1B and 2B

Peak	Position	Offset	Assignment	Band
A	12,450	0	$0_{0}^{0}$	${\tilde{X}}^{2} A' \leftarrow {\tilde{X}}^{1} A'$
B	12,682	232	$43_{0}^{1}$
C	12,838	388	$42_{0}^{1}$
D	12,964	514	$40_{0}^{1}$
E	13,049	599	$39_{0}^{1}$
F	13,057	607	$59_{0}^{1} 61_{0}^{1}$
G	13,071	621	$38_{0}^{1}$
H	13,203	753	$36_{0}^{1}$
I	13,210	760	?
J	13,278	828	$39_{0}^{1} 43_{0}^{1}$
K	13,286	836	$43_{0}^{1} 59_{0}^{1} 61_{0}^{1}$
L	13,303	853	$38_{0}^{1} 43_{0}^{1}$
M	13,341	891	$34_{0}^{1}$
N	13,436	986	$39_{0}^{1} 42_{0}^{1}$
O	13,444	994	$42_{0}^{1} 59_{0}^{1} 61_{0}^{1}$
P	13,459	1,010	$38_{0}^{1} 42_{0}^{1}$
Q	13,541	1,091	$30_{0}^{1}$
R	13,562	1,113	$39_{0}^{1} 40_{0}^{1}$
S	13,572	1,122	$40_{0}^{1} 59_{0}^{1} 61_{0}^{1}$
T	13,587	1,138	$38_{0}^{1} 40_{0}^{1}$
U	13,655	1,205	$39_{0}^{2}, 39_{0}^{1} 59_{0}^{1} 61_{0}^{1}$
V	13,669	1,219	$38_{0}^{1} 39_{0}^{1}$
W	13,678	1,228	$38_{0}^{1} 59_{0}^{1} 61_{0}^{1}$
X	13,693	1,243	$38_{0}^{2}$
Y	13,729	1,279	$34_{0}^{1} 42_{0}^{1}$
a	24,665	0	$0_{0}^{0}$	${\tilde{A}}^{2} A^{″} \leftarrow {\tilde{X}}^{1} A'$
b	24,920	255	$60_{0}^{1}$
c	25,161	496	$56_{0}^{1}$

Open in a new tab

The 2-anthracenyl radical $\tilde{X}$ band in Fig. 1C shows contamination from 1-anthracenyl ground-state features; those peaks that derive from 1-anthracenyl are shaded gray. The distinct binding energies and FC profiles of the two bands permit nearly full assignment of the 2-anthracenyl ground-state features despite the contamination. The instrumental resolution is sufficient to distinguish the ν₄₁ and ν₄₂ fundamentals (peaks C/D; Fig. 1C, Inset) that are split by 3 cm⁻¹; this is now possible only with our updated VMI electron optics. Nearly all other bands in the 2-anthracenyl $\tilde{X}$ spectrum can be assigned to progressions and combination bands of the observed fundamentals (Table S3). As was seen in the 1-anthracenyl $\tilde{X}$ band, we observe a feature not predicted by FC simulation (peak G), again 8 cm⁻¹ above the ν₃₉ fundamental (peak F).

The source of peaks F and G in the $\tilde{X}$ -state spectra of the 1- and 2-anthracenyl radicals is subtle. The PAD and threshold behavior of these peaks is the same as that of the neighboring peaks, suggesting that they have the same electronic character as the FC allowed peaks. However, the locations of peaks F and G with respect to the vibrational origins of 1- and 2-anthracenyl do not align with any calculated aʹ fundamentals or combinations of FC active modes. Each of these peaks is most likely a combination band of two a″ modes that together have aʹ vibrational symmetry, as $a^{″} \otimes a^{″} = a'$ . Such a combination band is FC allowed by symmetry, and if its frequency is very close to that of the ν₃₉ fundamental, it can borrow the intensity of the ν₃₉ fundamental through a Fermi resonance. Fermi resonances occur when two vibrational states of the same overall symmetry lie very close together and interact, leading to mixing of the two eigenstates and widening of the energy gap as the mixing states repel one another (39). Anharmonic terms in the potential energy provide the perturbation that allows energy levels associated with different normal modes to mix.

Based on calculated harmonic frequencies in Tables S7 and S8, the most likely a″ modes involved in the combination band in Fermi resonance with ν₃₉ are modes ν₅₉ and ν₆₁ for both isomers (displacements shown in Fig. S2). In 1-anthracenyl, the ν₃₉ mode is strongly FC active, and this Fermi resonance manifests several times, with combination bands and progressions of ν₃₉ appearing as doublets and multiplets. The ν₃₉ mode is less active in 2-anthracenyl and only the fundamental clearly demonstrates this extra feature. No analogous extra peaks appear in the 9-anthracenyl $\tilde{X}$ spectrum, as the ring distortion mode analogous to ν₃₉ (ν₆₁ in Table S7) is not FC allowed.

Table S7.

Vibrational frequencies (cm⁻¹) for the anion and neutral 9-anthracenyl states calculated with B3LYP/6-311+G*

Symmetry	Mode	Anion	Ground	Excited
a₁	ν₁	3,166	3,194	3,199
	ν₂	3,145	3,183	3,186
	ν₃	3,126	3,171	3,172
	ν₄	3,111	3,163	3,158
	ν₅	3,103	3,157	3,152
	ν₆	1,644	1,667	1,618
	ν₇	1,549	1,566	1,575
	ν₈	1,502	1,513	1,504
	ν₉	1,460	1,486	1,482
	ν₁₀	1,380	1,390	1,366
	ν₁₁	1,307	1,325	1,298
	ν₁₂	1,267	1,293	1,288
	ν₁₃	1,224	1,271	1,203
	ν₁₄	1,173	1,193	1,187
	ν₁₅	1,146	1,168	1,143
	ν₁₆	1,025	1,029	1,046
	ν₁₇	910	907	909
	ν₁₈	743	759	750
	ν₁₉	656	658	669
	ν₂₀	635	638	624
	ν₂₁	396	398	391
	ν₂₂	243	233	241
a₂	ν₂₃	965	984	999
	ν₂₄	929	958	973
	ν₂₅	830	854	880
	ν₂₆	748	759	760
	ν₂₇	736	739	741
	ν₂₈	507	498	495
	ν₂₉	478	470	466
	ν₃₀	233	232	217
	ν₃₁	99	114	100
b₁	ν₃₂	966	985	1,000
	ν₃₃	929	960	974
	ν₃₄	846	887	903
	ν₃₅	798	840	853
	ν₃₆	761	765	779
	ν₃₇	717	734	755
	ν₃₈	600	577	602
	ν₃₉	476	499	450
	ν₄₀	394	391	393
	ν₄₁	259	259	284
	ν₄₂	86	89	83
b₂	ν₄₃	3,165	3,194	3,199
	ν₄₄	3,144	3,183	3,186
	ν₄₅	3,126	3,171	3,172
	ν₄₆	3,106	3,159	3,155
	ν₄₇	1,630	1,660	1,593
	ν₄₈	1,593	1,622	1,574
	ν₄₉	1,522	1,573	1,513
	ν₅₀	1,461	1,478	1,451
	ν₅₁	1,405	1,417	1,418
	ν₅₂	1,358	1,377	1,339
	ν₅₃	1,304	1,364	1,303
	ν₅₄	1,230	1,238	1,261
	ν₅₅	1,178	1,200	1,197
	ν₅₆	1,153	1,170	1,175
	ν₅₇	1,107	1,125	1,114
	ν₅₈	1,021	1,025	1,046
	ν₅₉	929	933	929
	ν₆₀	820	829	837
	ν₆₁	620	614	620
	ν₆₂	536	537	529
	ν₆₃	404	384	394

Open in a new tab

Table S8.

Vibrational frequencies (cm⁻¹) for the anion and neutral 1-anthracenyl states calculated with B3LYP/6-311+G*

Symmetry	Mode	Anion	Ground	Excited
a′	ν₁	3,165	3,190	3,196
	ν₂	3,153	3,183	3,185
	ν₃	3,150	3,178	3,183
	ν₄	3,138	3,175	3,173
	ν₅	3,128	3,171	3,170
	ν₆	3,125	3,164	3,166
	ν₇	3,101	3,160	3,147
	ν₈	3,075	3,159	3,145
	ν₉	3,026	3,154	3,116
	ν₁₀	1,646	1,670	1,640
	ν₁₁	1,614	1,660	1,592
	ν₁₂	1,597	1,622	1,586
	ν₁₃	1,572	1,588	1,535
	ν₁₄	1,516	1,549	1,529
	ν₁₅	1,503	1,505	1,501
	ν₁₆	1,461	1,484	1,463
	ν₁₇	1,437	1,450	1,443
	ν₁₈	1,409	1,425	1,413
	ν₁₉	1,374	1,403	1,381
	ν₂₀	1,364	1,383	1,378
	ν₂₁	1,336	1,363	1,336
	ν₂₂	1,310	1,326	1,305
	ν₂₃	1,283	1,297	1,291
	ν₂₄	1,279	1,288	1,282
	ν₂₅	1,241	1,246	1,233
	ν₂₆	1,186	1,197	1,199
	ν₂₇	1,170	1,180	1,185
	ν₂₈	1,163	1,177	1,174
	ν₂₉	1,152	1,165	1,155
	ν₃₀	1,099	1,113	1,098
	ν₃₁	1,024	1,032	1,046
	ν₃₂	1,017	1,025	1,033
	ν₃₃	926	926	926
	ν₃₄	911	902	910
	ν₃₅	811	822	820
	ν₃₆	750	762	756
	ν₃₇	661	655	667
	ν₃₈	638	639	633
	ν₃₉	607	618	596
	ν₄₀	536	526	523
	ν₄₁	401	400	397
	ν₄₂	394	395	391
	ν₄₃	238	236	238
a′′	ν₄₄	950	985	1,000
	ν₄₅	947	962	993
	ν₄₆	935	958	970
	ν₄₇	914	904	944
	ν₄₈	863	889	919
	ν₄₉	856	861	890
	ν₅₀	819	843	860
	ν₅₁	765	772	768
	ν₅₂	744	761	756
	ν₅₃	739	746	750
	ν₅₄	715	714	748
	ν₅₅	589	570	597
	ν₅₆	502	509	504
	ν₅₇	476	483	478
	ν₅₈	468	470	441
	ν₅₉	388	390	385
	ν₆₀	247	262	256
	ν₆₁	223	233	210
	ν₆₂	105	118	107
	ν₆₃	89	91	83

Open in a new tab

Fig. S2. — (A and B) Out-of-plane vibrational modes appearing in combination as Fermi resonances in the ground-state photoelectron spectra of (A) 1-anthracenyl and (B) 2-anthracenyl.

The experimental and simulated values of β describing the anisotropies of the 9-, 1-, and 2-anthracenyl $\tilde{X}$ bands are shown in Fig. 3A. The PADs for detachment to the radical ground states differ based on relative s and p contributions to the s-p hybrid Dyson orbitals of the three isomers (40). Detachment from s-like orbitals yields predominantly $l = 1$ photoelectrons with parallel PADs, whereas detachment from p-like orbitals yields both isotropic $l = 0$ photoelectrons and perpendicularly polarized $l = 2$ photoelectrons. The relative contributions of the s and p detachment channels significantly affect the PAD (41), allowing for very different observed anisotropies for the three isomers.

By these metrics, the 9-anthracenyl $\tilde{X}$ -state Dyson MO has the largest contribution of s character, whereas 1- and 2-anthracenyl have respectively higher fractions of p character. We propose an explanation by considering a model sp²-hybridized σ-bonding network for the anthracenyl anions. In each anion, the deprotonated carbon has three sp² orbitals: two participating in C−C σ bonds and one containing the electron lone pair, from which an electron is detached to form the ground-state radical. In the 9-anthracenyl anion, the deprotonated C₉ atom is adjacent to two tertiary carbons, C_9a and C_8a (Fig. 1A). The two C₉ sp² bonding orbitals therefore have geometric overlap with the other four C−C σ bonds in which C_9a and C_8a participate. This overlap delocalizes and stabilizes the longer-range p character of the sp² bonding orbitals, leaving more s character in the lone pair orbital. In the 1-anthracenyl anion, C₁ is bonded to just one tertiary carbon, whereas in 2-anthracenyl, C₂ adjoins only secondary carbons. In these anions, the deprotonated carbon sp² bonding orbitals therefore have less p stabilization due to fewer nearby C–C σ bonds with which to overlap. The lone pair orbital thus exhibits more p character in the 1-anthracenyl anion and more yet in 2-anthracenyl.

The energetic ordering of the anthracenyl anions is closely related to these effects. Papas et al. (29) make the steric argument that the 1- and 2-anthracenyl anions are lifted in energy relative to 9-anthracenyl due to greater repulsion between the excess charge and H atoms bonded to secondary carbons adjacent to the deprotonation site. The repulsion of the lone pair also narrows the C−C−C interior bond angle at the deprotonated site, and the ability of each isomer structure to accommodate this strain contributes to the energetic ordering of the anions. It is compelling that the radical $\tilde{A}$ states follow the same energetic ordering as the anions, owing to similar hybridization and steric arguments in accommodating the doubly occupied lone pair MO. With no full lone pair to accommodate, the ground-state anthracenyl radicals fall much closer together in energy.

Radical Excited States.

The anthracenyl radical $\tilde{A}$ bands have very intense vibrational origins and weak FC activity beyond the origins, indicating little change in geometry upon photodetachment from the anion to the radical excited state. The calculated excited-state geometries accordingly show little displacement from those of the anions (Tables S4–S6). Dyson MOs for these transitions (Fig. 4) are delocalized over the ring system, consistent with little perturbation in geometry upon removal of an electron.

The $\tilde{A}$ bands are poorly resolved compared with the $\tilde{X}$ bands, making spectral assignments challenging beyond identification of the vibrational origin. The low $\tilde{A}$ -state photodetachment cross sections close to threshold cause features to vanish before they can be narrowly resolved with SEVI. According to the Wigner threshold law for photodetachment, the near-threshold cross section (σ) is a function of eKE and l, the angular momentum of the nascent photoelectron (42):

σ \propto {(eKE)}^{l + 1 / 2} .

[2]

The Dyson MOs for the anthracenyl radical $\tilde{A}$ state, shown in Fig. 4, have nodal structure analogous to that of an atomic g orbital, in which the orbital angular momentum of the electron is $l = 4$ . In a one-photon, one-electron process, a photoelectron is detached from an atomic orbital with $Δ l = \pm 1$ (43). By extension, after detachment to form the anthracenyl $\tilde{A}$ state, the outgoing photoelectron must have at least $l = 3$ , which yields vanishing σ at small eKE according to Eq. 2. The Dyson orbitals for the $\tilde{X}$ states, on the other hand, have localized s-p hybrid character, enabling partial $l = 0$ detachment, and significant intensity at small eKE (40). Photodetachment cross sections calculated as a function of eKE are shown in Fig. 4 and reproduce this difference in threshold behavior for the $\tilde{X}$ and $\tilde{A}$ states of all three isomers.

The 1-anthracenyl $\tilde{A}$ state displays some interesting behavior that warrants additional comment. At photodetachment wavelengths of 27,500–27,600 cm⁻¹, the $\tilde{A}$ band shows increased cross section and enhanced intensity of peaks b and c relative to the origin, demonstrated by the purple trace in Fig. 2B. Additionally, peaks b and c have distinct parallel polarization, in contrast to the consistent slightly perpendicular PAD of the vibrational origin.

The strong wavelength-dependent change in photodetachment cross section in combination with the appearance of features with distinct PADs suggests a contributing autodetachment mechanism. If a metastable anion state is embedded in the neutral-plus-free-electron continuum, resonant transitions to this state can compete with direct photodetachment. The anion excited state can then autodetach, resulting in electron signals whose cross section and anisotropy are not governed by the considerations discussed above. The closed-shell anthracenyl anions should have similar electronic structure to that of anthracene, whose S₁←S₀ band origin lies at 27,687 cm⁻¹ (22), so the presence of a 1-anthracenyl anion resonance at excitation wavelengths of 27,500–27,600 cm⁻¹ is not surprising.

Conclusions

We have acquired isomer-specific slow photoelectron velocity-map imaging spectra of cold 9-, 1-, and 2-anthracenyl anions, using a newly designed VMI electrostatic lens. We observe detailed vibronic structure of the largely unstudied neutral radicals. The radical ground-state spectra are fully vibrationally resolved, allowing measurement of precise electron affinities and Franck–Condon active vibrational frequencies. Transitions to the radical first excited states are also measured and term energies reported, although poor threshold photodetachment cross sections prevent full vibrational resolution as achieved for the radical ground states.

This work shows how improvements in high-resolution photoelectron imaging, in combination with cryogenic ion cooling and techniques for isomer selection, can be used to untangle the vibronic structure of increasingly complex and diverse radical systems. The subtle structural and electronic differences between the 9-, 1-, and 2-anthracenyl radicals are made clear in the cryo-SEVI spectra and photoelectron angular distributions presented here. These results inform our fundamental understanding of the chemistry of these species, as well as their behavior in the context of astrochemistry and combustion.

Materials and Methods

Experimental.

In cryo-SEVI, cold, mass-selected anions are photodetached with a tunable laser. The kinetic energies of the resulting photoelectrons are measured with a velocity-map imaging spectrometer. Low extraction voltages are used to preferentially detect slow electrons, magnifying their image on a position-sensitive detector. Cryo-SEVI and the current configuration of our instrument have been described elsewhere (33, 44, 45).

Specific anthracenyl anion isomers are prepared by flowing trace NF₃ in helium gas over a reservoir containing 9-, 1-, or 2-(trimethylsilyl)-anthracene heated to 40–60 °C. We expand this mixture through an Even–Lavie pulsed valve (46) operating at 20 Hz and fitted with a circular filament ionizer. Electrons from the ionizer induce dissociative electron attachment of NF₃ to produce F⁻. F⁻ then reacts with 9-, 1-, or 2-(trimethylsilyl)-anthracene, selectively forming the corresponding anthracenyl anion due to the strength of the silicon–fluorine bond (47).

The three trimethylsilyl-anthracene precursors were synthesized from the corresponding bromoanthracenes according to the procedure described by Marcinow et al. (48). The 9-bromoanthracene (TCI; >99%) and 1-bromoanthracene (TCI; >97%) were obtained commercially, whereas 2-bromoanthracene was synthesized from 2-aminoanthracene (Sigma-Aldrich; 96%). The details of these syntheses and NMR characterization of products are reported in SI Materials and Methods.

After production, the anthracenyl anions are directed through a radio-frequency (RF) hexapole ion guide and a quadrupole mass filter and into a linear RF octopole ion trap. The trap is held at 5 K and filled with pulsed bursts of precooled buffer gas in a mixture of 80% He and 20% H₂. The ions are stored in the trap for around 40 ms, enabling collisional cooling to their ground vibrational state (45). The cooled ions are extracted from the trap into an orthogonal time-of-flight mass spectrometer.

After mass selection, the anions reach the interaction region inside the VMI spectrometer, where they are photodetached with the output from an Nd:YAG-pumped tunable dye laser. The VMI electrostatic lens focuses the photoelectrons onto an imaging detector comprising a pair of chevron-stacked microchannel plates coupled to a phosphor screen (49). We have recently updated our VMI lens to improve the focusing of photoelectrons onto the detector; the details of this design are described in Materials and Methods, Modification of the VMI Lens.

Electron events on the imaging detector are recorded with a 768 × 1,024-pixel CCD camera. We use event-counting software to identify single electron events and compute their centroids during data acquisition (50). It is essential to bin centroids into a grid that is fine enough not to limit the resolution of the spectrometer. The radial and angular electron distributions are reconstructed from the accumulated image, using the maximum-entropy velocity Legendre reconstruction method (51).

We calibrate the velocity and hence eKE of the photoelectrons as a function of radial displacement, R, using SEVI images of atomic O⁻ and F⁻ (52, 53) taken at many photon energies. The distance of an electron spot from the center of the reconstructed image is largely linearly proportional to its velocity following photodetachment. However, the added length and lensing properties of the modified VMI design slightly distort the electron velocities, so an additional polynomial term is needed; we calibrate it by fitting $eKE = a R^{2} + b R^{4}$ . The eKE spectrum can then be converted to eBE, using the energy conservation expression $eBE = h ν - eKE$ , where $h ν$ is the photon energy.

Modification of the VMI Lens.

We have recently modified the original three-electrode Eppink–Parker VMI design (54) to improve the energy resolution of our cryo-SEVI spectra. The redesigned VMI lens is shown in Fig. 5. The major goal of this modification was to improve the energy resolution for eKEs in the range of a few hundred to a few thousand wavenumbers. This is the regime in which SEVI is forced to operate for anions with poor threshold photodetachment cross sections. Additionally, a SEVI spectrum over a particular energy range can now be obtained with fewer high-resolution windows.

Fig. 5. — Configuration of the VMI lens, with electrodes in gray and Vespel spacers in brown. Ions enter the lens from the left and interact with the laser halfway between plates 1 and 2, indicated in red. All lengths are in millimeters.

Several alternative VMI designs with excellent focusing capability for photoelectrons have been reported in the literature (55–57). The key changes made in these designs include adding a fourth electrode whose voltage is adjusted to optimize focus (55, 57) and increasing the total length of the VMI stack (56). Increasing the length of the VMI lens along the electron flight axis can significantly improve its focusing ability, as the volume of the interaction region is smaller relative to the size of the lens. Additionally, lengthening the stack while maintaining a fixed repeller voltage results in a lower electric field gradient in the interaction region, which can reduce the importance of laser positioning and stabilize the velocity calibration of the resulting images (58). However, adding distance between electrodes allows fringe fields to penetrate into the lens and distort electron trajectories. This effect can be corrected for either by adding guard rings to the existing electrodes (56) or by adding additional guarding plates to maintain a smooth voltage gradient in each acceleration region (57).

We chose to implement a design comprising three electrodes and four guarding plates for simplicity and ease of optimization. The seven-plate geometry shown in Fig. 5 keeps the repeller (plate 1), extractor (plate 4), and grounding (plate 7) electrodes of the Eppink–Parker design, but adds four guarding electrodes (plates 2, 3, 5, and 6). The plate apertures and spacings were optimized using electron trajectory simulations in SIMION 8.0 (59). All VMI plates were machined from stainless steel 304 and are coated with colloidal graphite. Four sets of Vespel spacers maintain the gap between neighboring plates. Using resistors wired in series, the guarding plates are held at voltages linearly interpolated between the repeller, extractor, and ground electrodes. The VMI is focused by tuning the ratio of the extractor voltage (V_E) to the repeller voltage (V_R).

The VMI spectrometer has a roughly constant value for $Δ eKE / eKE$ , yielding the narrowest features for the lowest-eKE transitions. Using detachment of atomic O⁻ and F⁻ to characterize the lens, we obtain our narrowest feature with a 1.2-cm⁻¹ FWHM at eKE = 5 cm⁻¹ and $V_{R} = - 170 V$ . Farther from threshold, $V_{R} = - 340 V$ provides the best compromise between image magnification on the detector and invulnerability to stray fields and yields typical resolution of 6 cm⁻¹ FWHM at eKE = 190 cm⁻¹ and 20 cm⁻¹ FWHM at eKE = 1,350 cm⁻¹, comparable to the performance reported by Wang and coworkers (57). Our resolution with the original Eppink–Parker design with a comparable V_R was 14 cm⁻¹ FWHM at eKE = 180 cm⁻¹ and 36 cm⁻¹ FWHM at eKE = 1,000 cm⁻¹. With $V_{R} = - 670 V$ , we can achieve $Δ eKE / eKE \sim 1.4 %$ for faster electrons, compared with an optimal resolution of $\sim 3 %$ with the Eppink–Parker lens.

Electronic Structure Calculations.

Density functional theory calculations were carried out at the B3LYP/6-311+G* level of theory to find the energetics, optimized geometries, normal modes, and harmonic vibrational frequencies for the 9-, 1-, and 2-anthracenyl anions and the ground and first excited states of the corresponding radicals. The maximum overlap method (60) was used for the radical excited states. Dyson orbitals for photodetachment transitions were calculated with EOM-IP-CCSD/6-311G. All ab initio work was done using Q-Chem 4.0 (61, 62).

Calculated energetics corrected for vibrational zero-point energy and relevant harmonic vibrational frequencies are summarized in Table 1; isomer energetics are summarized in Table 2. Full lists of optimized geometries and harmonic frequencies can be found in Tables S4–S9. All states considered were found to have planar equilibrium geometries, in agreement with the literature (6, 29, 31).

Table S9.

Vibrational frequencies (cm⁻¹) for the anion and neutral 2-anthracenyl states calculated with B3LYP/6-311+G*

Symmetry	Mode	Anion	Ground	Excited
a′	ν₁	3,165	3,190	3,197
	ν₂	3,150	3,184	3,184
	ν₃	3,137	3,177	3,173
	ν₄	3,130	3,164	3,168
	ν₅	3,123	3,161	3,163
	ν₆	3,120	3,160	3,161
	ν₇	3,081	3,158	3,134
	ν₈	3,033	3,157	3,108
	ν₉	3,011	3,154	3,091
	ν₁₀	1,646	1,670	1,638
	ν₁₁	1,615	1,643	1,602
	ν₁₂	1,580	1,611	1,591
	ν₁₃	1,557	1,589	1,546
	ν₁₄	1,538	1,567	1,515
	ν₁₅	1,497	1,508	1,494
	ν₁₆	1,457	1,473	1,459
	ν₁₇	1,438	1,455	1,436
	ν₁₈	1,411	1,424	1,427
	ν₁₉	1,401	1,413	1,393
	ν₂₀	1,375	1,373	1,378
	ν₂₁	1,337	1,366	1,323
	ν₂₂	1,312	1,308	1,314
	ν₂₃	1,292	1,291	1,305
	ν₂₄	1,290	1,287	1,293
	ν₂₅	1,277	1,266	1,263
	ν₂₆	1,201	1,206	1,205
	ν₂₇	1,183	1,191	1,200
	ν₂₈	1,175	1,170	1,185
	ν₂₉	1,166	1,159	1,159
	ν₃₀	1,125	1,126	1,119
	ν₃₁	1,024	1,039	1,047
	ν₃₂	993	1,029	999
	ν₃₃	923	926	920
	ν₃₄	898	894	903
	ν₃₅	812	819	821
	ν₃₆	760	760	765
	ν₃₇	659	658	663
	ν₃₈	643	639	634
	ν₃₉	620	609	606
	ν₄₀	528	533	515
	ν₄₁	398	398	395
	ν₄₂	393	396	388
	ν₄₃	234	238	233
a′′	ν₄₄	960	985	999
	ν₄₅	945	960	993
	ν₄₆	928	953	964
	ν₄₇	903	905	932
	ν₄₈	874	891	915
	ν₄₉	858	849	888
	ν₅₀	817	822	856
	ν₅₁	776	780	819
	ν₅₂	768	767	768
	ν₅₃	741	741	755
	ν₅₄	724	713	744
	ν₅₅	577	574	568
	ν₅₆	472	503	480
	ν₅₇	466	477	446
	ν₅₈	422	461	407
	ν₅₉	363	378	357
	ν₆₀	236	271	216
	ν₆₁	215	237	189
	ν₆₂	109	119	99
	ν₆₃	92	93	86

Open in a new tab

Photoelectron spectra for detachment to the ground and excited states of the 9-, 1-, and 2-anthracenyl radicals were simulated at 0 K, using the ezSpectrum program (63). The ab initio geometries, normal modes, and harmonic frequencies were given as input, and Franck–Condon overlap factors were calculated in the harmonic approximation with Duschinsky mixing of all modes (64). The vibrational origins of all bands were shifted in eBE to align with their experimental values. Those frequencies of the neutral ground states that were observed in the high-resolution SEVI spectra were scaled to their experimental values; the others were left at their calculated B3LYP/6-311+G* values.

The photoelectron angular distributions and photodetachment cross sections for transitions to the neutral ground and excited states were calculated as functions of eKE with the ezDyson program (38, 65). ezDyson takes as input the ab initio Dyson orbitals for the relevant photodetachment transitions and finds the contribution of partial spherical waves with angular momentum $l \leq 4$ to the wavefunction of the outgoing photoelectron.

SI Materials and Methods

Synthesis of 2-Bromoanthracene.

A suspension of 2-aminoanthracene (3.0 g, 11.6 mmol) in bromoform (15 mL) was heated to 100 °C. Amyl nitrite (1.89 g, 16.1 mmol) was added portion-wise and the reaction mixture was stirred for 2 h at 100 °C. The volatile materials were removed under vacuum to afford a dark residue, which was dissolved in hexanes and passed through a pad of silica gel to remove most of the colored impurities. The hexanes were removed by rotary evaporation to afford 2-bromoanthracene (15% yield).

2-Bromoanthracene: ¹H NMR (500 MHz, CDCl₃) δ 8.40 (s, 1H), 8.32 (s, 1H), 8.17 (br s, 1H), 8.07–7.96 (m, 2H), 7.88 (d, J = 9.0 Hz, 1H), 7.53–7.45 (m, 3H).

Synthesis of 9-, 1-, and 2-(Trimethylsilyl)-Anthracene.

9-Bromoanthracene (2.5 g, 9.72 mmol) was dissolved in dry THF (100 mL) and cooled to −78 °C under a N₂ atmosphere. A solution of n-BuLi (1.6 M in hexanes) was added dropwise and the reaction mixture was stirred at −78 °C for 1 h. The previously colorless reaction mixture turned orange and then became cloudy. Trimethylsilylchloride (1.6 mL, 12.6 mmol) was added dropwise at −78 °C and the reaction mixture was stirred for 20 min. The reaction mixture was allowed to warm to room temperature with stirring for 1 h, at which point no orange color remained, and then poured into a separatory funnel. Water (100 mL) and diethyl ether (20 mL) were added, and the organic phase was washed with water (2 × 100 mL) and dried over MgSO₄. Volatile materials were removed by rotary evaporation to afford the pure solid product (2.25 g, 92% yield). The 1- and 2-(trimethylsilyl)-anthracenes were prepared by the same method and were both obtained as pure solids.

9-(Trimethylsilyl)-anthracene: ¹H NMR (600 MHz, CDCl₃) δ 8.50–8.45 (m, 3H), 8.05–7.99 (m, 2H), 7.53–7.42 (m, 4H), 0.74 (s, 9H). ¹³C NMR (151 MHz, CDCl₃) δ 137.01, 135.69, 131.34, 129.92, 129.50, 128.70, 124.76, 124.50, 4.57.

1-(Trimethylsilyl)-anthracene: ¹H NMR (400 MHz, CDCl₃) δ 8.64 (s, 1H), 8.44 (d, J = 4.4 Hz, 2H), 8.07–7.92 (m, 3H), 7.69 (dd, J = 6.5, 1.3 Hz, 1H), 7.55–7.37 (m, 2H), 0.55 (s, 9H). ¹³C NMR (151 MHz, CDCl₃) δ 138.33, 134.64, 133.26, 131.85, 131.42, 131.15, 130.06, 128.47, 127.91, 127.35, 126.80, 125.41, 125.36, 124.68, 0.36.

2-(Trimethylsilyl)-anthracene: ¹H NMR (500 MHz, CDCl₃) δ 8.44 (br s, 1H), 8.41 (br s, 1H), 8.19 (br s, 1H), 8.08–7.94 (m, 3H), 7.58 (br d, J = 8.4 Hz, 1H), 7.47 (m, 2H), 0.39 (s, 9H). ¹³C NMR (151 MHz, CDCl₃) δ 137.45, 134.38, 131.97, 131.28, 128.86, 128.35, 128.17, 127.07, 126.44, 126.22, 125.96, 125.41, 125.35, 125.23, −1.11.

Acknowledgments

This work was supported by the Director, Office of Basic Energy Sciences, Chemical Sciences Division of the US Department of Energy under Contract DE-AC02-05CH11231. M.L.W. and D.S.L. thank the National Science Foundation for graduate research fellowships.

Footnotes

The authors declare no conflict of interest.

See QnAs on page 1679.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1520862113/-/DCSupplemental.

References

1.Barckholtz C, Barckholtz TA, Hadad CM. C−H and N−H bond dissociation energies of small aromatic hydrocarbons. J Am Chem Soc. 1999;121(3):491–500. [Google Scholar]
2.Calcote HF. Mechanisms of soot nucleation in flames—a critical review. Combust Flame. 1981;42(3):215–242. [Google Scholar]
3.Richter H, et al. Detailed kinetic study of the growth of small polycyclic aromatic hydrocarbons. 1. 1-naphthyl + ethyne. J Phys Chem A. 2001;105(9):1561–1573. [Google Scholar]
4.Harvey RG, Geacintov NE. Intercalation and binding of carcinogenic hydrocarbon metabolites to nucleic acids. Acc Chem Res. 1988;21(2):66–73. [Google Scholar]
5.Durant JL, Busby WF, Jr, Lafleur AL, Penman BW, Crespi CL. Human cell mutagenicity of oxygenated, nitrated and unsubstituted polycyclic aromatic hydrocarbons associated with urban aerosols. Mutat Res. 1996;371(3–4):123–157. doi: 10.1016/s0165-1218(96)90103-2. [DOI] [PubMed] [Google Scholar]
6.Gao J, Berden G, Oomens J. Laboratory infrared spectroscopy of gaseous negatively charged polyaromatic hydrocarbons. Astrophys J. 2014;787(2):170–180. [Google Scholar]
7.Puget JL, Léger A. A new component of the interstellar matter: Small grains and large aromatic molecules. Annu Rev Astron Astrophys. 1989;27:161–198. [Google Scholar]
8.Allamandola LJ, Tielens AGGM, Barker JR. Interstellar polycyclic aromatic hydrocarbons: The infrared emission bands, the excitation/emission mechanism, and the astrophysical implications. Astrophys J Suppl Ser. 1989;71(4):733–775. doi: 10.1086/191396. [DOI] [PubMed] [Google Scholar]
9.Tielens AGGM. Interstellar polycyclic aromatic hydrocarbon molecules. Annu Rev Astron Astrophys. 2008;46:289–337. [Google Scholar]
10.Parker DSN, et al. Low temperature formation of naphthalene and its role in the synthesis of PAHs (polycyclic aromatic hydrocarbons) in the interstellar medium. Proc Natl Acad Sci USA. 2012;109(1):53–58. doi: 10.1073/pnas.1113827108. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Kaiser RI, Parker DSN, Mebel AM. Reaction dynamics in astrochemistry: Low-temperature pathways to polycyclic aromatic hydrocarbons in the interstellar medium. Annu Rev Phys Chem. 2015;66:43–67. doi: 10.1146/annurev-physchem-040214-121502. [DOI] [PubMed] [Google Scholar]
12.Cherchneff I. The formation of polycyclic aromatic hydrocarbons in evolved circumstellar environments. EAS Publications Ser. 2011;46:177–189. [Google Scholar]
13.Demarais NJ, et al. Gas-phase reactions of polycyclic aromatic hydrocarbon anions with molecules of interstellar relevance. Astrophys J. 2012;746(1):32–37. [Google Scholar]
14.Vuong MH, Foing BH. Dehydrogenation of polycyclic aromatic hydrocarbons in the diffuse interstellar medium. Astron Astrophys. 2000;363(1):L5–L8. [Google Scholar]
15.Le Page V, Snow TP, Bierbaum VM. Hydrogenation and charge states of polycyclic aromatic hydrocarbons in diffuse clouds. II. Results. Astrophys J. 2003;584(1):316–330. [Google Scholar]
16.Wakelam V, Herbst E. Polycyclic aromatic hydrocarbons in dense cloud chemistry. Astrophys J. 2008;680(1):371–383. [Google Scholar]
17.Weichman ML, Kim JB, DeVine JA, Levine DS, Neumark DM. Vibrational and electronic structure of the α- and β-naphthyl radicals via slow photoelectron velocity-map imaging. J Am Chem Soc. 2015;137(4):1420–1423. doi: 10.1021/ja5124896. [DOI] [PubMed] [Google Scholar]
18.Szczepanski J, Vala M, Talbi D, Parisel O, Ellinger Y. Electronic and vibrational spectra of matrix isolated anthracene radical cations: Experimental and theoretical aspects. J Chem Phys. 1993;98(6):4494–4511. [Google Scholar]
19.Hudgins DM, Sandford SA. Infrared spectroscopy of matrix isolated polycyclic aromatic hydrocarbons. 1. PAHs containing two to four rings. J Phys Chem A. 1998;102(2):329–343. [Google Scholar]
20.Huneycutt AJ, et al. Infrared cavity ringdown spectroscopy of jet-cooled polycyclic aromatic hydrocarbons. ChemPhysChem. 2004;5(3):321–326. doi: 10.1002/cphc.200300776. [DOI] [PubMed] [Google Scholar]
21.Martin JML, El-Yazal J, François J-P. Structure and vibrational spectrum of some polycyclic aromatic compounds studied by density functional theory. 1. Naphthalene, azulene, phenanthrene, and anthracene. J Phys Chem. 1996;100(38):15358–15367. [Google Scholar]
22.Staicu A, Rouillé G, Sukhorukov O, Henning T, Huisken F. Cavity ring-down laser absorption spectroscopy of jet-cooled anthracene. Mol Phys. 2004;102(16–17):1777–1783. [Google Scholar]
23.DeCoster J, et al. PAH emissions from high-temperature oxidation of vaporized anthracene. Proc Combust Inst. 2007;31(1):491–499. [Google Scholar]
24.Ananthula R, Yamada T, Taylor PH. Kinetics of OH radical reaction with anthracene and anthracene-d10. J Phys Chem A. 2006;110(10):3559–3566. doi: 10.1021/jp054301c. [DOI] [PubMed] [Google Scholar]
25.Kasai PH, Clark PA, Whipple EB. The electronic ground states of aryl radicals. J Am Chem Soc. 1970;92(9):2640–2644. [Google Scholar]
26.Chen RH, Kafafi SA, Stein SE. Reactivity of polycyclic aromatic aryl radicals. J Am Chem Soc. 1989;111(4):1418–1423. [Google Scholar]
27.Syka JEP, Coon JJ, Schroeder MJ, Shabanowitz J, Hunt DF. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc Natl Acad Sci USA. 2004;101(26):9528–9533. doi: 10.1073/pnas.0402700101. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Cioslowski J, Liu G, Martinov M, Piskorz P, Moncrieff D. Energetics and site specificity of the homolytic C−H bond cleavage in benzenoid hydrocarbons: An ab initio electronic structure study. J Am Chem Soc. 1996;118(22):5261–5264. [Google Scholar]
29.Papas BN, Wang S, DeYonker NJ, Woodcock HL, Schaefer HF. Naphthalenyl, anthracenyl, tetracenyl, and pentacenyl radicals and their anions. J Phys Chem A. 2003;107(32):6311–6316. [Google Scholar]
30.Unterreiner BV, Sierka M, Ahlrichs R. Reaction pathways for growth of polycyclic aromatic hydrocarbons under combustion conditions, a DFT study. Phys Chem Chem Phys. 2004;6(18):4377–4384. [Google Scholar]
31.Kokkin DL, Schmidt TW. On the electronic properties of dehydrogenated polycyclic aromatic hydrocarbons. J Phys Chem A. 2006;110(18):6173–6177. doi: 10.1021/jp057441v. [DOI] [PubMed] [Google Scholar]
32.Rossi MJ. Thermochemical properties from ab initio calculations: π- and σ-free radicals of importance in soot formation: •C3H3 (propargyl), •C4H3, •C13H9 (phenalenyl), •C6H5 (phenyl), •C10H7 (naphthyl), •C14H9 (anthryl), •C14H9 (phenanthryl), •C16H9 (pyrenyl), •C12H7 (acenaphthyl), and •C12H9 (biphenylyl) Int J Chem Kinet. 2008;40(7):395–415. [Google Scholar]
33.Neumark DM. Slow electron velocity-map imaging of negative ions: Applications to spectroscopy and dynamics. J Phys Chem A. 2008;112(51):13287–13301. doi: 10.1021/jp807182q. [DOI] [PubMed] [Google Scholar]
34.Lineberger WC. Once upon anion: A tale of photodetachment. Annu Rev Phys Chem. 2013;64:21–36. doi: 10.1146/annurev-physchem-032511-143753. [DOI] [PubMed] [Google Scholar]
35.Kim JB, Weichman ML, Neumark DM. Structural isomers of Ti2O4 and Zr2O4 anions identified by slow photoelectron velocity-map imaging spectroscopy. J Am Chem Soc. 2014;136(19):7159–7168. doi: 10.1021/ja502713v. [DOI] [PubMed] [Google Scholar]
36.Weichman ML, Kim JB, Neumark DM. Slow photoelectron velocity-map imaging spectroscopy of the ortho-hydroxyphenoxide anion. J Phys Chem A. 2015;119(23):6140–6147. doi: 10.1021/acs.jpca.5b00768. [DOI] [PubMed] [Google Scholar]
37.Cooper J, Zare RN. Angular distribution of photoelectrons. J Chem Phys. 1968;48(2):942–943. [Google Scholar]
38.Oana CM, Krylov AI. Cross sections and photoelectron angular distributions in photodetachment from negative ions using equation-of-motion coupled-cluster Dyson orbitals. J Chem Phys. 2009;131(12):124114. doi: 10.1063/1.3231143. [DOI] [PubMed] [Google Scholar]
39.Herzberg G. Vol 2 Van Nostrand; New York: 1945. Infrared and Raman Spectra of Polyatomic Molecules, Molecular Spectra and Molecular Structure. [Google Scholar]
40.Culberson LM, Blackstone CC, Sanov A. Photoelectron angular distributions of pyridinide: A benchmark application of the mixed s-p model to a truly polyatomic anion. J Phys Chem A. 2013;117(46):11760–11765. doi: 10.1021/jp402507v. [DOI] [PubMed] [Google Scholar]
41.Culberson LM, Blackstone CC, Wallace AA, Sanov A. Aromatic stabilization and hybridization trends in photoelectron imaging of heterocyclic radicals and anions. J Phys Chem A. 2015;119(38):9770–9777. doi: 10.1021/acs.jpca.5b04672. [DOI] [PubMed] [Google Scholar]
42.Wigner EP. On the behavior of cross sections near thresholds. Phys Rev. 1948;73(9):1002–1009. [Google Scholar]
43.Mabbs R, Grumbling ER, Pichugin K, Sanov A. Photoelectron imaging: An experimental window into electronic structure. Chem Soc Rev. 2009;38(8):2169–2177. doi: 10.1039/b815748k. [DOI] [PubMed] [Google Scholar]
44.Osterwalder A, Nee MJ, Zhou J, Neumark DM. High resolution photodetachment spectroscopy of negative ions via slow photoelectron imaging. J Chem Phys. 2004;121(13):6317–6322. doi: 10.1063/1.1787491. [DOI] [PubMed] [Google Scholar]
45.Hock C, Kim JB, Weichman ML, Yacovitch TI, Neumark DM. Slow photoelectron velocity-map imaging spectroscopy of cold negative ions. J Chem Phys. 2012;137(24):244201. doi: 10.1063/1.4772406. [DOI] [PubMed] [Google Scholar]
46.Even U, Jortner J, Noy D, Lavie N, Cossart-Magos C. Cooling of large molecules below 1 K and He clusters formation. J Chem Phys. 2000;112(18):8068–8071. [Google Scholar]
47.DePuy CH, et al. Gas-phase reactions of anions with substituted silanes. J Am Chem Soc. 1980;102(15):5012–5015. [Google Scholar]
48.Marcinow Z, Clawson DK, Rabideau PW. Silicon-modified Birch reduction and reductive alkylation of polynuclear aromatics. Tetrahedron. 1989;45(17):5441–5448. [Google Scholar]
49.Chandler DW, Houston PL. Two-dimensional imaging of state-selected photodissociation products detected by multiphoton ionization. J Chem Phys. 1987;87(2):1445–1447. [Google Scholar]
50.Doyle MB, Abeyasera C, Suits AG. NuACQ. Wayne State University; Detroit: 2012. [Google Scholar]
51.Dick B. Inverting ion images without Abel inversion: Maximum entropy reconstruction of velocity maps. Phys Chem Chem Phys. 2014;16(2):570–580. doi: 10.1039/c3cp53673d. [DOI] [PubMed] [Google Scholar]
52.Blondel C, et al. The electron affinities of O, Si, and S revisited with the photodetachment microscope. Eur Phys J D. 2005;33(3):335–342. [Google Scholar]
53.Blondel C, Delsart C, Goldfarb F. Electron spectrometry at the µeV level and the electron affinities of Si and F. J Phys B At Mol Opt Phys. 2001;34(9):L281–L288. [Google Scholar]
54.Eppink ATJB, Parker DH. Velocity map imaging of ions and electrons using electrostatic lenses: Application in photoelectron and photofragment ion imaging of molecular oxygen. Rev Sci Instrum. 1997;68(9):3477–3484. [Google Scholar]
55.Townsend D, Minitti MP, Suits AG. Direct current slice imaging. Rev Sci Instrum. 2003;74(4):2530–2539. [Google Scholar]
56.Cavanagh SJ, et al. High-resolution velocity-map-imaging photoelectron spectroscopy of the O− photodetachment fine-structure transitions. Phys Rev A. 2007;76(5):052708. [Google Scholar]
57.León I, Yang Z, Liu H-T, Wang L-S. The design and construction of a high-resolution velocity-map imaging apparatus for photoelectron spectroscopy studies of size-selected clusters. Rev Sci Instrum. 2014;85(8):083106. doi: 10.1063/1.4891701. [DOI] [PubMed] [Google Scholar]
58.Li W, Chambreau SD, Lahankar SA, Suits AG. Megapixel ion imaging with standard video. Rev Sci Instrum. 2005;76(6):063106. [Google Scholar]
59.Manura DJ. SIMION 8.0. Scientific Instrument Services; Ringoes, NJ: 2008. [Google Scholar]
60.Gilbert ATB, Besley NA, Gill PMW. Self-consistent field calculations of excited states using the maximum overlap method (MOM) J Phys Chem A. 2008;112(50):13164–13171. doi: 10.1021/jp801738f. [DOI] [PubMed] [Google Scholar]
61.Shao Y, et al. Advances in methods and algorithms in a modern quantum chemistry program package. Phys Chem Chem Phys. 2006;8(27):3172–3191. doi: 10.1039/b517914a. [DOI] [PubMed] [Google Scholar]
62.Krylov AI, Gill PMW. Q-Chem: An engine for innovation. WIREs Comput Mol Sci. 2013;3(3):317–326. [Google Scholar]
63.Mozhayskiy VA, Krylov AI. ezSpectrum 3.0. iOpenShell Center for Computational Studies of Electronic Structure and Spectroscopy of Open-Shell and Electronically Excited Species; Los Angeles: 2009. [Google Scholar]
64.Duschinsky F. On the interpretation of electronic spectra of polyatomic molecules. I. Concerning the Franck-Condon principle. Acta Physicochimica URSS. 1937;7:551–566. [Google Scholar]
65.Gozem S, Krylov AI. ezDyson 3.0. iOpenShell Center for Computational Studies of Electronic Structure and Spectroscopy of Open-Shell and Electronically Excited Species; Los Angeles: 2015. [Google Scholar]

[r1] 1.Barckholtz C, Barckholtz TA, Hadad CM. C−H and N−H bond dissociation energies of small aromatic hydrocarbons. J Am Chem Soc. 1999;121(3):491–500. [Google Scholar]

[r2] 2.Calcote HF. Mechanisms of soot nucleation in flames—a critical review. Combust Flame. 1981;42(3):215–242. [Google Scholar]

[r3] 3.Richter H, et al. Detailed kinetic study of the growth of small polycyclic aromatic hydrocarbons. 1. 1-naphthyl + ethyne. J Phys Chem A. 2001;105(9):1561–1573. [Google Scholar]

[r4] 4.Harvey RG, Geacintov NE. Intercalation and binding of carcinogenic hydrocarbon metabolites to nucleic acids. Acc Chem Res. 1988;21(2):66–73. [Google Scholar]

[r5] 5.Durant JL, Busby WF, Jr, Lafleur AL, Penman BW, Crespi CL. Human cell mutagenicity of oxygenated, nitrated and unsubstituted polycyclic aromatic hydrocarbons associated with urban aerosols. Mutat Res. 1996;371(3–4):123–157. doi: 10.1016/s0165-1218(96)90103-2. [DOI] [PubMed] [Google Scholar]

[r6] 6.Gao J, Berden G, Oomens J. Laboratory infrared spectroscopy of gaseous negatively charged polyaromatic hydrocarbons. Astrophys J. 2014;787(2):170–180. [Google Scholar]

[r7] 7.Puget JL, Léger A. A new component of the interstellar matter: Small grains and large aromatic molecules. Annu Rev Astron Astrophys. 1989;27:161–198. [Google Scholar]

[r8] 8.Allamandola LJ, Tielens AGGM, Barker JR. Interstellar polycyclic aromatic hydrocarbons: The infrared emission bands, the excitation/emission mechanism, and the astrophysical implications. Astrophys J Suppl Ser. 1989;71(4):733–775. doi: 10.1086/191396. [DOI] [PubMed] [Google Scholar]

[r9] 9.Tielens AGGM. Interstellar polycyclic aromatic hydrocarbon molecules. Annu Rev Astron Astrophys. 2008;46:289–337. [Google Scholar]

[r10] 10.Parker DSN, et al. Low temperature formation of naphthalene and its role in the synthesis of PAHs (polycyclic aromatic hydrocarbons) in the interstellar medium. Proc Natl Acad Sci USA. 2012;109(1):53–58. doi: 10.1073/pnas.1113827108. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r11] 11.Kaiser RI, Parker DSN, Mebel AM. Reaction dynamics in astrochemistry: Low-temperature pathways to polycyclic aromatic hydrocarbons in the interstellar medium. Annu Rev Phys Chem. 2015;66:43–67. doi: 10.1146/annurev-physchem-040214-121502. [DOI] [PubMed] [Google Scholar]

[r12] 12.Cherchneff I. The formation of polycyclic aromatic hydrocarbons in evolved circumstellar environments. EAS Publications Ser. 2011;46:177–189. [Google Scholar]

[r13] 13.Demarais NJ, et al. Gas-phase reactions of polycyclic aromatic hydrocarbon anions with molecules of interstellar relevance. Astrophys J. 2012;746(1):32–37. [Google Scholar]

[r14] 14.Vuong MH, Foing BH. Dehydrogenation of polycyclic aromatic hydrocarbons in the diffuse interstellar medium. Astron Astrophys. 2000;363(1):L5–L8. [Google Scholar]

[r15] 15.Le Page V, Snow TP, Bierbaum VM. Hydrogenation and charge states of polycyclic aromatic hydrocarbons in diffuse clouds. II. Results. Astrophys J. 2003;584(1):316–330. [Google Scholar]

[r16] 16.Wakelam V, Herbst E. Polycyclic aromatic hydrocarbons in dense cloud chemistry. Astrophys J. 2008;680(1):371–383. [Google Scholar]

[r17] 17.Weichman ML, Kim JB, DeVine JA, Levine DS, Neumark DM. Vibrational and electronic structure of the α- and β-naphthyl radicals via slow photoelectron velocity-map imaging. J Am Chem Soc. 2015;137(4):1420–1423. doi: 10.1021/ja5124896. [DOI] [PubMed] [Google Scholar]

[r18] 18.Szczepanski J, Vala M, Talbi D, Parisel O, Ellinger Y. Electronic and vibrational spectra of matrix isolated anthracene radical cations: Experimental and theoretical aspects. J Chem Phys. 1993;98(6):4494–4511. [Google Scholar]

[r19] 19.Hudgins DM, Sandford SA. Infrared spectroscopy of matrix isolated polycyclic aromatic hydrocarbons. 1. PAHs containing two to four rings. J Phys Chem A. 1998;102(2):329–343. [Google Scholar]

[r20] 20.Huneycutt AJ, et al. Infrared cavity ringdown spectroscopy of jet-cooled polycyclic aromatic hydrocarbons. ChemPhysChem. 2004;5(3):321–326. doi: 10.1002/cphc.200300776. [DOI] [PubMed] [Google Scholar]

[r21] 21.Martin JML, El-Yazal J, François J-P. Structure and vibrational spectrum of some polycyclic aromatic compounds studied by density functional theory. 1. Naphthalene, azulene, phenanthrene, and anthracene. J Phys Chem. 1996;100(38):15358–15367. [Google Scholar]

[r22] 22.Staicu A, Rouillé G, Sukhorukov O, Henning T, Huisken F. Cavity ring-down laser absorption spectroscopy of jet-cooled anthracene. Mol Phys. 2004;102(16–17):1777–1783. [Google Scholar]

[r23] 23.DeCoster J, et al. PAH emissions from high-temperature oxidation of vaporized anthracene. Proc Combust Inst. 2007;31(1):491–499. [Google Scholar]

[r24] 24.Ananthula R, Yamada T, Taylor PH. Kinetics of OH radical reaction with anthracene and anthracene-d10. J Phys Chem A. 2006;110(10):3559–3566. doi: 10.1021/jp054301c. [DOI] [PubMed] [Google Scholar]

[r25] 25.Kasai PH, Clark PA, Whipple EB. The electronic ground states of aryl radicals. J Am Chem Soc. 1970;92(9):2640–2644. [Google Scholar]

[r26] 26.Chen RH, Kafafi SA, Stein SE. Reactivity of polycyclic aromatic aryl radicals. J Am Chem Soc. 1989;111(4):1418–1423. [Google Scholar]

[r27] 27.Syka JEP, Coon JJ, Schroeder MJ, Shabanowitz J, Hunt DF. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc Natl Acad Sci USA. 2004;101(26):9528–9533. doi: 10.1073/pnas.0402700101. [DOI] [PMC free article] [PubMed] [Google Scholar]

[r28] 28.Cioslowski J, Liu G, Martinov M, Piskorz P, Moncrieff D. Energetics and site specificity of the homolytic C−H bond cleavage in benzenoid hydrocarbons: An ab initio electronic structure study. J Am Chem Soc. 1996;118(22):5261–5264. [Google Scholar]

[r29] 29.Papas BN, Wang S, DeYonker NJ, Woodcock HL, Schaefer HF. Naphthalenyl, anthracenyl, tetracenyl, and pentacenyl radicals and their anions. J Phys Chem A. 2003;107(32):6311–6316. [Google Scholar]

[r30] 30.Unterreiner BV, Sierka M, Ahlrichs R. Reaction pathways for growth of polycyclic aromatic hydrocarbons under combustion conditions, a DFT study. Phys Chem Chem Phys. 2004;6(18):4377–4384. [Google Scholar]

[r31] 31.Kokkin DL, Schmidt TW. On the electronic properties of dehydrogenated polycyclic aromatic hydrocarbons. J Phys Chem A. 2006;110(18):6173–6177. doi: 10.1021/jp057441v. [DOI] [PubMed] [Google Scholar]

[r32] 32.Rossi MJ. Thermochemical properties from ab initio calculations: π- and σ-free radicals of importance in soot formation: •C3H3 (propargyl), •C4H3, •C13H9 (phenalenyl), •C6H5 (phenyl), •C10H7 (naphthyl), •C14H9 (anthryl), •C14H9 (phenanthryl), •C16H9 (pyrenyl), •C12H7 (acenaphthyl), and •C12H9 (biphenylyl) Int J Chem Kinet. 2008;40(7):395–415. [Google Scholar]

[r33] 33.Neumark DM. Slow electron velocity-map imaging of negative ions: Applications to spectroscopy and dynamics. J Phys Chem A. 2008;112(51):13287–13301. doi: 10.1021/jp807182q. [DOI] [PubMed] [Google Scholar]

[r34] 34.Lineberger WC. Once upon anion: A tale of photodetachment. Annu Rev Phys Chem. 2013;64:21–36. doi: 10.1146/annurev-physchem-032511-143753. [DOI] [PubMed] [Google Scholar]

[r35] 35.Kim JB, Weichman ML, Neumark DM. Structural isomers of Ti2O4 and Zr2O4 anions identified by slow photoelectron velocity-map imaging spectroscopy. J Am Chem Soc. 2014;136(19):7159–7168. doi: 10.1021/ja502713v. [DOI] [PubMed] [Google Scholar]

[r36] 36.Weichman ML, Kim JB, Neumark DM. Slow photoelectron velocity-map imaging spectroscopy of the ortho-hydroxyphenoxide anion. J Phys Chem A. 2015;119(23):6140–6147. doi: 10.1021/acs.jpca.5b00768. [DOI] [PubMed] [Google Scholar]

[r37] 37.Cooper J, Zare RN. Angular distribution of photoelectrons. J Chem Phys. 1968;48(2):942–943. [Google Scholar]

[r38] 38.Oana CM, Krylov AI. Cross sections and photoelectron angular distributions in photodetachment from negative ions using equation-of-motion coupled-cluster Dyson orbitals. J Chem Phys. 2009;131(12):124114. doi: 10.1063/1.3231143. [DOI] [PubMed] [Google Scholar]

[r39] 39.Herzberg G. Vol 2 Van Nostrand; New York: 1945. Infrared and Raman Spectra of Polyatomic Molecules, Molecular Spectra and Molecular Structure. [Google Scholar]

[r40] 40.Culberson LM, Blackstone CC, Sanov A. Photoelectron angular distributions of pyridinide: A benchmark application of the mixed s-p model to a truly polyatomic anion. J Phys Chem A. 2013;117(46):11760–11765. doi: 10.1021/jp402507v. [DOI] [PubMed] [Google Scholar]

[r41] 41.Culberson LM, Blackstone CC, Wallace AA, Sanov A. Aromatic stabilization and hybridization trends in photoelectron imaging of heterocyclic radicals and anions. J Phys Chem A. 2015;119(38):9770–9777. doi: 10.1021/acs.jpca.5b04672. [DOI] [PubMed] [Google Scholar]

[r42] 42.Wigner EP. On the behavior of cross sections near thresholds. Phys Rev. 1948;73(9):1002–1009. [Google Scholar]

[r43] 43.Mabbs R, Grumbling ER, Pichugin K, Sanov A. Photoelectron imaging: An experimental window into electronic structure. Chem Soc Rev. 2009;38(8):2169–2177. doi: 10.1039/b815748k. [DOI] [PubMed] [Google Scholar]

[r44] 44.Osterwalder A, Nee MJ, Zhou J, Neumark DM. High resolution photodetachment spectroscopy of negative ions via slow photoelectron imaging. J Chem Phys. 2004;121(13):6317–6322. doi: 10.1063/1.1787491. [DOI] [PubMed] [Google Scholar]

[r45] 45.Hock C, Kim JB, Weichman ML, Yacovitch TI, Neumark DM. Slow photoelectron velocity-map imaging spectroscopy of cold negative ions. J Chem Phys. 2012;137(24):244201. doi: 10.1063/1.4772406. [DOI] [PubMed] [Google Scholar]

[r46] 46.Even U, Jortner J, Noy D, Lavie N, Cossart-Magos C. Cooling of large molecules below 1 K and He clusters formation. J Chem Phys. 2000;112(18):8068–8071. [Google Scholar]

[r47] 47.DePuy CH, et al. Gas-phase reactions of anions with substituted silanes. J Am Chem Soc. 1980;102(15):5012–5015. [Google Scholar]

[r48] 48.Marcinow Z, Clawson DK, Rabideau PW. Silicon-modified Birch reduction and reductive alkylation of polynuclear aromatics. Tetrahedron. 1989;45(17):5441–5448. [Google Scholar]

[r49] 49.Chandler DW, Houston PL. Two-dimensional imaging of state-selected photodissociation products detected by multiphoton ionization. J Chem Phys. 1987;87(2):1445–1447. [Google Scholar]

[r50] 50.Doyle MB, Abeyasera C, Suits AG. NuACQ. Wayne State University; Detroit: 2012. [Google Scholar]

[r51] 51.Dick B. Inverting ion images without Abel inversion: Maximum entropy reconstruction of velocity maps. Phys Chem Chem Phys. 2014;16(2):570–580. doi: 10.1039/c3cp53673d. [DOI] [PubMed] [Google Scholar]

[r52] 52.Blondel C, et al. The electron affinities of O, Si, and S revisited with the photodetachment microscope. Eur Phys J D. 2005;33(3):335–342. [Google Scholar]

[r53] 53.Blondel C, Delsart C, Goldfarb F. Electron spectrometry at the µeV level and the electron affinities of Si and F. J Phys B At Mol Opt Phys. 2001;34(9):L281–L288. [Google Scholar]

[r54] 54.Eppink ATJB, Parker DH. Velocity map imaging of ions and electrons using electrostatic lenses: Application in photoelectron and photofragment ion imaging of molecular oxygen. Rev Sci Instrum. 1997;68(9):3477–3484. [Google Scholar]

[r55] 55.Townsend D, Minitti MP, Suits AG. Direct current slice imaging. Rev Sci Instrum. 2003;74(4):2530–2539. [Google Scholar]

[r56] 56.Cavanagh SJ, et al. High-resolution velocity-map-imaging photoelectron spectroscopy of the O− photodetachment fine-structure transitions. Phys Rev A. 2007;76(5):052708. [Google Scholar]

[r57] 57.León I, Yang Z, Liu H-T, Wang L-S. The design and construction of a high-resolution velocity-map imaging apparatus for photoelectron spectroscopy studies of size-selected clusters. Rev Sci Instrum. 2014;85(8):083106. doi: 10.1063/1.4891701. [DOI] [PubMed] [Google Scholar]

[r58] 58.Li W, Chambreau SD, Lahankar SA, Suits AG. Megapixel ion imaging with standard video. Rev Sci Instrum. 2005;76(6):063106. [Google Scholar]

[r59] 59.Manura DJ. SIMION 8.0. Scientific Instrument Services; Ringoes, NJ: 2008. [Google Scholar]

[r60] 60.Gilbert ATB, Besley NA, Gill PMW. Self-consistent field calculations of excited states using the maximum overlap method (MOM) J Phys Chem A. 2008;112(50):13164–13171. doi: 10.1021/jp801738f. [DOI] [PubMed] [Google Scholar]

[r61] 61.Shao Y, et al. Advances in methods and algorithms in a modern quantum chemistry program package. Phys Chem Chem Phys. 2006;8(27):3172–3191. doi: 10.1039/b517914a. [DOI] [PubMed] [Google Scholar]

[r62] 62.Krylov AI, Gill PMW. Q-Chem: An engine for innovation. WIREs Comput Mol Sci. 2013;3(3):317–326. [Google Scholar]

[r63] 63.Mozhayskiy VA, Krylov AI. ezSpectrum 3.0. iOpenShell Center for Computational Studies of Electronic Structure and Spectroscopy of Open-Shell and Electronically Excited Species; Los Angeles: 2009. [Google Scholar]

[r64] 64.Duschinsky F. On the interpretation of electronic spectra of polyatomic molecules. I. Concerning the Franck-Condon principle. Acta Physicochimica URSS. 1937;7:551–566. [Google Scholar]

[r65] 65.Gozem S, Krylov AI. ezDyson 3.0. iOpenShell Center for Computational Studies of Electronic Structure and Spectroscopy of Open-Shell and Electronically Excited Species; Los Angeles: 2015. [Google Scholar]

PERMALINK

Isomer-specific vibronic structure of the 9-, 1-, and 2-anthracenyl radicals via slow photoelectron velocity-map imaging

Marissa L Weichman

Jessalyn A DeVine

Daniel S Levine

Jongjin B Kim

Daniel M Neumark

Series information

Significance

Abstract

Fig. 1.

Results and Discussion

Experimental Photodetachment Spectra.

Fig. 2.

Table 1.

Table S1.

Table S3.

Fig. 3.

Assignment of Electronic Structure.

Fig. 4.

Table 2.

Radical Ground States.

Table S4.

Table S6.

Table S5.

Fig. S1.

Table S2.

Table S7.

Table S8.

Fig. S2.

Radical Excited States.

Conclusions

Materials and Methods

Experimental.

Modification of the VMI Lens.

Fig. 5.

Electronic Structure Calculations.

Table S9.

SI Materials and Methods

Synthesis of 2-Bromoanthracene.

Synthesis of 9-, 1-, and 2-(Trimethylsilyl)-Anthracene.

Acknowledgments

Footnotes

References

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases