Conformational Flexibility Determines Electronic Coupling and Charge

Subscriber access provided by Kaohsiung Medical University

C: Energy Conversion and Storage; Energy and Charge Transport

Conformational flexibility determines electronic coupling and charge transfer character in single propeller-shaped perylene diimide tetramer arrays David J. Walwark, Benjamin Douglas Datko, Qinghe Wu, Andriy Neshchadin, Margaret Berrens, Luping Yu, and John K. Grey J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b05709 • Publication Date (Web): 25 Sep 2018 Downloaded from http://pubs.acs.org on September 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 35 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

Conformational Flexibility Determines Electronic Coupling and Charge Transfer Character in Single Propeller-Shaped Perylene Diimide Tetramer Arrays David J. Walwark Jr, Benjamin D. Datko, Qinghe Wu, Andriy Neshchadin, Margaret L. a

a

b, †

b

Berrens, Luping Yu, * John. K. Grey * a

a

b

a

Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, NM 87131. Department of Chemistry and The James Franck Institute, University of Chicago, b

Chicago, IL 60637.

ABSTRACT

We investigate the effect of molecular geometry and conformational flexibility on electronic coupling and charge transfer interactions within propeller-shaped perylene diimide (PDI) tetramer arrays differing by the number of covalent linkages to a central spirobifluorene core. Electronic spectra of tetramers with one (‘floppy’) or two (‘rigid’) bay covalent linkages display evidence of charge transfer character in either ground or excited states. Floppy tetramers exhibit marked redshifted and broadened absorption features that we assign as overlapping inter-PDI charge transfer and PDI-centered p-p* transitions whereas rigid tetramers retain features similar to single PDI molecules albeit broader linewidths. Interestingly, both tetramers exhibit charge transfer character

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 35

in their fluorescence emission but this is most prominent in the rigid tetramer which displays dominant long-lived excimer behavior in addition to a minority component resembling single PDIlike emission. We then use single molecule spectroscopy and imaging to understand how conformational-dependent charge transfer properties influence tetramer photophysics. Over 90% of single rigid tetramers display telegraphic (i.e., 2-level) blinking behavior with relatively short “on” times compared to ~60% of single floppy tetramer transients which tends to exhibit emission from multiple levels.

Electronic structure simulations were next performed to aid in the

assignment of electronic transitions and photophysical behavior. Floppy tetramer canonical and natural transition orbitals reveal remarkable similarities with significant charge transfer character in the lowest energy excited states involving transverse PDI units and appreciable spirobifluorene contributions in the ground electronic state.

Rigid tetramers exhibit greater electronic

delocalization and calculated absorption transition energies show good agreement with experiment although excited state interactions are less straightforward to discern from simulations. Raman spectroscopy and polarization dependent single molecule spectroscopy were also performed that support assignments based on theoretical predictions and electronic spectroscopy results. Overall, we demonstrate the importance of molecular geometry and conformational flexibility of multichromophore arrays in determining the nature of electronic interactions in ground and excited states, which can eventually be harnessed to improve performance attributes at the materials level.


2

Page 3 of 35 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


Introduction Good electronic communication between conjugated organic molecules is essential for achieving efficient charge transport and minimal losses in solid-state optoelectronic device applications. However, solution spin casting processing methods typically used to fabricate device thin film active layers do not allow for accurate control of molecular connectivity and order over large length scales. The inability to reliably selectively control molecular conformation and packing 1

which dictate electronic coupling strength and directionality is perhaps best exemplified in solar cells based on blends of conjugated polymers as electron donors with soluble fullerenes as electron acceptors. In particular, material performance suffers from the severe tortuosity of charge migration pathways due to morphological heterogeneity over a wide range of length scales leading to unwanted charge trapping and recombination. In order to overcome such losses, contiguous 2

percolation pathways are required to achieve efficient charge transport and extraction in solar cells, although these efforts have relied chiefly on trial and error optimization schemes. 3-6

7

Molecular structure and geometric factors have, surprisingly, received far less consideration as input parameters for achieving good electronic communication at the materials level. The slower evolution of molecular engineering can perhaps by traced to the effectiveness of fullerenes as electron acceptors in organic solar cells, which is evident from the intense focus on varying conjugated polymer donor structure and electronic properties. Band gap tuning by introducing 8

charge transfer character into the backbone (i.e., push-pull systems), control of packing 9

interactions using side group engineering,

10-11

polydispersity effects

12-13

and small molecule variants that avoid issues with

are among the most notable focus areas on this front. While substantial

improvements in performance, e.g., power conversion efficiencies surpassing 10%, have been achieved largely by morphological tuning, there is still room for improvement. In particular, much


3


Page 4 of 35

can be gained by obtaining molecular-level perspectives of structural factors regulating electronic interactions in conjugated organic materials for achieving reliable synthetic (bottom-up) control of material performance. Along these lines, there has been greater interest in replacing fullerenes with more cost-effective alternatives without sacrificing stability or power conversion efficiencies. Among the most promising systems to date are perylene diimide (PDI) arrays that, in addition to their high electron affinities, may be synthetically configured in various geometries (e.g., linear or higher dimensional arrays).

14-20

Multi-PDI arrays offer the prospect of tunable packing and electronic coupling over

larger distance scales by design

21-22

and their relatively large absorptivities present greater

opportunities for energy harvesting not often available in fullerene-based acceptors. Similar to fullerenes, long-range electronic communication through good intermolecular connectivity is essential for PDI arrays to realize their full potential in solid state devices. This characteristic is ultimately dictated by molecular level structure and conformational qualities yet often difficult to discern using ensemble level materials and device characterization techniques. To this end, we synthesized two structurally similar PDI tetramers, differing only by the number of covalent linkages to a spirobifluorene core in the PDI bay position, to determine how molecular geometry and conformational flexibility impact intramolecular electronic communication and photophysics. We used electronic absorption and emission spectroscopy to investigate ground and excited electronic states of tetramers with one and two covalent links classified herein as “floppy” and “rigid”, respectively. These molecules exhibit varying degrees of charge transfer character in either ground and excited electronic states depending on their conformational characteristics (i.e., flexibility). We next investigated photophysics at the single molecule level where tetramers displayed distinct intermittency behaviors that reflect the nature of emitting state(s) and


4

Page 5 of 35 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


interactions within tetramers and their surroundings. Lastly, we use density functional theoretical simulations, Raman spectroscopy and polarization-dependent single molecule spectroscopy to confirm assignments of conformation-dependent electronic transitions. Our results provide useful molecular-level perspectives of the critical role of molecular geometry and conformational flexibility on electronic coupling within these light harvesting arrays, which have already demonstrated promise in solar cells.

23

Experimental

Chart 1. Synthesis of the rigid tetramer using floppy precursors.

The floppy tetramer was first synthesized using a similar procedure described in an earlier report. The rigid derivative was prepared by dissolving FeCl (2 g) in 6 mL CH NO was added 23

3

3

2

to 15 mL CH Cl solution of the floppy tetramer (212 mg) at 0 °C, which is outlined in Chart 1. 2

2

After one hour stirring at room temperature, 20 mL 1 M HCl was added. After stirring overnight, the organic part was separated and the solvent was removed under reduced pressure. The product


5


Page 6 of 35

was purified by column chromatography, using dichloromethane/hexane = 3:2 as the eluent. 108 mg of pure rigid tetramer (51%) was obtained. The rigid compound was characterized by H NMR 1

spectroscopy in CDCl solution. However, peak assignments were complicated although features 3

were similar to other PDI arrays. Mass spectrometry (MALDI-TOF) was also performed on both tetramers confirming the structures outlined in the Supporting Information. C H N O m/z: 225

248

8

16

3320.50; Found: (M)+; Anal. calcd for C H N O : C, 81.39 %; H, 7.53 %; N, 3.37 %. 225

248

8

16

Steady-state electronic spectroscopy was performed on dilute solutions of both tetramers dissolved in chlorobenzene. Electronic absorption and fluorescence emission and excitation spectra were recorded using a scanning spectrometer with broadband and monochromatic excitation sources (Edinburgh instruments). Synchronously scanned fluorescence excitation and emission spectroscopy were also measured, wherein excitation wavelength and detected emission wavelength are simultaneously changed while being held offset at a fixed interval. Fluorescence lifetimes were measured using time-correlated single photon counting techniques (TCSPC) and all decays were fitted using an iterative convolution approach. Raman spectra were collected using a microscope-based imaging spectrometer (Nicolet) and backgrounds were removed by fitting a spline function. All spectra were corrected for instrument response. Dilute solid dispersions of both tetramers were prepared by dispersing dilute chlorobenzene solutions into polystyrene (3% w/w, chlorobenzene) then spin casting onto rigorously cleaned glass coverslips. Single molecule images were acquired using a widefield microscope system equipped with an electron multiplying CCD camera (Andor iXon). Samples were excited using 488 nm light and excitation power densities were typically in the range of ~1 W/cm —100 W/cm 2

2

and scattered excitation light was removed using long-pass filters. Areal densities scaled linearly with concentration and excitation intensities were varied to optimize photostability and signal-to-


6

Page 7 of 35

noise ratios. No evidence of aggregation was apparent, and all samples showed diffraction-limited spots with discrete intermittency behavior. In order to facilitate particle tracking over the duration of the acquisition (~300 s), the imageJ MATLAB plug-in was used. Single molecule intensity transients were generated by calculating the particle centroid followed by integrating intensities on a per-frame basis for each resolved particle. Background thresholds were set to three standard deviations of the noise floor and comparisons of several samples were used to parameterize a stepfinding algorithm to de-noise (idealize) the fluorescence transients. An adaptive kernel density estimation (AKDE) was then performed to determine the amount of emission levels in each transient.

Polarization-dependent single molecule spectroscopic imaging was performed by

rotating linearly polarized excitation light from 0 to 180 degrees synchronously with image acquisition.

Polarization modulation depths were measured by first removing instrumental

distortions determined by comparing systems with known excitation polarization dependence.

Results and Discussion

Wavelength (nm) 600 500

700

450 PDI

Norm. Int.

Rigid

Norm. Abs.

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


Floppy

14

16 18 20 22 3 -1 Energy (cm ) x10

24


7


Page 8 of 35

Figure 1. Electronic absorption and fluorescence spectra of “rigid” and “floppy” spirobifluorene PDI tetramers, respectively. An absorption spectrum of a single PDI derivative (dotted trace) is included alongside the rigid absorption spectrum for comparison.

Figure 1 displays electronic absorption and fluorescence spectra of dilute tetramer solutions (2000 cm ) consistent with appreciable geometric rearrangements in the excited state or multiple -1

overlapping transitions.

31-32

On the other hand, the floppy tetramer displays a broad and featureless

fluorescence line shape at ca. 16000 cm and Stokes shifts of ~1500 cm consistent with more -1

-1

displaced vibrational modes, i.e., lower frequency modes. These features are also reminiscent of charge transfer type emission whereby resolved vibronic structure is usually broadened and transition energies are significantly red-shifted. We rule out self-absorption as a cause for the appearance of large Stokes shifts based on concentration dependent studies (data not shown). To get a clearer picture of the emitting state(s) in both molecules we measured time-resolved fluorescence decays at different emission energies and fit results are tabulated in Tables S1 and S2 of the Supporting Information. Briefly, the rigid tetramer exhibits bi-exponential decay behavior across the fluorescence lineshape and the decay time constant of the minority emitting component (~14%) was ~4 ns in contrast to the dominant emitting component (~85%) with an average decay time constant of 33 ns (Table S2). The floppy tetramer, on the other hand, displayed a single emitting component (>95%) with a time constant of ~8 ns. The minority emitting component in the rigid tetramer is in good agreement with lifetimes reported from single PDI molecules (~4 ns),

33

in contrast to the dominant long-lived component, which corresponds to excimer emission. The 34

origins of excited state interactions producing excimeric features are not immediately clear although these results shed light on the unusual fluorescence lineshape of the rigid tetramer. We


9


Page 10 of 35

posit that the more intense, partially resolved component most likely corresponds to transitions centered on single PDI(s) and the broad, overlapping component is consistent with excimer-like transitions, which also explains the large observed Stokes shifts. The single exponential decays of the floppy tetramer are outside the range of typical excimer lifetimes in PDI systems (e.g., >10 ns), but, the broadened, red-shifted absorption and emission spectra are indicative of charge 34

transfer character. Comparison to related systems, e.g., spirobifluorene substituted PDI molecules, support this view where transitions involving substantial PDI-spiro admixture were observed.

28

We next performed single molecule spectroscopic imaging on both PDI tetramers dispersed in polystyrene hosts to help overcome potential ambiguities in assigning transitions from ensemble spectra (i.e., congested absorption and emission transitions).

These experiments also have a

distinct advantage for resolving how specific tetramer structural factors impact electronic communication within the tetramer molecules, which further aid in understanding photophysics and offer useful views of how tetramers respond to fluctuations in the local environment otherwise masked in ensemble measurements.

35-36

Fluorescence emission was excited using 488 nm light with

intensities in the range of ~1¾100 W/cm and Figures 2a,b show representative fluorescence 2

images of isolated rigid and floppy tetramers, respectively. In general, individual rigid tetramers were significantly brighter than their floppy counterparts measured under the same conditions but were less photostable (vide infra).


10

Page 11 of 35

Int. (cts) a) 1370

Rigid

b)

Floppy

Int. (cts) 1130 1090

1220

1050

Int. (arb. units)

1295

600 c)

Rigid

d)

Floppy

400 200 0 0

50

100 150 200 time (s)

50

100 150 200 time (s)

3.0 e) 2.5

PDF

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


Rigid Floppy

2.0 1.5 1.0 0.5 0.0 0.2

0.4 0.6 ΔI (Norm.)

0.8

1.0

Figure 2. Fluorescence images of single rigid (a) and floppy (b) tetramers (scale bar = 3 µm). Representative intensity transients for rigid (c) and floppy (d) molecules with resolved emitting levels shown as solid gray lines and background levels as solid green lines. e) Kernel density estimation of the probability distribution function (PDF) for the largest change in fluorescence intensity (normalized) for rigid and floppy single tetramer transients. Histogram areas are normalized to unity.

We examined this behavior in detail by measuring fluorescence intensity transients of many single rigid and floppy tetramers and representative examples are shown in Figures 2c,d, respectively, measured at lower excitation intensities. It is first informative to consider the nature of tetramer


11


Page 12 of 35

structures shown in Chart 1 as a basis for expected photophysical responses that may be reconciled later with trends in electronic spectra in Fig. 1. For example, if interactions amongst PDI units and PDI-spiro interactions are weak, then all four PDIs are expected to behave as independent chromophores whereas strong coupling usually manifests as single-step (telegraphic) blinking.

25, 37

Molecules with substantial charge transfer character tend to exhibit a broader range of responses that may resemble either of these limits. Over 2000 transients of each tetramer were collected and we employed a step-fitting idealization/denoising algorithm and adaptive kernel density estimation (AKDE) to resolve the range of discrete intensity fluctuations. This method effectively groups similar emission intensities (i.e., levels) over the total transient acquisition time (~200 s) and examples of the AKDE analysis results are depicted with representative experimental transients in Figs. 2c,d appearing as red dotted lines. We next constructed probability distribution histograms of the largest intensity changes (DI) from the AKDE analysis for each tetramer shown in Fig. 2e in which the area is normalized (additional examples of fluorescence blinking lability analyses are provided in Figs. S1, S2 of the Supporting Information). This metric reports the tendency of either molecule to switch to an “off” state during the transient acquisition in addition to transitions between emitting levels. Floppy tetramers exhibit a broad range of intermittency behaviors with a sizeable fraction of molecules only fluctuating between different “on” levels and never switching to an “off” state (i.e., intensities drop to background levels). As seen in Fig. 2e, floppy tetramers have nearly equal likelihood to undergo transitions between different emitting levels and visiting an “off” state. Conversely, the probability of experiencing a blinking is skewed heavily toward larger probabilities in rigid tetramers due to a higher frequency of only telegraphic blinking behavior leading to faster photobleaching.

Several possible explanations exist for the appearance of


12

Page 13 of 35 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


blinking and emitting level switching in these tetramers, such as reversible interactions with oxygen and charges, conformational dynamics, energy transfer, singlet-triplet interactions, and fluctuations in the local environment.

24, 31, 35, 38

Tentatively, the greater propensity for rigid tetramers

to undergo a blinking event typically corresponding to long “off” periods can be attributed to a higher probability of forming longer-lived triplet excited states. Likewise, the lower blinking frequency in floppy tetramers suggest that triplet levels are much lower although we presently are not able to quantify triplet population dynamics. We next construct statistical distributions of “on” and “off” times for molecules displaying distinct blinking events to gain a broader perspective of transition probabilities between levels. First, the fraction of tetramers experiencing blinking behavior obtained from the AKDE analysis for molecules that visit an “off” state during the acquisition varies significantly between floppy and rigid systems, namely, >90% of rigid tetramers blink compared to ~60% at excitation intensities ~100 W/cm . This disparity becomes much less pronounced at lower excitation 2

intensities where the floppy tetramer fluorescence is much longer-lived and tends to only fluctuate between different “on” levels (see Fig. S3, Supporting Information). Complementary cumulative probability distribution functions (CCDF) are displayed in Figs. 3a,b for “on” and “off” time distributions, respectively. These behaviors reflect the statistical likelihood of remaining in either “on” or “off” states which are represented here in terms of characteristic times (i.e., t and t ). We on

off

do not distinguish between different “on” levels when constructing these plots, which effectively only reports the tendency for tetramers to blink via triplet-induced oxygen sensitization.


13


Page 14 of 35

Figure 3. Complementary cumulative distribution functions (CCDF) of “on” (a) and “off” (b) times for floppy and rigid molecule sub-ensembles displaying blinking behavior.

Qualitatively, both systems show deviations from idealized power law behavior as observed previously for single PDI chromophore emitters,

36, 39-40

which is most evident from the “on” time

distributions of both tetramers. A systematic error emerges at longer times from the finite transient acquisition times of the experiment where curves tend toward the time axis intercept. Deviations from power law behavior (i.e., single emitter behavior) are especially pronounced for the rigid tetramer that probably reflect the presence of isolated PDI-like and excimer emitters. Comparisons of “off’ time distributions reveal greater similarities and indicate a similar mechanism, i.e., triplet mediated photochemistry, although the large differences between “on” time distributions in Fig. 3a confirm the rigid tetramer is much less stable.


14

Page 15 of 35

We further rule out ground state interactions as a cause for these behaviors due to the fact that CCDF behaviors are excitation intensity dependent although similar trends in “on” time behaviors exist at lower excitation intensities (ca. 10 W/cm ) for the rigid tetramer (see Fig. S3, Supporting 2

Information). The greater frequency of blinking in these molecules further suggest that excimer interactions are promoting triplet formation leading to larger yields of oxygen sensitization. For example, these longer-lived states increase the likelihood of transitioning into the triplet manifold,

41

which is also enhanced when PDI units aggregate.

42

Additionally, recent investigations of PDI

solids reported efficient singlet fission. However, it is premature at this stage to assign the larger 43

blinking frequency in the rigid tetramer to singlet fission although the fluorescence intermittency statistics in Fig. 3 are consistent with larger triplet occupancies on long time scales.

a) "Floppy"

b) "Rigid"

LUMO

LUMO

HOMO

HOMO

c)

Energy (eV)

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


-3 -4 -5 -6

Floppy Rigid


15


Page 16 of 35

Figure 4. HOMO and LUMO levels for floppy (a) and rigid (b) tetramers. c) HOMO(-4) and LUMO(+4) levels. In order to gain a deeper perspective of structural influences on electronic properties and photophysics, we perform theoretical simulations to resolve the dependence of the ground and lowest energy excited electronic states on tetramer conformational qualities. Density functional theory (DFT) was used to calculate electronic structures of both systems in optimized ground state molecular geometries at the B3LYP/6-31G level. Figures 4a,b show the canonical HOMO and LUMO isosurfaces for floppy and rigid tetramers, respectively. Frontier orbitals of the floppy tetramer are mostly localized on individual PDI units and transitions but, HOMO and LUMO electron densities reside on different PDI units, specifically, transverse positions, indicating ground state charge transfer character.

There are also significant contributions from the

spirobifluorene core in the floppy HOMO as opposed to the LUMO which is concentrated entirely on the transverse PDI unit. In contrast, the rigid HOMO level is delocalized over the entire molecule whereas its LUMO localizes to one PDI-spiro-PDI ‘wing’. Electronic delocalization effects are well known for mainly linear PDI arrays and assemblies although its extent in the tetramer systems is limited by the specific conformation and environment.

22, 44

It is important to

note that the omission of the branched alkyl groups removes possible steric contributions that can lead to significant deviations from this idealized case.

The four highest/lowest

occupied/unoccupied orbital energies (up to HOMO-4, LUMO+4) are also shown in Fig. 4c for comparison. Energy level dispersion is very small in the rigid tetramer (

Conformational Flexibility Determines Electronic Coupling and Charge

Recommend Documents