Charge Localization after Ultrafast Photoexcitation of a Rigid Perylene

Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Article

Charge Localization after Ultrafast Photoexcitation of a Rigid Perylene Perylenediimide Dyad Visualized by Transient Stark Effect Marius Koch, Mykhaylo Myahkostupov, Daniel G. Oblinsky, Siwei Wang, Sofia Garakyaraghi, Felix N. Castellano, and Gregory D. Scholes J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b01630 • Publication Date (Web): 29 Mar 2017 Downloaded from http://pubs.acs.org on March 29, 2017

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 free 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 accessible to all readers and 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.

Journal of the American Chemical Society 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 10

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

Journal of the American Chemical Society

Charge Localization after Ultrafast Photoexcitation of a Rigid Perylene Perylenediimide Dyad Visualized by Transient Stark Effect Marius Koch†, Mykhaylo Myahkostupov§, Daniel G. Oblinsky†, Siwei Wang†, Sofia Garakyaraghi§, Felix N. Castellano§*, and Gregory D. Scholes†* †

Department of Chemistry, Princeton University, Washington Road, 08540 Princeton, NJ/USA §

Department of Chemistry, North Carolina State University, Raleigh, NC 27695-8204, USA

ABSTRACT: The intramolecular charge transfer (CT) dynamics of a rigid and strongly conjugated perylenediimidebridge-perylene dyad (PDIPe) has been investigated in dichloromethane using ultrafast transient electronic absorption spectroscopy and quantum chemical calculations. The strong electronic coupling between the dyad units gives rise to a CT band. Its photoexcitation forms a delocalized CT state with well-preserved ion bands despite the strong coupling. In the dyad, the electronic transition dipole moment of the electron donor perylene is aligned along the axis of the electric field vector with respect to the CT species. This alignment makes the donor sensitive to the Stark effect and thus charge density fluctuations in the CT state. Charge localization on the picosecond time scale manifests as a timedependent Stark shift in the visible region. Quantum chemical calculations reveal a twist around the acetylene bridging unit to be the responsible mechanism generating a partial to an almost complete CT state. An estimate of the electric field strength in the CT state yields approximately 25 MV/cm, which increases to around 31 MV/cm during charge localization. Furthermore, the calculations illustrate the complexity of electronic structure in this strongly delocalized superchromophore and reflect the complications in the interpretation of transient absorption results when compared to steady-state approaches such as spectroelectrochemistry and model chromophore experiments such as photoinduced bimolecular charge transfer. Exciplexes (or CT species) are common intermediates in photochemistry, since they represent the connection Introduction between neutral and charged species. Consequently, they have been observed in intra-8-11 and Photoinduced charge transfer (CT) or charge intermolecular12-18 CS in solution, OPVs19-25 and separation (CS) are key reactions towards the effective OLEDs,26-30 among others. One important goal in conversion of solar light into chemical energy. Since •– these research areas is to understand how the initially Weller’s detection of the perylene radical anion, Pe , delocalized charge density ultimately localizes in as non-radiative deactivation product of excited state 1 order to facilitate the generation of full charges and to processes in 1961 the topic has evolved into many 2-5 hinder recombination to the neutral species, a loss different directions. For photoinduced processes in mechanism in solar energy conversion. A powerful organic compounds, the fundamental interest and technique for the understanding of these ultrafast importance for potential applications in organic processes is pump-probe spectroscopy. Here, a photovoltaics (OPVs), organic light emitting diodes femtosecond pump pulse triggers an excited state (OLEDs), molecular electronics, sensing or artificial reaction whose dynamics is subsequently probed by a photosynthesis has led to a well-established second pulse also on the order of femtoseconds. A understanding over the years.6 Many of these common approach is to time-resolve electronic processes can take place on ultrafast time scales and transition bands with this type of spectroscopy using necessitate a detailed view of non-equilibrium visible light pulses. However, broad overlapping dynamics and intermediate steps that are traversed en signals in the congested visible window make it route to efficient harvesting of photon energy. difficult to distinguish species that differ only by their A typical intermediate in these explorations is an CT character (incomplete vs. complete). excited state complex (exciplex), which is an In this respect, researchers have extracted the incomplete CS state that either recombines to the information regarding CT character from vibrational neutral ground state or separates into free charges. In signatures which are narrower, usually spectrally this manuscript, CS is used when the transferred isolated and more sensitive to the local charge charge can be fully assigned to an individual molecule. density.31 In these cases, molecular vibrations have CT is used if the transfer is only partial and in this 7 provided clear transient signatures of bimolecular respect CT is an incomplete CS or exciplex. exciplexes and given structural insight not easily

ACS Paragon Plus Environment

1


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

(Scheme 1, left),35 the frequency shift has been used to probe the electrostatics of solvent fields,36 protein environments,35 amino acid residuals,37 38 transmembrane potentials and conformational changes for example in rhodopsins.39 Other areas concern solvent modes in CT processes40 or research on dye-sensitized41-42 and polymer solar cells43-44 to monitor diffusing charges. Further potential using the Stark effect as a probe was demonstrated for charge motion through delocalized states,45 based on foregoing efforts in this area,46-47 that helped tremendously to understand electron/hole separation at electron donor/acceptor (D/A) interfaces. In molecular D/A dyads the Stark effect visualizes the internal 𝐸 produced in a photogenerated ion pair.48 The cation and anion, separated by a few angstroms generate an electric field strength of multiple MV/cm if not largely shielded by the solvent.48 Using the change of the 𝐸 amplitudes, in going from neutral to ionic states after photoexcitation, was thus proposed as a possibility for ultrafast molecular switches.49-50

obtained using transient electronic absorption spectroscopy. In this article, we introduce a structural design concept for probing charge localization through an exciplex intermediate in the visible region. We visualize the intermediate through a transient Stark effect, already widely used in vibrational Stark spectroscopy.32 The Stark effect33 describes the modification of energy levels in the vicinity of an electric field, 𝐸. As a consequence, transition dipole moments, 𝜇, are influenced and detected as frequency shifts in a spectroscopy experiment. In general, for an ensemble measurement as in pumpprobe spectroscopy, the Stark effect leads to band broadening because of the measurement of isotropic distributions. For anisotropic distributions, as in constricted environments of enzymes and proteins,34 the observed frequencies are necessarily red- or blueshifted with maximized displacements for parallel and antiparallel orientation between 𝜇 and 𝐸. Since the latter is generated, for example, by charges or dipoles Orientations

µ

E

strong coupling

weak coupling

µ

E

Page 2 of 10

µ

µ

E

Ground State

E

µ

µ

µ

µ

µ δ-

µ µ

Energy Levels Higher State

E

E µ

Charge Transfer State

µ

µ δ-

1

E

1

ΔA 0

λ

-1

A

Photo Product

1

0

λ

-1

B

0

E

µ µ

δ+

E

µ

Ground State

TA Spectrum

δ+

µ -

µ

+

µ

µ -

+

λ

-1

µ

C

D

E

E

µ

F

E

G

Scheme 1: Left, A-C: Illustration of the effect of an electric field (𝐸) on molecular energy levels as a function of the angle with the transition dipole moment (𝜇). The resulting transient absorption difference spectra (DA) either shift to lower (red) or higher (blue) wavelengths, l. Right, D-G: A simplified illustration of 𝐸 and 𝜇 orientations in weakly and strongly coupled electron donor/acceptor pairs before and after photoexcitation and in possible intermediate states. Grey denotes neutral charge density, while brown and green indicate positive and negative charge density, respectively. The color contrast illustrates the amount of density on the respective molecule.

In the case of the evolution from an exciplex to a CS state in molecular D/A pairs, the magnitude of 𝐸 would increase as a function of time and thus report accordingly on the process via a time-dependent Stark shift of the bands. Both species, usually observed as one band in the visible region could then be directly distinguished, without the need to observe, for example, vibrational markers.

Why is it then, that exciplex absorption in the visible region is not observed despite the numerous investigations on D/A systems? A reasonable explanation is related to the commonly used molecular structures that do not usually incorporate three necessary structural prerequisites simultaneously.


2

Page 3 of 10

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) Electronic coupling: Weak electronic coupling between the D and A populates the CS state directly and exciplex intermediates will not be traversed (Scheme 1, D & E).51-52 Thus, sufficient electronic interaction is necessary for exciplex studies and can be present in stacked or sandwiched D/A structures at low driving forces for CT, ∆𝐺%& . 2) Alignment: However, 𝐸 orientation for stacked or sandwiched structures is usually perpendicular to the molecular plane and thus perpendicular to the probed 𝜇 (Scheme 1 F). Therefore, this geometry does not allow for the observation of a transient Stark effect.53 (Anti)parallel alignment of 𝐸 with 𝜇 is thus required and mandates a third consideration. 3) Rigidity: The constraint of conformational flexibility preserves the alignment between 𝐸 and PDI-Ref O

R1 N

𝜇 that would otherwise vanish during molecular motion and lead to an isotropic arrangement. Interestingly, a simple reflection of the molecular structure provides a way to implement the three prerequisites simultaneously. To demonstrate this concept, we have synthesized a dyad that comprises the electron donor, perylene (Pe), and the electron acceptor, perylenediimide (PDI), connected at the bay position through an acetylene bridge (Figure 1). A description of the synthesis and characterization is given in Section 3 of the Supporting Information and Figures S1-S6. We hypothesize that this dyad structure has the potential to reveal the D ()/A (–. ® D•+/A•– process in the visible region by the time-dependent influence of 𝐸 on 𝜇 of either D•+ or A•– (Scheme 1 G). Here, “𝛿” stands for partial and “•” for complete charge localization.

Pe-Ref Eox1 = 1.67 V Ered1 = -0.60 V Ered2 = -0.79 V

O

TMS

TMS

Eox1 = 1.04 V Eox2 = 1.52 V O

N R1

Ered1 = -1.55 V

O

PDIPe

R1 = R1 O

N

O

Eox1 = 1.08 V Eox2 = 1.48 V Ered1 = -0.56 V Ered2 = -0.75 V Ered3 = -1.62 V

O

N R1

O

Figure 1: (Left) Molecular structures of perylenediimide-bridge-perylene dyad, PDIPe, and reference molecules perylenediimideethynyltrimethylsilane, PDI-Ref, perylene-ethynyltrimethylsilane, Pe-Ref, with their corresponding redox potentials in dichloromethane (vs. SCE). See also Table S2. (Right) Normalized steady-state absorption and emission spectra of PDI-Ref (solid), Pe-Ref (dashed) (top) and PDIPe (bottom). Used solvents are heptane, Hep; cyclohexane, CHX; diethylether, Et2O; ethyl acetate, EtAc; tetrahydrofuran, THF; and dichloromethane, DCM; with their corresponding dielectric constants in parenthesis. Emission was detected after excitation at 410 nm (Pe-Ref), 460 nm (PDI-Ref) and 470 nm (PDIPe).

Results and discussion Steady-state Spectroscopy. The intensity-normalized absorption and fluorescence spectra of the individual PDI and Pe molecules in solution, named PDI-Ref and Pe-Ref hereafter, feature a pronounced vibronic structure (Figure 1) with splittings on the order of Δ𝜈 = 1400 cm-1 that correspond to the aromatic -C=C- stretching vibrations. In general, the electronic structure of PDI is only slightly perturbed after chemical modification at its imide-nitrogen(s).54-56

However, it is well known that strong changes in the electronic properties and molecular photophysics occur after core substitution of PDI.55-57 Linking Pe to the PDI core via a conjugated bridge introduces substantial through-bond coupling,58-60 and the mixing of electronic wavefunctions in the excited state imparts oscillator strength to a CT transition in the PDIPe dyad (Figure 1). The newly formed CT transition between 550 to 700 nm (18000-14000 cm-1) is absent in the PDI-Ref and Pe-Ref and exhibits vibronic band structure at low solvent polarity. The spacing is similar to the reference molecules and is


3


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

approximately Δ𝜈 = 1360 cm-1 (0.17 eV) in cyclohexane (CHX). In such a nonpolar solvent, the CT state has a pronounced emission (Figure 1) with a lifetime in the nanosecond regime that shows a bi-exponential intensity decay profile as anticipated (Figure S7 & Table S1). The CT emission is red-shifted with respect to the PDI-Ref and Pe-Ref fluorescence (Figure 1) and the excitation scan suggests that it is emanating from only one state (Figure S8). Based on the experimental redox potentials (Figure 1, Table S2), the lowest energy CT state has its electron density biased towards PDI, formally generating a CT state of.(– PDIPe ().or •– PDIPe•+ character. Although the description is qualitative, complete CT character for the emitting species can be excluded because of the small radiative rate constant of ionic species. This is in agreement with the observation at higher solvent polarity for which the exciplex-like emission is no longer detectable in steady-state fluorescence experiments due to the increase in CT character. We note that strictly speaking, we are not looking at a classical exciplex because of the through-bond coupling. Therefore, we prefer to use the term exciplex-like in the following description. The intensity decay of the fluorescence emission is not measurable at time delays above 200 ps after photoexcitation, which is the case for our timecorrelated single photon counting apparatus. This does not exclude the presence of an exciplex-like intermediate at earlier times. It is also evident from Figure 1 that the vibronic structure of the emission disappears at higher solvent polarity, e.g. in Et2O (higher CT character in the excited state) contrary to the PDI-like fluorescence at higher energy (lower CT character in the excited state).61 Briefly, the broad and red-shifted emission originates from a CT state, while the structured emission band on the blue side is better characterized as singlet fluorescence from the neutral excited state of PDIPe. Ultrafast Transient Absorption Spectroscopy (TAS). Formation of •–PDIPe•+ in a linear PDIPe analog has been shown to occur on ultrafast time scales.54 Accordingly, we use TAS to follow the spectral characteristics and photoinduced dynamics of PDIPe (for details about TAS see Section 2.3 of the Supporting Information). As a starting point, we identified the ion absorptions of Pe-Ref and PDI-Ref. The excited state lifetimes, t, of these reference chromophores are 3.7 and 4.7 ns in dichloromethane (DCM), respectively (Figure S7). It is well established that the presence of electron donors or acceptors can drastically shorten excited state lifetimes due to efficient deactivation via CT or CS.1 In this respect, we use photoinduced bimolecular deactivation by the electron donor N,N-

Page 4 of 10

dimethylaniline (DMA) or the electron acceptors dicyanoethylene (DCNE) and tetracyanoethylene (TCNE) to generate the reference anions and cations, respectively (Figure S9). The main absorption peaks observed in DCM solution are summarized in Table 1 and are consistent with literature reports.62-64 Table 1: Ion absorption of reference molecules after photoinduced bimolecular CT with DMA, TCNE or DCNE, measured in DCM. Also shown are the band maxima for the dyad (values in nm). •– Pe•+ Pe•– PDI•+ PDI•– PDIPe•+ 577 603 604 715 722/580

Deviations from reported spectral positions, for example for Pe•+ and Pe•–,65 are explained by the extended π-conjugation in the reference molecules and dyad in this work. This extended π-conjugation decreases the energy gap between the electronic states and leads to the observation of the bands at longer wavelengths. In general, the reference band signatures represent the spectral positions with only small influence from an electric field of the ion pair, 𝐸 IP, because of a) the better shielding by the solvent and b) the assumed sandwich structure of bimolecular ion pairs.66 Here, 𝐸 IP is thus perpendicular to the detected 𝜇’s. In a strongly conjugated system, as in the case for PDIPe, excitation into the red-edge (670 nm) leads to the direct formation of the CT species. In medium polar solvents, such as DCM, an instantaneous appearance of the photoinduced absorption is observed with clearly defined maxima around 580 and 722 nm. This is concomitant with broad ground state bleaching (GSB) below 540 nm (Figure 2). The positive bands are assigned to Pe•+ around 580 nm and PDI•– at 722 nm by comparison to the reference bimolecular charge transfer experiments (Figure S9) and spectroelectrochemistry data (Figure S10). The discrepancy with the one electron oxidized [PDIPe]•+ absorption spectrum arises from the strong conjugation of PDIPe. The oxidation potential of PDIPe points to a localization of the hole on Pe, because its value differs only by 40 meV from the one for Pe•+-Ref. However, the combined experiments and simulations suggest that the charge is actually delocalized, giving a convolution of Pe•+ and PDI•+ absorption bands in the spectroelectrochemistry measurements (Figure S10). Note, that explaining the two bands by aggregation, could not be confirmed in diluted experiments. Therefore, if the given interpretation is correct, identifying specific ion bands in TA difference spectra by comparison to spectroelectrochemistry data alone is not possible. Returning to the TA difference spectrum of PDIPe, the CT species relaxes to its equilibrium position in the first few picoseconds (< 2 ps) with the surrounding solvent adapting to the newly established dipole of the


4

Page 5 of 10

dyad.67 The resulting spectral shift and narrowing of the bands in this solvent are in good agreement with the spectral shift measured for PDI-Ref (Figure S11) and reflect ultrafast solvent relaxation. Consequently, the transient signals of PDIPe decay to zero with an isosbestic point at 545 nm, indicating that charge recombination is the only operative pathway and that hot ground state features are not detectable.68 This argument is also made for alternative pathways, such as triplet formation, for which in addition the spectral PDI features at 471, 505 and 550 nm are not observed.69 However, even after solvent relaxation (time delay > 2 ps), the tentatively assigned Pe•+ band shifts by 17 nm

(450 cm-1) from 580 to 563 nm while narrowing and decaying. The apparent dynamic shift of the Pe•+ band is accompanied by a decrease of the Pe•+/PDI•– ratio, best seen when normalizing the signal amplitude at the PDI•– maximum (Figure 2). The ground state bleach of PDIPe has a timedependent shape and the good agreement with the ground state absorption spectrum at early times ultimately vanishes to an approximately 7-9 nm redshift at later times. A simple representation of the kinetics at different wavelengths illustrates that the dynamics are well modeled by two time constants once solvent relaxation is complete (Figure 2, right).

λ

λ @ 5 ps

Δ

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


λ

λ

λ

Figure 2: Left: (Top) Ground state absorption spectrum of PDIPe in DCM (black) and transient difference spectra of the relevant anion (red) and cation (blue) species obtained from bimolecular charge transfer experiments. The Pe•+-Ref spectrum was obtained by subtracting the early and late time delay measurement from each other after normalization to the excited state absorption band. (Bottom) Transient absorption difference spectra of PDIPe in DCM at different time delays after excitation at 670 nm. The absorbance of PDIPe at the excitation wavelength was 0.1 in a 2 mm path length optical cell with a pulse energy of ~ 0.56 mJ/cm2. Center: (Top) Temporal evolution of the cation/anion ratio (orange) and spectral position of the cation maximum (pink). (Bottom) Transient difference spectra after normalization between 710-730 nm. Right: (Top) Signal amplitude at 5 ps for 580/722 nm as a function of excitation energy. (Bottom) Dynamics of PDIPe in DCM at depicted wavelengths. The shown fits were obtained from global analysis of the experimental data using the sum of two exponentials.

Origin of the shift. Due to direct population of the CT state and femtosecond relaxation of the solvent medium, charge recombination should be the only expected process in the •–PDIPe•+ CS state. No band shifts or amplitude ratio changes would be expected to occur on the picosecond time scale. In the present case, we observe a time-dependent shift for the cation band, i.e. Pe•+ only. As a consequence of the dyad structure, 𝐸 CT is roughly parallel to 𝜇 Pe•+ and perpendicular to 𝜇 PDI•–. Therefore it acts almost exclusively on 𝜇 Pe•+. As observed by the ratio change of Pe•+/PDI•– and the shift of the Pe•+ frequency, the dynamics can be related to an influence of 𝐸 CT as explained above. This influence is evidently time-dependent and could therefore represent charge localization in an exciplexlike intermediate. However, in the dyad, geometrical rearrangements, including rotation around the triple bond, can occur. This is analogous to the well-known

twisted intramolecular CT (TICT)70 as in the classical example of 9,9’-bianthryl, in which CT is discussed in view of a twist of the former in-plane D/A pair.71-73 For PDIPe, no twist is required to initiate CT and therefore we avoid this terminology. Nevertheless, this does not exclude a twist as the potential origin of the transient Stark shift. A twist can change the relative angle between 𝜇 Pe•+ and 𝐸 CT and therefore the Stark influence even at constant 𝐸 CT amplitude. A large change of the angular orientation of 𝐸 CT is not evident when assuming the extreme case of a complete rotation of Pe around the bridge axis. Small variations between 𝜇 and 𝐸 at parallel or perpendicular orientations have only minor impact due to the cosine relationship. Additionally, when rotating Pe around the triple bond it can be argued that the orientation of 𝜇 Pe•+ changes similarly to the one of 𝐸 CT and the relative angular change is even smaller. This


5


qualitative assumption will be further discussed in the light of quantum chemical simulations in the next section. For now, we state that an angular change between 𝜇 Pe•+ and constant 𝐸 CT is probably not completely accounting for the shift of the Pe•+ band in the present dyad. We hypothesize that a structural motion interplays with a process that changes the amplitude of 𝐸 CT because zero changes in nuclear geometry would result in an identical electronic structure and therefore constant 𝐸 CT amplitude. It has been recently shown that symmetry-breaking charge separation, i.e. the localization of charge density, is guided by geometrical relaxation of side chains in quadrupolar molecules.74 Such a geometrical relaxation could be another reason for the timedependent localization of charge density in PDIPe. Also in this case, it would be expected that the angle between 𝜇 Pe•+ and 𝐸 CT should not significantly change, rendering its influence on the Pe•+ band static. A related shift would then be a sole probe of the CT character of •–PDIPe•+. We turn now to quantum chemical calculations to elucidate deeper insight into the origin of this energy shift. Quantum Chemical Calculations. Optimizing the ground state geometry of PDIPe at the density functional level of theory (B3LYP functional and 6311G** basis set) gives two stable conformers 1 and 2, with the latter having a slightly lower energy of 57 meV, i.e. ~2kBT. The calculated permanent dipole moment is similar for both conformers ~2 Debye. A detailed description of all computational procedures can be found in sections 2.5 and 6 of the Supporting Information. The structural difference between conformer 1 and 2 is largely characterized by torsional motion of almost 180° about the triple bond (Figure 3). Steric hindrance in the dyad forces the PDI subunit to slightly twist its core (Figure S12), an effect already seen for various bay-substituted PDIs.75-76 We define the dyad by the deviation from planarity that is induced by the torsional angle. Conformer 1 then has a torsional angle of 350° and conformer 2 an angle of 160°. The small energy barrier between the conformers raises the question as to whether or not torsional motion around the triple bond is at the origin of the transient Stark shift. Comparing the orientation of the simulated permanent dipole moments, 𝜇 PDIPe, of the two conformers gives an approximate angular difference of 14° (Figure S13). Geometrically, 𝜇 PDIPe equals 𝐸 CT and we can now qualitatively compare it with 𝜇 Pe. In order to do so, we draw a line between the bridge atoms and the center of the Pe-tail to schematically illustrate 𝜇 Pe. We see that it rotates in phase with 𝐸 CT and their mutual angles change

approximately from 11° to 9°, confirming the earlier assumption that angular changes between 𝜇 Pe•+ and 𝐸 CT are not likely responsible for the band shift. Upon excitation at 670 nm, the lowest and only accessible electronic state is the CT state of •–PDIPe•+, that has a predicted energy of 1.55 eV at the timedependent density functional level of theory (B3LYP functional and 6-311G** basis set, Table S3) and agrees well with values from the redox potentials of the dyad in DCM (Table S2). Similarly, the next 3 excited states that are higher in energy are predicted to be *PDIPe (2), PDIPe* (3) and •+PDIPe•– (4), with excitation energies from the ground state that again coincide with the results from electrochemistry experiments (Tables S2/S3 & Figure S14). The asterisk ‘*’ denotes the local excitation on either the PDI or Pe moiety. Concentrating on •–PDIPe•+, the transferred charge density during the photoinduced process is calculated for two geometries (160° & 350°). At the ground state geometry (i.e. fixed torsional angle and no geometrical relaxation), the CT species directly after excitation is predicted to involve a transferred charge density that accounts for ~68 % (for more details see Section 2.5 of the Supporting Information). At fixed torsional angle, but optimized CT geometry, this value only changes slightly to ~66 % (Figure 3).

CT0.66

CT0.82

exc. 670 nm ∆ESim= 33 kcal/mol

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 6 of 10

CT0.98

CT0.66

Figure 3: Quantum chemical calculations of the PDIPe energy in the ground (green) and first CT (red) state as a function of the torsional angle of Pe around the triple bond. Calculated geometries are shown in the figure, together with the CT character at discussed geometries, excluding the bridge. See Supporting Information for further details.

In the CT state, conformer 1 (350°) is in an energetic minimum and torsional motion is not observed (Figure S15). However, this is not the case for conformer 2 in


6

Page 7 of 10

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 CT state at 160°. Here, a torsional rotation guides .(– PDIPe ().to lower energy. During this rotation, .(– PDIPe ().crosses local energetic minima at 100° and 280°. At this geometry, Pe is almost perpendicular to the PDI plane. Recalculating the transferred charge density now yields 82% and 98%, respectively. This substantial change in 𝐸 CT amplitude illustrates the change in charge localization from an exciplex-like intermediate,.(– PDIPe ()., towards an almost fully charge separated state, •–PDIPe•+. The increased electric field strength that is now acting on 𝜇 Pe•+ leads to a Stark shift of 450 cm-1 and the torsional motion is assigned as the origin of this shift in the visible region. .(– PDIPe ().is however only distinguishable from •– PDIPe•+ as a result of the specific structural design of the dyad. An estimate of the electric field strength yields a value between 25-31 MV/cm at a distance between the charges of approximately 6 Å (see Section 6.4 of the Supporting Information and Figure S16). With respect to the different distances between the charges in different dyads, this value is in the expected region for 𝐸 CT magnitudes.48 Lastly, it should be noted that in the present work the excitation of an equilibrium of the two conformers and subsequent twist around the triple bond could be thought of as alternative explanation for the mechanism, especially since the steady-state fluorescence spectrum features emission from a neutral PDI-like species as well as from a species with CT character (Figure 1). Instead, this is best explained by a distribution around an average CT value in the spectroscopic ensemble measurement. Furthermore, the conformers have similar CT character in the ground state and directly after excitation. An excited equilibrium and following interchange between conformer 1 and 2 is a) hindered by an energy barrier and b) would lead to an increase and subsequent decrease of CT character after excitation. In this case, the Pe•+ signature in the transient spectra (Figure 2) would shift back to its initial position, which is not observed.

Summary We have demonstrated a structural design principle for molecular dyads that enables detecting picosecond charge localization in an exciplex-like charge transfer (CT) species in the visible region of the electromagnetic spectrum. To achieve this goal, three structural prerequisites were incorporated into a molecular dyad. It comprises the electron donor perylene, Pe, and the electron acceptor perylenediimide, PDI, connected via an acetylene bridge yielding the superchromophore PDIPe. The prerequisites include i) the electronic coupling, i.e.

through-bond coupling, ii) the alignment of the electric field of the CT species, 𝐸 CT with the transition dipole moment, 𝜇, and iii) a constant angle between 𝐸 CT and 𝜇. The strong through-bond coupling allows direct excitation of the CT transition of •–PDIPe•+ to follow charge localization in this bichromophore. Localization becomes visualized because the transition dipole moment of Pe•+, 𝜇 Pe•+, is aligned along the axis of 𝐸 CT, which itself increases in amplitude during charge localization. The resulting Stark effect shifts the Pe•+ band by 17 nm (450 cm-1). The estimated electric field strength created in the CT is around 25-31 MV/cm, a range that is also of relevance in biology.34,77 These classes of donoracceptor dyads could thus be an ultrafast way for perturbing transmembrane potentials or protein electrostatics that control electric signaling of biological cells. Quantum chemical calculations confirm the increase of CT character by shifting electron density from Pe towards PDI. The origin is identified as a torsional motion around the bridging triple bond orienting Pe almost perpendicular to the PDI plane. Furthermore, and of significant fundamental interest, our investigation suggests why exciplex formation in common molecular donor/acceptor pairs is difficult to observe when measured by transient electronic absorption spectroscopy. The sometimes observed band broadening of ion pair bands in the presence of exciplexes or excimers may be related to an angular deviation between 𝐸 and 𝜇. The development of new molecular design concepts could lead to structural information about CT complexes using electronic spectroscopy, which is usually discarded. The rigidity, strong absorption in the visible region and possibility for instantaneous population of the CT state in this class of superchromophores makes the present dyad suitable for investigations of ultrafast, possibly coherent, processes with two-dimensional electronic spectroscopy. However, our study has indicated that the complexity of this molecular class is increased tremendously, as it intrinsically involves exciplex-like intermediates concomitant with strong through-bond coupling.

Associated Content Experimental section and procedures, synthesis, timeresolved fluorescence and absorption spectra, computational results.

Author Information


7


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

Corresponding Authors. [email protected], [email protected] Notes. The authors declare no competing financial interest.

Acknowledgements G.D.S. and F.N.C. gratefully acknowledge the Division of Chemical Sciences, Geosciences, and

References 1. Leonhardt, H.; Weller, A., Z. Phys. Chem. N. F. 1961, 29, 277-280. 2. Gould, I. R.; Farid, S., Acc. Chem. Res. 1996, 29, 522-528. 3. Gust, D.; Moore, T. A.; Moore, A. L., Acc. Chem. Res. 2009, 42, 1890-1898. 4. Karlsson, S.; Boixel, J.; Pellegrin, Y.; Blart, E.; Becker, H.-C.; Odobel, F.; Hammarstrom, L., Faraday Discuss. 2012, 155, 233-252. 5. Wasielewski, M. R., Chem. Rev. 1992, 92, 435-461. 6. Kumpulainen, T.; Lang, B.; Rosspeintner, A.; Vauthey, E., Chem. Rev. 2016, DOI:10.1021/acs.chemrev.6b00491. 7. Reid, O. G.; Pensack, R. D.; Song, Y.; Scholes, G. D.; Rumbles, G., Chem. Mater. 2014, 26, 561-575. 8. Swinnen, A. M.; Van der Auweraer, M.; De Schryver, F. C.; Nakatani, K.; Okada, T.; Mataga, N., J. Am. Chem. Soc. 1987, 109, 321-330. 9. Wasielewski, M. R.; Minsek, D. W.; Niemczyk, M. P.; Svec, W. A.; Yang, N. C. C., J. Am. Chem. Soc. 1990, 112, 2823-2824. 10. Banerji, N.; Angulo, G.; Barabanov, I.; Vauthey, E., J. Phys. Chem. A 2008, 112, 9665-9674. 11. Al Subi, A. H. A.; Niemi, M. M.; Tkachenko, N. V. N.; Lemmetyinen, H. H., J. Phys. Chem. A 2012, 116, 96539661. 12. Knibbe, H.; Rehm, D.; Weller, A., Z. Phys. Chem. 1967, 56, 95-98. 13. Gould, I. R.; Young, R. H.; Mueller, L. J.; Farid, S., J. Am. Chem. Soc. 1994, 116, 8176-8187. 14. Iwai, S.; Murata, S.; Katoh, R.; Tachiya, M.; Kikuchi, K.; Takahashi, Y., J. Chem. Phys. 2000, 112, 7111-7117. 15. Kuzmin, M. G. M.; Soboleva, I. V. I.; Dolotova, E. V. E.; Dogadkin, D. N. D., Photochem. Photobiol. Sci. 2003, 2, 967-974. 16. Morandeira, A.; Fürstenberg, A.; Vauthey, E., J. Phys. Chem. A 2004, 108, 8190-8200. 17. Koch, M.; Letrun, R.; Vauthey, E., J. Am. Chem. Soc. 2014, 136, 4066-4074. 18. Koch, M.; Licari, G.; Vauthey, E., J. Phys. Chem. B 2015, 119, 11846-11857. 19. Morteani, A. C. A.; Sreearunothai, P. P.; Herz, L. M. L.; Friend, R. H. R.; Silva, C. C., Phys. Rev. Lett. 2004, 92, 247402. 20. Offermans, T.; van Hal, P. A.; Meskers, S. S.; Koetse, M. S.; Janssen, R. R., Phys. Rev. B 2005, 72, 045213.

Page 8 of 10

Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant No. DESC0015429 and DE-SC0011979, respectively. M.K. is grateful for support from a Swiss National Science Foundation postdoctoral fellowship and thanks Prof. Michael R. Wasielewski at Northwestern University for helpful discussions. D.G.O. acknowledges financial support from the Natural Sciences and Engineering Council of Canada. S.G. is a Chemistry Graduate Scholars Research Assistant at NC State University.

21. Yin, C. C.; Kietzke, T. T.; Neher, D. D.; Hoerhold, H. H.-H., Appl. Phys. Lett. 2007, 90, 092117. 22. Benson Smith, J. J.; Wilson, J. J.; Dyer Smith, C. C.; Mouri, K. K.; Yamaguchi, S. S.; Murata, H. H.; Nelson, J. J., J. Phys. Chem. B 2009, 113, 7794-7799. 23. Dyer Smith, C. C.; Benson Smith, J. J.; Murata, H. H.; Mitchell, W. J. W., J. Phys. Chem. C 2009, 113, 1453314539. 24. Shepherd, W. E. B.; Platt, A. D. A.; Kendrick, M. J. M.; Loth, M. A. M.; Anthony, J. E. J.; Ostroverkhova, O. O., J. Phys. Chem. Lett. 2011, 2, 362-366. 25. Stewart, D. J. D.; Dalton, M. J. M.; Swiger, R. N. R.; Cooper, T. M. T.; Haley, J. E. J.; Tan, L. L.-S., J. Phys. Chem. A 2013, 117, 3909-3917. 26. Chao, C. I.; Chen, S. A., Appl. Phys. Lett. 1998, 73, 426428. 27. Feng, J.; Li, F.; Gao, W. B.; Liu, S. Y.; Liu, Y., Appl. Phys. Lett. 2001, 78, 3947-3949. 28. D’Andrade, B. W.; Brooks, J.; Adamovich, V.; Thompson, M. E.; Forrest, S. R., Adv. Mater. 2002, 14, 1032-1036. 29. Morteani, A. C.; Friend, R. H.; Silva, C., Chem. Phys. Lett. 2004, 391, 81-84. 30. Kim, J.-S.; Lu, L.; Sreearunothai, P.; Seeley, A.; Yim, K.-H.; Petrozza, A.; Murphy, C. E.; Beljonne, D.; Cornil, J. r. m.; Friend, R. H., J. Am. Chem. Soc. 2008, 130, 1312013131. 31. Dereka, B.; Koch, M.; Vauthey, E., Acc. Chem. Res. 2017, 50, 426-434. 32. Fried, S. D.; Boxer, S. G., Acc. Chem. Res. 2015, 48, 998-1006. 33. Stark, J., Nature 1913, 92, 401-401. 34. Fried, S. D.; Bagchi, S.; Boxer, S. G., Science 2014, 346, 1510-1514. 35. Boxer, S. G., Stark Realities. J. Phys. Chem. B 2009, 113, 2972-2983. 36. van Wilderen, L. J. G. W.; Kern-Michler, D.; MullerWerkmeister, H. M.; Bredenbeck, J., Phys. Chem. Chem. Phys. 2014, 16, 19643-19653. 37. Slocum, J. D.; Webb, L. J., J. Am. Chem. Soc. 2016, 138, 6561-6570. 38. Hu, W.; Webb, L. J., J. Phys. Chem. Lett. 2011, 2, 19251930. 39. Mohrmann, H.; Kube, I.; Lórenz-Fonfría, V. A.; Engelhard, M.; Heberle, J., J. Phys. Chem. B 2016, 120, 4383-4387.


8

Page 9 of 10

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


40. Baiz, C. R.; Kubarych, K. J., J. Am. Chem. Soc. 2010, 132, 12784-12785. 41. Ardo, S.; Sun, Y.; Castellano, F. N.; Meyer, G. J., J. Phys. Chem. B 2010, 114, 14596-14604. 42. Ardo, S.; Sun, Y.; Staniszewski, A.; Castellano, F. N.; Meyer, G. J., J. Am. Chem. Soc. 2010, 132, 6696-6709. 43. Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E., Adv. Mater. 2007, 19, 1551-1566. 44. Caruso, D.; Troisi, A., Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 13498-13502. 45. Gélinas, S.; Rao, A.; Kumar, A.; Smith, S. L.; Chin, A. W.; Clark, J.; van der Poll, T. S.; Bazan, G. C.; Friend, R. H., Science 2014, 343, 512-516. 46. Cabanillas-Gonzalez, J.; Virgili, T.; Gambetta, A.; Lanzani, G.; Anthopoulos, T. D.; de Leeuw, D. M., Phys. Rev. Lett. 2006, 96, 106601. 47. Cabanillas-Gonzalez, J.; Virgili, T.; Gambetta, A.; Lüer, L.; Lanzani, G.; Anthopoulos, T. D.; de Leeuw, D. M., Phys. Rev. B 2007, 75, 045207. 48. Debreczeny, M. P.; Svec, W. A.; Wasielewski, M. R., New. J. Chem. 1996, 20, 815-828. 49. Debreczeny, M. P.; Svec, W. A.; Wasielewski, M. R., Science 1996, 274, 584-587. 50. Debreczeny, M. P.; Svec, W. A.; Marsh, E. M.; Wasielewski, M. R., J. Am. Chem. Soc. 1996, 118, 81748175. 51. Wu, Y.; Young, R. M.; Frasconi, M.; Schneebeli, S. T.; Spenst, P.; Gardner, D. M.; Brown, K. E.; Würthner, F.; Stoddart, J. F.; Wasielewski, M. R., J. Am. Chem. Soc. 2015, 137, 13236-13239. 52. Angulo, G.; Kattnig, D. R.; Rosspeintner, A.; Grampp, G.; Vauthey, E., Chem.-Eur. J. 2010, 16, 2291-2299. 53. Sung, J.; Nowak-Król, A.; Schlosser, F.; Fimmel, B.; Kim, W.; Kim, D.; Würthner, F., J. Am. Chem. Soc. 2016, 138, 9029-9032. 54. Brown, K. E.; Veldkamp, B. S.; Co, D. T.; Wasielewski, M. R., J. Phys. Chem. Lett. 2012, 3, 2362-2366. 55. Weiss, E. A.; Ahrens, M. J.; Sinks, L. E.; Gusev, A. V.; Ratner, M. A.; Wasielewski, M. R., J. Am. Chem. Soc. 2004, 126, 5577-5584. 56. Huang, C.; Barlow, S.; Marder, S. R., J. Org. Chem. 2011, 76, 2386-2407. 57. Yan, Q.; Zhao, D., Org. Lett. 2009, 11, 3426-3429. 58. Karki, L.; Vance, F. W.; Hupp, J. T.; LeCours, S. M.; Therien, M. J., J. Am. Chem. Soc. 1998, 120, 2606-2611.

59. LeCours, S. M.; Guan, H.-W.; DiMagno, S. G.; Wang, C. H.; Therien, M. J., J. Am. Chem. Soc. 1996, 118, 14971503. 60. Amanpour, J.; Hu, G.; Alexy, E. J.; Mandal, A. K.; Kang, H. S.; Yuen, J. M.; Diers, J. R.; Bocian, D. F.; Lindsey, J. S.; Holten, D., J. Phys. Chem. A 2016, 120, 7434-7450. 61. Mohammed, O. F.; Vauthey, E., J. Phys. Chem. A 2008, 112, 5804-5809. 62. Shida, T., Electronic Absorption Spectra of Radical Ions. Elsevier: Amsterdam, 1988; Vol. Physical Sciences data 34. 63. Kircher, T.; Lohmannsroben, H. G., Phys. Chem. Chem. Phys. 1999, 1, 3987-3992. 64. Gosztola, D.; Niemczyk, M. P.; Svec, W.; Lukas, A. S.; Wasielewski, M. R., J. Phys. Chem. A 2000, 104, 65456551. 65. Markovic, V.; Villamaina, D.; Barabanov, I.; Daku, L. M.; Vauthey, E., Angew. Chem. Int. Ed. 2011, 50, 75967598. 66. Mohammed, O. F.; Adamczyk, K.; Banerji, N.; Dreyer, J.; Lang, B.; Nibbering, E. T. J.; Vauthey, E., Angew. Chem. Int. Ed. 2008, 47, 9044-9048. 67. Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M., J. Phys. Chem. 1995, 99, 17311-17337. 68. Koch, M.; Rosspeintner, A.; Adamczyk, K.; Lang, B.; Dreyer, J.; Nibbering, E. T. J.; Vauthey, E., J. Am. Chem. Soc. 2013, 135, 9843-9848. 69. Ford, W. E.; Kamat, P. V., J. Phys. Chem. 1987, 91, 6373-6380. 70. Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W., Chem. Rev. 2003, 103, 3899-4032. 71. Kang, T. J.; Kahlow, M. A.; Giser, D.; Swallen, S.; Nagarajan, V.; Jarzeba, W.; Barbara, P. F., J. Phys. Chem. 1988, 92, 6800-6807. 72. Kang, T. J.; Jarzeba, W.; Barbara, P. F.; Fonseca, T., Chem. Phys. 1990, 149, 81-95. 73. Yamasaki, K.; Arita, K.; Kajimoto, O.; Hara, K., Chem. Phys. Lett. 1986, 123 (4), 277-281. 74. Kim, W.; Sung, J.; Grzybowski, M.; Gryko, D. T.; Kim, D., J. Phys. Chem. Lett. 2016, 7, 3060-3066. 75. Nagarajan, K.; Mallia, A. R.; Reddy, V. S.; Hariharan, M., J. Phys. Chem. C 2016, 120, 8443-8450. 76. Myahkostupov, M.; Prusakova, V.; Oblinsky, D.; Scholes, G.; Castellano, F., J. Org. Chem. 2013, 78, 86348644. 77. Hekstra, D. R.; White, K. I.; Socolich, M. A.; Henning, R. W.; Šrajer, V.; Ranganathan, R., Nature 2016, 540, 400405.


9


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

Page 10 of 10

TOC Ground State δ-

Incomplete CT δ-

δ+

δ+

Complete CT -

+

Electric Field

time / ps early late

Stark shift

δδ+

-

+

λ


10

Charge Localization after Ultrafast Photoexcitation of a Rigid Perylene

Recommend Documents