Nitrile Vibration Reports Induced Electric Field and Delocalization of

Subscriber access provided by UNIV OF WEST FLORIDA

A: Kinetics, Dynamics, Photochemistry, and Excited States

Nitrile Vibration Reports Induced Electric Field and Delocalization of Electron in Charge-Transfer State of Aryl Nitriles Tomoyasu Mani, and David C. Grills J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.8b08025 • Publication Date (Web): 24 Aug 2018 Downloaded from http://pubs.acs.org on August 25, 2018

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 28 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

The Journal of Physical Chemistry

Nitrile Vibration Reports Induced Electric Field and Delocalization of Electron in Charge-Transfer State of Aryl Nitriles Tomoyasu Mani*† and David C. Grills‡ † Department of Chemistry, University of Connecticut, Storrs, Connecticut 06269-3060, United States ‡ Chemistry Division, Brookhaven National Laboratory, Upton, New York 11973-5000, United States AUTHOR INFORMATION Corresponding Author

*[email protected]

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 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 2 of 28

ABSTRACT. An electric field is created upon photoinduced charge separation in electron donoracceptor (D-A) molecules. The photophysics of a prototypical D-A molecule, 4-(dimethylamino)benzonitrile (DMABN), has been under extensive investigation for decades. Here, by using the framework of the vibrational Stark effect (VSE), we show that the nitrile vibration quantifies a significant induced electric field in the intramolecular charge-transfer state of DMABN. We further demonstrate that such a phenomenon can be observed in a structurally similar aryl nitrile and that the VSE depends on solvent polarity due to dielectric screening. Our current work shows how the superb sensitivity of the nitrile vibration can be used to identify the nature of electron delocalization and quantify the induced electric field in photoinduced charge transfer processes.


2

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


1.

INTRODUCTION

Efficient utilization of solar energy requires a better understanding of the nature of electronic excited states, especially those involved in photoinduced charge transfer reactions. The separation of opposite charges, either by photoexcitation or electrochemically, produces an electric field that can play an important role in photocatalysis, solar energy conversion, and electrochemistry.1-4 4-(dimethylamino)benzonitrile (DMABN, Chart 1) is a prototype of electron donor-acceptor (D-A) molecules, in which photoexcitation leads to intramolecular charge separation. Since the discovery of its dual fluorescence more than half a century ago,5 DMABN has been under extensive investigation, but a detailed description of the photophysical mechanism still remains elusive.6-9

Chart 1. Molecular Structures of BN, DMABN, TPA-F1CN, and 15-crown-5-Ph-F1CN.


3


Page 4 of 28

The two main singlet excited states of DMABN are well established as a local excited (LE) state and an intramolecular charge-transfer (ICT) state, or charge-separated (CS) state in D-A terminology. However, the exact structures of these states and their dynamics remain under debate.8-9 Vibrational spectroscopy, including ultrafast IR and Raman spectroscopy, can provide a direct readout of structural and electronic information in the excited state.10-12 Two competing descriptions, a twisted ICT (TICT) state and a pseudo-Jahn-Teller ICT (PICT) state, were examined by studying the time-resolved vibrational signatures, most notably by probing the nitrile vibration of the BN moiety.13-17 Here, we show that the nitrile vibration in the CS state of DMABN reports a significant electric field induced by the charge separation, which can be explained in the framework of the vibrational Stark effect (VSE). We further show that a similar effect was observed in a structurally related aryl nitrile (TPA-F1CN, Chart 1), and that in a polar solvent, the


4

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


(C≡N) frequency of the CS state is identical to that of the free radical anion because of dielectric screening. Note that throughout the text, we refer to the ICT state of DMABN as the CS state, to be consistent with the case for TPA-F1CN, in which a partially separated charge-transfer (CT) state was observed prior to the fully charge-separated (CS) state.

2.

Experimental Methods 2.1 General Information.

All solvents and reagents used were obtained from standard

commercial sources and used as received unless otherwise noted. Tetrahydrofuran (THF) was purified with a Vacuum Atmospheres Solvent Purifier System installed inside an inert atmosphere glovebox. UV-vis absorption spectra were recorded with a Cary 50 Scan UV-vis spectrophotometer (Varian). Steady-state fluorescence spectra were recorded using a FluoroMax (Horiba) spectrometer. FTIR spectra were recorded on a Thermo Nicolet Nexus 670 FTIR spectrometer or JASCO FT/IR-6100 using a cell equipped with CaF2 windows. 2.2 Pulse Radiolysis.

Pulse radiolysis experiments were performed at the Laser-Electron

Accelerator Facility (LEAF) at Brookhaven National Laboratory. For UV-vis-NIR transient absorption detection, the experiments were performed as described before,18 using a quartz cell with a pathlength of 0.5 cm. For TRIR detection, a detailed description of the experimental setup is given elsewhere.19 A home-built, airtight IR solution flow cell was used (1.10 mm pathlength), with 0.35 mm thick CaF2 windows. A continuous wave external-cavity quantum cascade laser (Model 21047-MHF, Daylight Solutions, Inc.) was used as the IR probe source. The time resolution is limited to ~40 ns in the current setup. 2.3 Femtosecond IR Transient Absorption Spectroscopy (fs-IR). The fs-IR setup is based on HELIOS IR (Ultrafast Systems), coupled to a femtosecond laser system (Coherent). The system


5


Page 6 of 28

consists of an Astrella one-box Ti:Sapphire amplifier. The pulse width is ~35 fs FWHM at 800 nm with an energy of ~5.0 mJ/pulse at a 1 kHz repetition rate. The output of the amplifier is used to pump two OPerA Solo optical parametric amplifiers (Coherent), one which provides a 400 nm pump pulse and the other which provides a probe pulse in the mid-IR region (2.6-11 m). The spectra are acquired with a liquid N2-cooled dual channel (2 x 32) MCT array detector that is coupled to a HR320 monochoromator (Horiba). Datasets were processed and analyzed using the Surface Xplorer Software (Ultrafast Systems). A commercial IR cell (Harrick Scientific, 0.95 mm pathlength) was used. During the measurements, the sample was pseudo-randomly translated to avoid photo-degradation by using a translating sample holder (Ultrafast Systems). 2.4 Femtosecond UV-Visible Transient Absorption Spectroscopy (fs-TA). The fs-TA setup was based on HELIOS FIRE (Ultrafast Systems), coupled to a femtosecond laser system (Coherent). The output of the amplifier is divided by beam splitters. One beam was used as the input for an OPerA Solo optical parametric amplifier (Coherent), which provided a 330 or 400 nm pump pulse. The energy of the pump pulse was adjusted to between 1 and 4 J/pulse. Another beam (~2% of the total power) was used to generate white light continuum probe pulses in the HELIOS spectrometer. For a UV continuum (325-650 nm) the beam was focused into a CaF2 crystal, while the beam was focused into a sapphire plate for a visible continuum (400-800 nm). A 1024-pixel CMOS sensor was used for detection. A quartz cell with a path length of 0.2 cm was used. During the measurements, the sample was pseudo-randomly translated to avoid photodegradation by using a translating sample holder (Ultrafast Systems). Datasets were processed and analyzed using the Surface Xplorer Software (Ultrafast Systems) by subtracting fluorescence background (if any), applying chirp correction, performing single value decomposition, and global fitting.


6

Page 7 of 28 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


2.5 Nanosecond Step-Scan FTIR Spectroscopy. The TRIR spectrum of the S[DMA•+–BN•–] ICT state of DMABN was recorded by step-scan FTIR spectroscopy following 266 nm photoexcitation with the 4th harmonic of a pulsed Nd:YAG laser (Spectra Physics Lab-170, 6 ns, 3 mJ/pulse). A Bruker Vertex 80v FTIR spectrometer equipped with a fast risetime (20 ns) HgCdTe detector was used in step-scan mode, recording the AC-coupled detector signal with a 14-bit, 400 MS/s digitizer. A 4 mM argon-saturated solution was flowed cyclically by a Micropump through a commercial IR flow cell (Harrick Scientific, DLC-S25, 0.5 mm pathlength) in an air-tight flow system for the duration of the experiment. 2.6 Electrochemistry.

Cyclic voltammetry measurements were conducted with a 600E

Electrochemical Analyzer/Workstation (CH Instruments) in a standard three-electrode cell consisting of a 3 mm glassy carbon disk working electrode, a platinum wire counter electrode, and a pseudo Ag reference electrode in N,N-dimethylformamide (DMF) solution with 0.1 M tetrabutylammonium hexafluorophosphate (TBA+PF6-). Data was processed and analyzed using the CHI600e Software. Potentials are reported vs Fc+/0; ferrocene was added after the measurements. 2.7 Computations. Computations were carried out with Gaussian09 D.01.20 The geometries were optimized with the LC-ωPBE21-23 (ω = 0.1 bohr-1) or ωB97X-D24 functional in density functional theory (DFT) calculations. The 6-31G(d) basis set was used. Polarizable continuum models (PCM)25-27 for solvation were used when noted. All calculations on anions were spinunrestricted. All n-hexyl groups in fluorenes were replaced by ethyl groups. The geometry optimizations were performed without symmetry constraints. Frequency calculations were performed at the optimized geometries without anharmonicity corrections. Linear response time-


7


Page 8 of 28

dependent DFT (TD-DFT) calculations were performed for low-lying excited states to determine transition energies at the ground state geometries.

3.

RESULTS AND DISCUSSION

It was long assumed that the nitrile vibration of the CS state of DMABN (S[DMA•+–BN•–]) resembles that of the benzonitrile radical anion (BN•–), even in nonpolar solvents,28-31 based on IR data of BN•– obtained by chemical reduction in THF.32 Such a similarity suggested a “decoupling” of the DMA and BN moieties, indicating the formation of a TICT state.9, 31 Pulse radiolysis is a powerful technique that uses a short pulse of high energy electrons from an accelerator to initiate chemical reactions, conferring the ability to produce charged species even in solvents of low polarity. The recent development of time-resolved infrared (TRIR) spectroscopy coupled with pulse radiolysis (PR-TRIR),18-19, 33 enables us to study the IR spectra of radical ions free from the effects of ion pairing.34 Here, using a combination of PR-TRIR and laser flash photolysis-TRIR, we unambiguously show that the nitrile vibration of S[DMA•+–BN•–] is distinct from that of BN•– in the relatively nonpolar solvent, tetrahydrofuran (THF,  = 7.52): νmax = 2116 cm-1 vs 2077 cm1

DMABN BN , respectively; ∆𝜈obs = 𝜈CS − 𝜈anion = 39 cm-1, Table 1 (Figure 1).


8

Page 9 of 28 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


Figure 1. TRIR spectra of DMABN•–, BN•–, and S[DMA•+–BN•–] in THF. The spectra of DMABN•– and BN•– were recorded ~50 ns after pulse radiolysis. The spectrum of S[DMA•+–BN•– ] was recorded ~50 ns after photoexcitation with a 266 nm laser pulse. The observed nitrile band of S[DMA•+–BN•–] is consistent with a previous report.31

Table 1. Observed (C≡N) (cm−1) of Neutral, Radical Anion, CS, and CT Forms of DMABN, BN, TPA-F1CN, 15-crown-5-Ph-F1CN, and F1CNa Neutral Name

CS (or ICT)

CT

νmax

FWHM (cm-1) b

νmax

FWHM (cm-1) b

∆anc

max

FWHM (cm-1) b

∆CSd

DMABN

2215

8.39

2071

14.7

-144

2116

12.1

-99

BNf

2230

6.48

2077

10.8

-153

TPA-F1CN

2224

5.03

2132

16.9

-92

2149

20.2

-75

2222

6.40

2120

28.0

-102

2120

11.0

-102

2224

4.49

2124

18.4

-100

F1CNf

2224

4.97

2110

12.1

-114

F1CN (in DMF)h

2223

6.07

2101

21.8

-122

TPA-F1CN (in DMF) 15-crown-5Ph-F1CNg

a

Anion

max

FWHM (cm-1) b

∆CTe

2160

10.3

-64

Values are in THF unless otherwise noted. The structures of 15-crown-5-Ph-F1CN and F1CN are shown in Chart 1 and Chart S1, respectively. b Full width half maximum (FWHM) was determined by fitting the spectra with a single Voigt function, except for neutral BN which was fit by a Lorentzian. c an = anion - neutral. d  CS = CS - neutral. e CT = CT - neutral. f Reference 35. g Reference 36. h Reference 37.

The observed difference between the nitrile vibrations of S[DMA•+–BN•–] and BN•– was not previously clearly recognized, likely due to a lack of proper reference data. Two main effects contribute to the observed 39 cm-1 shift; 1) a change in electronic structure or electron density, and 2) an induced electric field (VSE) in the CS state of DMABN. Upon charge separation within DMABN, a hole is left behind in the localized amino nitrogen lone pair orbital while an electron


9


Page 10 of 28

becomes delocalized in the BN unit.38-39 This intramolecular charge separation produces an electric field within the molecule and such separation of charge is the origin of the larger dipole moment observed in the CS state compared to in the ground and LE states.40-43 When we assume that the electronic structure of the BN unit in the CS state is close to that of BN•– (the same degree of electron delocalization), the effect of a change in electronic structure becomes negligible, and the dominant effect on the IR shift comes from an induced electric field. A calculation by Dreyer and Kummrow showed that the electronic wavefunction of the CS state is fully characterized by a single electron excitation from the localized amino nitrogen lone pair to the BN unit.38 Acknowledging the complex nature of nitrile vibrational probes,44-45 we apply a Stark interpretation given that the experiments are performed in “well-behaved” non-hydrogen bonding solvents. Assuming a linear dependence of vibrational frequency on electric field, the relationship between the observed IR shift of the nitrile vibration (∆νobs) and the electric field can be expressed within the VSE framework by the following equation,45-46 Δ𝜈obs = −Δ𝐹⃗ ∆𝜇⃗CN

(1)

where Δ𝐹⃗ is the difference in the absolute electric fields that the two different electrostatic environments exert onto the nitrile probe, and ∆𝜇⃗CN is the nitrile probe’s difference in dipole moment between the vibrational ground and excited states, often expressed as the linear Stark tuning rate.47 The equation describes the change in vibrational frequency due to a difference in electric field between two states. In the current case, BN•– serves as the reference state in the absence of an electric field (𝐹⃗BN = 0 MV/cm) to determine the induced electric field exerted on the nitrile in the BN•– moiety within DMABN in the CS state (𝐹⃗DMABN ). We estimate the induced electric field within DMABN to be 𝐹⃗DMABN ~ 43 ± 5 MV/cm, based on the Stark tuning rate determined by using an Onsager-like model of solvation following the treatment by Boxer and co-


10

Page 11 of 28 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


workers (see Supporting Information)48; |∆𝜇⃗CN | = 0.90 ± 0.1 (cm−1 /(MV/cm)), assuming the quadratic Stark tuning rate is small.45, 49 Our own DFT calculations support this notion (see below). The estimated value lies within the electric fields induced by dipole-dipole interaction.45 Putting our value in the context of the electric fields in various fields previously measured by VSE, it is larger than those in lipid membranes,50 and comparable to the observed local electric field in plasmonic nanogap junctions,51 and those in the Betaine-30 excited states probed by NaSCN,52 but much smaller than those due to strong hydrogen bonding in the ketosteroid isomerase.53 The observed blue-shift of the nitrile band of S[DMA•+–BN•–] relative to that of BN•– is consistent with the positive charge (a hole), located on the DMA side of the DMABN molecule opposite to the nitrile group, inducing an electric field antiparallel to the direction of ∆𝜇⃗CN , thus increasing the transition energy, in a similar manner to the case we previously observed for a chemically-reduced, nitrile-substituted fluorene with a covalently-tethered, metal-chelating ligand (Na(15-crown-5)+Ph-F1CN•–).36 Our measured 𝐹⃗DMABN value is in good agreement with the estimated electric field comp (𝐹⃗DMABN ~ −36 MV/cm) using the dipole moment of the ICT state measured in THF (D = 19.5

Debye).54 Note here that we assumed a complete charge transfer, or decoupling, of the DMA and the BN units in our analysis (see Supporting Information). Our current result does not contradict the previous IR studies that suggest such a decoupling.28-31 Instead, the analysis further augments their arguments by showing clear spectroscopic evidence of the induced electric field that originates from a complete charge separation. Another possible interpretation of the observed shift is that it comes from the difference in electron density on the nitrile between S[DMA•+–BN•–] and BN•–. If this were the case and the shift does not reflect a VSE contribution, then it is likely that decoupling of the DMA and BN units is not complete, which would call into question the formation of a TICT state.


11


Page 12 of 28

It is also worth mentioning that the red-shift observed for the (C≡N) of DMABN•–, compared to BN•– (Figure 1), is due to the electron donating ability of the DMA moiety. A similar red-shift is observed among the neutral states as well (Table 1). The shift of the nitrile IR band between the anion and the neutral (an) is due to the degree of electron delocalization, 35, 37 and it is larger in magnitude for BN than DMABN, as the excess electron in DMABN can be more spread into the DMA unit, and extending its electron delocalization results in a smaller an. To investigate if the observed effect is general, we synthesized a structurally related aryl nitrile compound (TPA-F1CN, Chart 1) that is expected to produce a CS state like that in DMABN. Here, TPA and F1CN act as an electron donor and acceptor, respectively. The synthetic procedure and the characterization of the target compound are reported in the Supporting Information, along with absorption and emission spectra (Figure S1) and FTIR spectra of the neutral species (Figure S2). The large observed Stokes shift and featureless nature of the emission band indicate the chargetransfer nature of the emission. Reduction and oxidation potentials also suggest a favorable Gibbs energy change for photoinduced intramolecular electron transfer (Table S1). A femtosecond UVvisible transient absorption spectroscopy (fs-TA) study in THF reveals that photoexcitation at 400 nm leads to formation of the charge-transfer excited state in

Nitrile Vibration Reports Induced Electric Field and Delocalization of

Recommend Documents