It takes more than an imine: The role of the central atom on the electron-accepting ability of benzotriazole and benzothiadiazole oligomers

Article

pubs.acs.org/JACS

It Takes More Than an Imine: The Role of the Central Atom on the

Electron-Accepting Ability of Benzotriazole and Benzothiadiazole

Oligomers

Dinesh G. (Dan) Patel, Fude Feng, Yu-ya Ohnishi,^†Khalil A. Abboud, So Hirata,^†Kirk S. Schanze,*

and John R. Reynolds*^,#

Department of Chemistry, Center for Macromolecular Science and Engineering, University of Florida, Gainesville, Florida 32611,

United States

S

* Supporting Information

ABSTRACT: We report on the comparison of the electronic and

photophysical properties of a series of related donor−acceptor−

donor oligomers incorporating the previously known 2H-benzo[d]-

[1,2,3]triazole (BTz) moiety as the acceptor and the recently

reported BTzTD acceptor, a hybrid of BTz and 2,1,3-benzothiadia-

zole (BTD). Although often implied in the polymer literature that

BTz has good acceptor character, we show that this moiety is best

described as a weak acceptor. We present electrochemical, computa-

tional, and photophysical evidence supporting our assertion that

BTzTD is a strong electron acceptor while maintaining the alkylation

ability of the BTz moiety. Our results show that the identity of the central atom (N or S) in the benzo-fused heterocyclic ring

plays an important role in both the electron-accepting and the electron-donating ability of acceptor moieties with sulfur

imparting a greater electron-accepting ability and nitrogen affording greater electron-donating character. We report on the X-ray

crystal structure of a BTzTD trimer, which exhibits greater local aromatic character in the region of the triazole ring and contains

an electron-deficient sulfur that imparts strong electron-accepting ability. Additionally, we examine the transient absorption

spectra of BTzTD and BTz oligomers and report that the BTz core promotes efficient intersystem crossing to the triplet state,

while the presence of the thiadiazole moiety in BTzTD leads to a negligible triplet yield. Additionally, while BTz does not

function as a good acceptor, oligomers containing this moiety do function as excellent sensitizers for the generation of singlet

oxygen.

leading to easily polymerizable monomers. It is usually assumed

that the polarizable imine unit is responsible for the electron-

accepting character. The incorporation of long alkyl chains on

the donor groups, however, is crucial to gain the required

solubility necessary for solution processing, as these acceptors

themselves cannot be symmetrically alkylated. Within the

context of small molecule DSSCs, a large fraction of all-organic

sensitizers rely on the cyanoacrylic acid group⁵(CAA, Figure

1), which acts as both the electron acceptor and the anchoring

group to the metal oxide film. Because the CAA anchor is not a

particularly strong acceptor,⁶such dyes typically incorporate

additional acceptor moieties or have extended conjugation to

shift the dye absorption maximum into the red region of the

spectrum. Some research groups are now turning to stronger

acceptors such as quinoxaline⁷and BTD to induce this spectral

shift.⁸The use of BTD, however, necessitated a thiophene

spacer containing the requisite carboxylate group for binding to

titanium dioxide.⁸

INTRODUCTION

■

Visible light absorbing conjugated polymers, oligomers, and

small molecules are critical components for the efficient

function of photonic devices. For example, solar energy

conversion devices utilizing dye-sensitized nanostructured

metal oxide thin films (dye-sensitized solar cells, DSSCs)

have exceeded 10% efficiency,¹while organic polymeric devices

are fast approaching a similar metric.²In all-organic polymer

and oligomer systems, visible light absorption can be achieved

by the interaction of covalently linked, π-conjugated electron

donor and electron acceptor groups,³and the wide variety of

donor and acceptor units available allows for the tuning of

electronic and optical properties.⁴This donor−acceptor (D−

A) interaction, where the HOMO of the donor and the LUMO

of the acceptor are largely responsible for the location of the

frontier orbitals in the D−A system, is illustrated in Figure 1.

The frontier orbitals in turn determine the light absorbing

characteristics of the material.

In polymeric and oligomeric systems, 2,1,3-benzothiadiazole

(BTD, Figure 1) and similar diimine containing units such as

benzobis(thiadiazole) (BBT, Figure 1) are often employed as

the acceptor because they can be coupled with donor units

Received: August 29, 2011

Published: January 31, 2012

2599

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

simultaneously with the Grimsdale group,¹⁰of a new acceptor,

benzo(triazole-thiadiazole), which is a hybrid of BTD and BTz

and that we abbreviate as BTzTD.

The study herein combines experimental and comprehensive

computational data, which we show calls into question the

conventional idea that inclusion of an imine or diimine group in

the molecular framework of a π-conjugated system will

necessarily lead to strong electron-accepting ability. We show

that the identity of the central atom in benzofused systems,

such as those presented here, is of critical importance in

determining the LUMO energy level but also has a surprising

effect on the HOMO level in D−A systems as well. While

Grimsdale¹⁰and others¹¹have discussed the HOMO level for

sulfur, selenium, and tellurium containing BTD and BBT

analogues, a crucial qualitative and quantitative discussion

comparing sulfur and nitrogen analogues (BTz vs BTD and

BTzTD vs BBT) has not yet been put forth, and this is

important given that they are more stable and more commonly

found in organic materials than their selenium and tellurium

analogues.

The new BTzTD acceptor introduced by Grimsdale’s

group¹⁰and ours helps fill a gap in that it is a strong electron

acceptor and is capable of further functionalization via N-

alkylation; therefore, we believe it will find much utility in

organic electronics. Importantly, our route introduces the

nitration of BTD before any acid labile groups are introduced,

while the Grimsdale route subjects the substituted BTz moiety

to nitration; therefore, acid-sensitive groups (such as those used

for binding to metal oxides) cannot be used as substituents

using the latter approach. As an example, esters, silanes, and

alcohols are relevant if BTzTD containing oligomers and

polymers are to find use in surface modification applications

and as other materials such as conjugated polyelectrolytes. In

our study, the photophysical and electrochemical properties of

ester-substituted BTzTD are put into perspective by compar-

ison of its properties to analogous oligomers containing BTz,

BTD, BBT, and tetrazine (oligomer 7) cores. We abbreviate

the tetrazine core as TZ and use it for comparison purposes in

our study. The family of oligomers under study is collectively

shown in Chart 1.

Figure 1. Energy level diagram showing how the molecular oribtals of

a donor and acceptor interact in a D−A system with a reduced

HOMO−LUMO gap. Also shown are the molecular structures of the

cyanoacrylic acid (CA) acceptor and anchoring group commonly used

in DSSCs, the benzothiadiazole (BTD) acceptor, the benzotriazole

(BTz) moiety, the benzobis(thiadiazole) (BBT) acceptor, and the

benzo(triazole-thiadiazole) (BTzTD) hybrid acceptor.

We were interested in the use of symmetric, alkylatable

electron acceptors as part of our larger research program on

visible light absorbing conjugated poly- and oligo-electrolytes.

Our investigation began using the benzotriazole moiety (BTz,

Figure 1) because of the potential for alkylation at the central

nitrogen and because recent reports suggest that in polymer

systems it is electron deficient and therefore acts as an electron-

accepting moiety.⁹As part of this present study, however, we

report that BTz does not function as a particularly strong

electron acceptor, which resulted in our design and synthesis,

Chart 1. Molecular Structures of BTz (Series 1 Compounds and Compound 3), BTzTD (Series 2 Compounds and Compound

a

4), BTD (5), BBT (6), and Tetrazine (7)-Containing Oligomers

a

The hyphenated C, E, and H descriptors are used to denote oligomers used for computations, electrochemistry experiments, and those that are

unalkylated, respectively.

2600

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

goals. Accordingly, we noted that synthetically the BTz, BTD,

and BBT moieties are accessible via ortho-diaminobenzene (o-

DAB) precursors (compound 10 for BTz and BTD and

compound 16 for BBT in Schemes 1 and 2). To arrive at BTD

RESULTS AND DISCUSSION

■

Synthetic Design. The BTz moiety has been explored as

an acceptor only within the past few years, usually within the

context of polymeric systems.⁹The groups of Toppare^9a−c,12

and others^9ehave reported direct bromination of N-alkylated

BTz to allow for subsequent metal-catalyzed coupling reactions

in the construction of monomer units.^9bWe found that an

indirect procedure utilizing commercially available 2,1,3-

benzothiadiazole (8) as a precursor for BTz oligomers gave

high product yield and purity in our hands. As outlined in

Scheme 1, 8 is brominated giving 4,7-dibromobenzo[c][1,2,5]-

Scheme 2. Synthetic Route to Alkylated BTzTD Containing

Trimers 2-H and 2

Scheme 1. Synthetic Route to BTz Oligomers 1 and 3

or BBT, the appropriate precursor is treated with thionyl

chloride¹³or N-thionylaniline in pyridine.¹⁷If compound 16

were instead treated with sodium nitrite under acidic

conditions, a BTzTD containing oligomer would be obtained.

Given that BTzTD contains the strongly accepting BTD

moiety, effectively a hybrid of BTD and BTz, we hoped that it

would fulfill our design goals.

The synthesis of BTzTD containing oligomers 2-H and 2 is

outlined in Scheme 2. We use the descriptor H to indicate that

the acceptor is in its protonated, unalkylated form. As with

BTz, dibrominated compound 9 is a key precursor. Compound

9 is nitrated to give 14 using a mixture of fuming nitric acid and

fuming sulfuric acid^17,18or via a new nitration route developed

by Andersson and co-workers,¹⁹which employs fuming nitric

and trifluoromethane sulfonic acids. This new route affords

higher yields and obviates the need for tedious chromatography

to separate 14 from the mononitrated side product. We next

use Stille methodology to couple 14 with 2-(tributylstannyl)-

thiophene to afford trimer 15.¹⁷The nitro groups are reduced

using iron powder in acetic acid to give diamino-trimer 16,^17,20

which, as mentioned earlier, is similar to the precursor required

for BBT compounds. Trimer 16 is cyclized using sodium nitrite

in acetic acid^9fto give the unalkylated BTzTD trimer 2-H.

Compound 2-H is a maroon colored solid with limited

solubility in common organic solvents. Fortuitously, and unlike

molecule 11, trimer 2-H has C_2hsymmetry that allows for

alkylation yielding a single symmetric product. We assign this

symmetric structure for 2-H based on ¹H and ¹³C NMR

spectra.

To alkylate the BTzTD acceptor core, we initially tried using

the aza-Michael addition that had succeded in giving BTz

containing oligomers 12a and 12b. We found that the reaction

proceeded sluggishly, and attempts to drive the reaction toward

the product by adding excess t-butyl acrylate gave complicated

NMR spectra. Analysis of the crude material by mass

spectrometry suggests that adding excess acrylate leads to the

formation of BTzTD initiated acrylate oligomers. This

oligomeric mixture was not characterized further. The

reversibility of the aza-Michael addition coupled with the

enhanced electron-accepting ability, and therefore stability of

deprotonated 2-H, likely meant that the equilibrium tended

toward the starting material. To circumvent this, we hoped that

a deprotonation step followed by an S_N2 reaction with a

halogenated t-butyl protected carboxylic acid would afford the

desired alkylated oligomer. In fact, this route worked in near

thiadiazole (9), which is reduced with sodium borohydride in

ethanol to give 1,2-diamino-3,6-dibromobenzene (10).¹³

Compound 10 is then cyclized under acidic conditions with

sodium nitrite to give the asymmetric dibrominated benzo-

triazole 11.^9fWe assign this asymmetric structure based on the

¹H and ¹³C NMR spectra of this compound. To alkylate 11, we

chose to use the aza-Michael addition reaction employing tert-

butyl protected acrylic acid and N,N-dimethylaminopyridine

catalyst, for which there is literature precedence.¹⁴This reaction

is reversible and, as a consequence of BTz functioning as a

good leaving group,¹⁵only afforded the alkylated product in

moderate yield. We were able to isolate desired symmetric

product 12a in 53% yield, while the asymmetric product 12b

was obtained in 6% yield. We surmise that the bromine atoms

in 11 sterically hinder the benzylic nitrogen atoms, thus

inhibiting reaction at those sites and affording 12a as the major

product. Compound 12a can be coupled with commercially

available 2-(tributylstannyl)thiophene under Stille conditions to

give trimer 1. Because 12b forms in lower yield, we chose not

to pursue oligomers based on this asymmetric BTz core.

Trimer 1 is brominated with N-bromosuccinimide (NBS) in

chloroform to give dibrominated compound 13, which is then

coupled with 2-methyl-5-(trimethylstannyl)thiophene¹⁶under

Stille conditions to give pentamer 3.

The aforementioned oligomers were appealing because they

have the ability both for symmetric N-alkylation on the

acceptor and for symmetric disubstitution with aromatic donor

moieties in keeping with our desire for conjugated D−A−D

type systems. Equally important for our design goals is the

ability of these D−A−D systems to efficiently absorb long

wavelength visible light. Preliminary spectral and electro-

chemical experiments suggested that oligomers incorporating

the BTz acceptor would not fulfill the second half of our design

2601

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

quantitative yield when employing triethylamine as the base in

DMF solvent to give alkylated oligomer 2 (Scheme 2). The

alkylated oligomer 2 is dark blue in color.

Synthesis of oligomer 4 (Scheme 3) proceeds by

bromination of 15 with NBS using an acetonitrile/DMF

Scheme 3. Synthesis of BTzTD Containing Oligomer 4

solvent system.²⁰Catalytic aqueous HBr helps drive the

reaction. Reduction of the resulting dibromo compound 17²⁰

is effected in acetic acid using iron powder, and resulting

diamine 18 is cyclized using sodium nitrite in acetic acid to give

oligomer 19. The alkylation of 19 in DMF and promoted by

triethylamine gives 20, which is coupled under Stille conditions

with 5-methyl-2-trimethylstanylthiophene to give oligomer 4.

Oligomer 4 is slightly soluble in chlorinated solvents and THF.

Purification is achieved first by column chromatography, and

then reprecipitation of a concentrated chloroform solution of 4

into rapidly stirring hot hexane.

For electrochemical characterization, methyl terminated

pentamers 3 and 4 are used in addition to methyl terminated

trimers 1-E and 2-E (shown in Chart 1). The synthesis of 1-E

and 2-E is detailed in the Supporting Information. Methylation

at the terminal α-position on the thiophene rings prevents

polymerization during the cyclic voltammetry (CV) and

differential pulse voltammetry (DPV) experiments used to

assign the HOMO and LUMO energies.²¹

Computational Analysis: Part 1. In tandem with our

synthetic efforts, we computationally analyzed and compared a

series of five oligomers that include the well-known BTD

(oligomer 5) and BBT (oligomer 6) acceptors as well as

methylated BTz and BTzTD containing oligomers 1-C and 2-

C (Chart 1). We chose density functional theory (DFT)

methods, known to have a good balance between accuracy and

the computational time required.²²Methyl substituents were

chosen to simplify the computations with the computational

details supplied in the Experimental Section. Tetrazine (TZ)

containing oligomer 7²³(Chart 1) was examined for

comparison purposes because it contains only electron-

withdrawing imine nitrogens and does not include trivalent

nitrogen or divalent sulfur atoms as part of a fused heterocyclic

ring. As we shall show, the presence and identity of the central

atom (N or S in BTz or BTD) has a substantial impact on the

LUMO level, and, surprisingly, also the HOMO level of the

oligomeric system.

Figure 2. Graphical representation of the gas-phase calculated HOMO

and LUMO values for oligomers 1-C, 2-C, and 5−7.

For the discussions that follow, we will mainly discuss the

values in vacuo.

Using the calculated LUMO level as a crude predictor of

electron-accepting ability²⁴in accordance with Koopmans’

theorem, the computational trends show that BTz oligomer 1-

C is a weak electron acceptor. Its calculated LUMO level, at

−2.23 eV, is higher than BTD containing oligomer 5, the next

weakest acceptor, by 0.66 eV. While oligomers 1-C and 5 are

isoelectronic, it is clear from both our experimental results and

the computations that the substitution of NCH₃for S has a

profound impact on the electron-accepting ability with sulfur

imparting greater electron-accepting character as we detail

below. Notably, the results for BTzTD oligomer 2-C are

encouraging as the LUMO level calculated for this oligomer is

between that of the BTD and BBT oligomers and considerably

lower than that for the BTz oligomer. Consequently, we believe

that BTzTD would indeed make a good electron acceptor. On

the basis of our computations, we assign acceptor strength as

BTz < BTD < TZ < BTzTD < BBT.

Oligomers: Spectral Characterization. In a polymeric

context, BTz has been used as an electron acceptor to afford

polymers that absorb in the visible region of the electro-

magnetic spectrum.^9,25Initially, we reasoned that BTz would

also function as an acceptor in oligomeric systems given that it

has an electron-accepting diimine component as part of its

benzannulated structure. A recent report by You and co-

workers,²⁵however, points out that the electron-rich nature of

BTz results in an increase in the LUMO level, thus decreasing

its electron-accepting ability. No references were provided for

this premise, however. Our own expeiments presented here,

including spectral data and electrochemical measurements,

suggest that BTz acts more as an electron neutral, rather than

electron accepting, moiety in oligomeric systems. In polymeric

systems, we believe that BTz presents the appearance of a good

electron acceptor because of the band structure and decrease of

the HOMO−LUMO gap that results when oligomeric units are

brought together to give extended conjugated structures.

In methylene chloride solution, BTz oligomer 1 has an

absorption band in the near-UV region with λ_maxat 278 and

387 nm and an absorption band edge at 440 nm (Figure 3,

The HOMO and LUMO levels obtained computationally in

the gas phase are summarized in Figure 2. It is noted that while

both the HOMO and the LUMO values computed in

dichloromethane using the polarized continuum model

(PCM) are slightly lower than those in vacuo, the trends of

these values in dichloromethane parallel those in vacuo. These

exact numerical values are given in the Supporting Information.

2602

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

the actinometer.²⁸While the observed changes in the

absorption spectra on going from 1 and 3 are consistent with

a molecular structure containing extended conjugation, a

significant portion of the absorptive transitions still occurs in

the UV region.

Given the position of the absorption maxima, the absorption

spectra for both 1 and 3 do not appear to be consistent with a

D−A−D type structure where there is effective charge transfer,

leading us to conclude that shorter BTz containing oligomers

will likely not make effective visible light absorbing

chromophores, despite the peak extinction coefficients of 3.1

× 10⁴and 4.0 × 10⁴M⁻¹cm⁻¹, for 1 and 3, respectively.

By contrast to the properties of BTz-based oligomers, the

BTzTD-based oligomers exhibit spectral features as predicted

by the computations, with transitions in the long wavelength

region of the visible and near-infrared (NIR) spectrum,

Figure 3. Absorption (top) and normalized emission (bottom) spectra

of oligomers 1 (−) and 3 (− − −) in methylene chloride solution.

top). We ascribe the 287 nm band to a transition polarized

along the short axis of the molecule, while the 387 nm

absorption band is due to a long-axis transition analogous to

what was proposed by Fraind and Tovar²⁶for a thiophene−

anthracene−thiophene trimer. For comparison, terthiophene

has a long wavelength absorption maximum at 355 nm in

chloroform solution, with a shorter wavelength transition at 245

nm.²⁷Oligomer 1, in methylene chloride solution, is also

fluorescent with λ_emat 456 nm and a quantum yield of 0.44 in

ethanol when using coumarin 480 as the actinometer²⁸(Figure

3). These high-energy transitions suggest that the donor−

acceptor interaction is minimal in this three-ring oligomer.

For use in organic photonic devices, a material must be able

to adequately absorb in the longer wavelength region of the

visible spectrum. As oligomer 1 does not effectively accomplish

this goal, we sought to decrease the energy of the absorption

band of BTz containing oligomers by extending the

conjugation through successive addition of donor groups.

Oligomer 3 is a pentamer where the BTz core is flanked by a

total of four thiophene donor units. The absorption of 3 in

methylene chloride shows two major bands with λ_maxvalues of

320 and 450 nm (Figure 3). The absorption band edge appears

at approximately 512 nm. Additionally, oligomer 3 shows

fluorescence emission with emission maxima at 516 and 546

nm with a quantum yield of 0.32 when using coumarin 480 as

Figure 4. Absorption (top) and emission (bottom) spectra of

oligomer 2-H (−), 2 (− − −), and 4 (· · ·) in methylene chloride.

indicating they have comparatively small HOMO−LUMO

gaps. In particular, in methylene chloride solution, unalkylated

oligomer 2-H has absorption maxima at 330 and 540 nm

(Figure 4, top). The prominent charge transfer band is well-

defined and separated from the higher energy absorption band,

which we attribute to a short-axis transition due to the BTzTD

moiety. Similar to BTz oligomers 1 and 3, we attribute the

short wavelength absorption to a transition along the short axis

2603

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

Table 1. Electrochemical Data for D−A−D Oligomers

ox

red

onset

oligomer

acceptor

E_1/2(V)

E_ox

(V)

0.42

E_red

(V)

HOMO (eV)

LUMO (eV)

ref

ab

,

1-E

BTz

0.55

−2.44

−2.30

(−2.25)

−5.54

(−5.50)

[−5.44]

−5.32

(−5.24)

−5.34

(−5.25)

[−5.23]

−5.17

(−5.05)

−5.90

−2.82

(−2.87)

[−2.23]

−2.99

(−3.06)

−3.85

(−3.92)

[−3.25]

−3.89

(−3.97)

−3.49

c

(0.38)

ab

,

3

BTz

0.32

0.38

−2.23

−1.37

0.20

(0.12)

0.22

−2.13

(−2.06)

−1.27

c

a

2-E

BTzTD

(0.13)

(−1.20)

4

5

6

7

BTzTD

0.20

−1.26

−1.68

−0.93

−1.22

0.05

−1.23

(−1.15)

−1.63

c

(−0.07)

0.78

d

g

BTD

−

30

31

23

[−5.66]

−5.32

[−5.40]

−7.30

[−2.89]

−4.28

[−3.78]

−4.10

e

g

BBT

−

0.55

1.18

−0.84

−1.02

f

g

TZ

−

[−6.43]

[−2.95]

a

Energies are reported relative to the Fc/Fc⁺redox couple and are obtained from CV experiments with the corresponding DPV experiment values

shown in parentheses and the energies computed by DFT methods in this work shown in brackets; the solvent employed is dichloromethane unless

b

ox

otherwise noted. Values are obtained from THF solutions with the exception of the E_1/2values, which are obtained from dichloromethane

c

d

solutions. Values are from data presented in this work. Values are obtained by CV in CH₂Cl₂solution; an SCE reference electrode calibrated

e

against Fc/Fc⁺was used; the onset of oxidation and reduction defines the HOMO and LUMO levels, respectively. Values are obtained by CV in

benzonitrile solution; an SCE reference electrode was used; the onset of oxidation and reduction defines the HOMO and LUMO levels, respectively.

f

The value E_1/2^redand onset values reported here are converted to vs Fc/Fc⁺. The LUMO value is obtained by CV in dichloromethane solution using

an Ag/AgCl reference electrode; the HOMO is calculated using the onset of absorption added to the LUMO level. The onset values reported here

g

have been converted to vs Fc/Fc⁺. Half-wave potentials were not provided as the oxidation was not reversible.

of the molecule.²⁶Oligomer 2-H fluoresces with λ_emat 659 nm

and a quantum yield of 0.87 when using 3,3′-diethyloxadi-

carbocyanine iodide in ethanol as the actinometer (Figure 4,

bottom).

may also find utility in solid-state devices such as solar cells, we

also present data on the onset of oxidation and reduction versus

Fc/Fc⁺and on the absolute values of the HOMO and LUMO

levels relative to vacuum. As such, the results presented may be

used for direct comparisons with existing D−A molecules. A

compilation of the electrochemical data is given in Table 1.

Our analysis began with oligomers 1-E and 3 in dilute

methylene chloride solution using a platinum button working

electrode, nonaqueous Ag/AgNO₃reference electrode, and a

platinum flag counter electrode with tetrabutylammonium

hexafluorophosphate as the supporting electrolyte. We

observed that the solvent/electrolyte system reduced before

any reductive processes associated with the oligomers.

Specifically, sweeping the potential from 0 to −2.25 V vs Fc/

Fc⁺gave rise to a sharp increase in current flow, but this was

observed regardless of whether or not the oligomers were

present in solution.

While not commonly used as a solvent for electrochemical

experiments, THF does allow access to a larger electrochemical

window for reductive processes.³²With this in mind, both 1-E

and 3 were examined in THF solution and exhibit reductive

processes by CV. Both compounds were also examined by

DPV, and these data are presented in Figure 5. By CV,

oligomer 1-E shows quasi-reversible reduction centered at

−2.44 V versus Fc/Fc⁺(onset of reduction at −2.30 V by CV

and −2.26 V by DPV). Oligomer 3 also exhibits a quasi-

reversible reduction centered at −2.23 V versus Fc/Fc⁺(onset

of reduction at −2.13 V by CV and −2.06 V by DPV). In

summary, the electrochemistry shows that that BTz oligomers

1-E and 3 have relatively high energy LUMO levels as reflected

by the especially negative cathodic reduction potentials.

The HOMO level of each BTz oligomer was also

investigated electrochemically (Figure 5). In THF solution,

oligomer 1-E shows an irreversible oxidation wave with an

onset at approximately 0.42 V by CV (0.38 V by DPV). In

Upon alkylation of oligomer 2-H to give 2, we observe a

significant bathrochromic shift in the charge transfer band that

has a peak at 630 nm, while the higher energy band is only

shifted by 6 nm and is now at 336 nm (Figure 4, top).

Oligomer 2 fluoresces with λ_emat 723 nm and a quantum yield

of 0.57, also with 3,3′-diethyloxadicarbocyanine iodide in

ethanol as the actinometer (Figure 4, bottom). As with 2-H,

the two absorption bands are well-defined with no overlap. The

onset of absorption for 2 is 730 nm (1.7 eV), close to the DFT

computed HOMO−LUMO gap of 2.03 eV in dichloro-

methane. When comparing the absorption spectra of 2-H and

2 to BTz analogue 1, it is clear that there is greater charge

transfer character associated with the BTzTD containing

systems.

Extension of the conjugation length and moving from

oligomer 2 to oligomer 4 is accompanied by a further red-shift

in the absorption maxiumum. Oligomer 4 shows three well-

defined absorption transitions at 338, 409, and 729 nm and has

a fluorescence maximum at 870 nm (Figure 4). A fluorescence

quantum yield of 0.06 is measured when using ZnTPP as the

actinometer.²⁹Worthy of note is the fact that oligomer 4

exhibits low solubility in most polar organic solvents, but even

when present at low concentration it produced darkly colored

solutions due to its high extinction coefficient (3.6 × 10⁴M⁻¹

cm⁻¹at 409 nm and 3.3 × 10⁴M⁻¹cm⁻¹at 729 nm).

Oligomers: Electrochemical Characterization. To char-

acterize the oligomers under study, we utilize both cyclic

voltammetry (CV) and differential pulse voltammetry (DPV)

methods and assign the half-wave potentials versus the

ferrocene−ferrocenium (Fc/Fc⁺) redox couple as is common

in the literature. Because the D−A−D oligomers in our work

2604

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

Figure 6. Cyclic voltammetry (CV, solid lines) and differential pulse

voltammetry (DPV, dashed lines) data for for 2-E (bottom) and 4

(top). The data were obtained in dichloromethane solution at 5 mM

concentration using 0.1 M tetrabutylammonium hexafluorophosphate

as the supporting electrolyte, a platinum button working electrode, a

nonaqueous Ag/AgNO₃reference electrode, and a platinum flag

counter electrode. The scan rate was 20 mV/s for 2-E and 50 mV/s for

4.

Figure 5. Cyclic voltammetry (CV, solid lines) and differential pulse

voltammetry (DPV, dashed lines) data for for 1-E (bottom) and 3

(top). The data were obtained in THF solution at 5 mM

concentration using 0.1 M tetrabutylammonium hexafluorophosphate

as the supporting electrolyte, a platinum button working electrode, a

nonaqueous Ag/AgNO₃reference electrode, and a platinum flag

counter electrode. The scan rate was 20 mV/s.

contrast, oligomer 3 shows a quasi-reversible oxidation centered

at 0.25 V (onset of oxidation at 0.20 V by CV and 0.12 V by

DPV). The lower potential required to oxidize 3 when

compared to 1-E is consistent with a system that is more

conjugated. Interestingly, 1-E and 3 both show reversible

oxidation in methylene chloride solution with their oxidation

onsets being 0.48 and 0.28 V, respectively, by CV (Supporting

BTz analogues. The electrochemical energy gap, defined as the

difference between the onset of oxidation and reduction, is thus

determined to be 1.49 eV by CV and 1.33 eV by DPV for 2-E.

These values are smaller than the HOMO−LUMO gap

determined by DFT (2.03 eV) and the optical energy gap

determined experimentally (1.7 eV). The discrepancy with the

computed values can be accounted for by solvation effects,

which are expected to reduce the value of E_gby stabilizing the

molecular dipoles in both the ground and the excited states.

The HOMO−LUMO energy gap for oligomer 4 was not

computed by DFT methods, although the electrochemical data

gives a HOMO−LUMO gap of 1.28 eV with the corresponding

optical energy gap being 1.4 eV.

ox

Information, Figure S1), similar to the THF figures. The E_1/2

value for 1-E in methylene chloride is 0.55 V, while the value

for 3 is 0.32 V. The fact that the oxidation process for 1-E is

fully reversible in methylene chloride but not in THF implies

that THF reacts with the cation radical, possibly by acting as a

nucleophile or as a base to promote decomposition.

The electrochemical results for oligomers 1-E and 3 are

summarized in Table 1 along with values for oligomers 2-E and

4, which will be discussed shortly. With both spectral and

electrochemical evidence suggesting that BTz containing

oligomers would not function as effective visible light absorbing

materials, we turned our attention to BTzTD containing

oligomers.

The compilation of our experimental electrochemical results

as well as the data previously reported for 5, 6, and 7 is also

given in Table 1. The onset potentials are used to estimate the

HOMO and LUMO values relative to vacuum level for the

ox

oligomers by using the conversion HOMO (eV) = E_onset

+

5.12 V and LUMO (eV) = E_onset^red+ 5.12 V.³³Worthy of note

is the difference between the HOMO levels of 1-E and 2-E.

The more electron-rich oligomer 1-E would be predicted to be

easier to oxidize relative to 2-E, which incorporates the better

electron-accepting BTzTD group. However, 2-E is easier to

oxidize by about 0.2 V. This supports the notion that BTzTD is

not only a strong electron-accepting group, but also functions

as an effective electron-donating group. The enhanced electron-

donating ability of BTzTD relative to BTz was predicted by our

DFT calculations. As stated earlier, and based on crystallo-

graphic and computational evidence (vide infra), the electron

donation is likely enhanced due to the combined effects of the

alkylated central nitrogen, which imparts “pyrrole-like” electron

donor character to the hybrid acceptor, and the electropositive

sulfur atom. The electron-accepting order predicted by DFT

methods parallels the order obtained experimentally. Further-

more, computations correctly predicted that 7 would have the

lowest energy HOMO, a fact confirmed by experiment.

When examined by CV and DPV in methylene chloride,

BTzTD oligomers 2-E and 4 show reversible or quasi-reversible

oxidative and reductive processes (Figure 6). Compound 2-E

has a reversible oxidation with E_1/2^oxof 0.38 V vs Fc/Fc⁺(onset

at 0.22 V by CV and 0.13 V by DPV). Upon extension of the

donor portion, compound 4 shows quasi-reverisble oxidation at

a lower potential centered at 0.20 V vs Fc/Fc⁺(onset at 0.05 V

by CV and −0.07 V by DPV). Remarkably, both 2-E and 4

exhibit lower oxidation potentials when compared to their BTz

analogues, 1-E and 3. Compounds 2-E and 4 both show

red

reversible reduction with E_1/2values of −1.37 and −1.26 V,

respectively, vs Fc/Fc⁺. The onset of reduction for 2-E is −1.27

V by CV (1.20 V by DPV), while the onset for 4 is −1.23 V by

CV (−1.15 V by DPV). In summary, the BTzTD oligomers

both show lower reduction potentials than the BTz oligomers.

Electrochemically, both of the BTzTD oligomers under

examination have reduced energy gaps when compared to their

2605

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

Figure 7. Graphical representation of the computed molecular orbitals for oligomers 1-C, 2-C, and 5−7, which are ordered according to the acceptor

core.

Computational Analysis: Part 2. To understand how

replacement of NCH₃with S in comparing BTz and BTD in

the core heterocycle ring system increases electron-accepting

ability, we chose to visualize the HOMO and LUMO of our

oligomer series, shown collectively in Figure 7. When

comparing BTz oligomer 1-C and BTD oligomer 5, both

molecules show delocalization of the HOMO across the

molecule, with the LUMO being shifted more toward the

acceptor core. The LUMO for both oligomers assumes a more

quinoidal form as shown by the presence of orbital lobes in the

bonds between the donor and acceptors. This orbital lobe

pattern is similar to the HOMO of a series of thiophene

oligomers that are quinoidal in the ground state.³⁴As indicated

by the bond length and Wiberg bond indices of the C−C bond

between the acceptor and thiophene (Table 2), it is considered

energy is computed to be −2.84 eV in vacuo, which is closer to

the computed LUMO energy of 5 (−2.89 eV) than to the

computed LUMO value of 1-C (−2.23 eV). Therefore, we can

conclude that, at least partially, the difference in accepting

ability between BTz and BTD stems from the difference in the

imine−central atom bond length.

Another major difference between the 1-C and 5 pair of

oligomers is the size of the molecular orbital lobe on the central

atom in the LUMO. Specifically, the lobe on the central

nitrogen atom in BTz is much smaller than the lobe on sulfur in

BTD. We would expect the imine nitrogens to act as electron-

accepting atoms, and sulfur, being larger and more polarizable,

likely helps to stabilize any charge buildup on these imine

nitrogens that arises in the charge transfer state produced by

photoexcitation. The central nitrogen in the case of BTz is

unable to provide as much stabilization as it is smaller and less

polarizable, therefore limiting the electron-accepting ability of

BTz.

a

Table 2. Donor−Acceptor C−C Bond Lengths and Wiberg

Bond Indices

The molecular orbitals of BTzTD and BBT oligomers 2-C

and 6 show features similar to those of the BTz and BTD

oligomers 1-C and 5. The HOMO levels are deocalized across

the whole of the molecules with the LUMO being localized

mostly on the acceptor (Figure 7). Quinoidal character is

observed in the LUMO molecular orbitals. However, because

the BTzTD unit in oligomer 2-C also has a sulfur atom like the

BTD and BBT units present in 5 and 6, it has enhanced

electron-accepting ability as compared to the BTz unit in 1-C.

It should also be noted that the bond lengths and Wiberg bond

indices between the acceptor and the donor thiophene are

slightly shorter and larger, respectively, in 2-C and 6 than in 1-

C and 5 as shown in Table 2. Therefore, it is considered that

BTzTD and BBT enhance the quinoidal character as compared

to BTz and BTD.

As a final component to our computational analysis, we

compare key bond lengths and Wiberg bond indices, a measure

of bond order, for 1-C and 5 to understand the differences in

the electron-accepting ability in this pair (Table 3). A full table

of computed bond lengths and bond indices is available in the

Supporting Information; we focus here on the C−N and

annulated C−C bonds. The annulated C−C bond, which we

define as the bond shared by the five- and six-membered rings

in the acceptors, in BTz oligomer 1-C is computed to be

compound

acceptor

bond length (Å)

Wiberg bond index

1-C

5

BTz

1.454

1.455

1.445

1.121

1.124

1.156

1.167

BTD

BTzTD

BBT

2-C

6

a

The C−C bond is defined as the bond between the 2-position on

thiophene and the acceptor.

that BTz oligomer 1-C and BTD oligomer 5 have the same

degree of quinoidal character, and, thus, this is not the reason

for the difference in the LUMO level of these oligomers.

Because the LUMOs consist mainly of the π* orbital on the

acceptors, we focused on the computed geometrical difference

between 1-C and 5. The largest difference between these two

molecules is observed in the bond length between the imine

nitrogen and the central atoms: the N−N bond length in 1-C is

1.319 Å, and the N−S bond length in 5 is computed to be

1.630 Å. Other geometrical parameters are summarized in the

Supporting Information. It is reasonable to consider that the

longer N−S bond decreases the π* orbital energy. In fact, we

find that when we take the optimized geometry of oligomer 5

and replace the central S atom of the BTD moeity with NCH₃,

while holding all geometrical parameters constant, the LUMO

Table 3. Selected Bond Lengths and Wiberg Bond Indices for D−A−D Oligomers

bond length (Å)

Wiberg bond index

annulated C−C

compound

acceptor

annulated C−C

C−N

1-C

5

BTz

1.425

1.456

1.349

1.336

1.223

1.152

1.359

1.452

BTD

2606

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

shorter by 0.031 Å than the corresponding bond in BTD

oligomer 5. This correlates with a larger annulated C−C bond

index in 1-C, indicating that this bond has greater double bond

character in 1-C when compared to 5. Correspondingly, the

C−N bond lengths are shorter by 0.013 Å in 5 as compared to

1-C. The bond index for the C−N bonds in 5 is also greater,

indicating that they have more double bond character. While

specific figures are reported in the Supporting Information, the

trend observed for BTz and BTD extends to the BTzTD and

BBT oligomers examined computationally.

heteroatom. TZ containing oligomer 7 has the lowest HOMO

level among the series regardless of whether the value in vacuo

or in dichloromethane is considered. As an explanation, we

believe that the HOMO energy is lowered because TZ contains

only sp²hybridized, electron-deficient imine nitrogens and

because it lacks any atoms capable of electron donation through

lone pairs as in the case N or S in BTz, BTD, BTzTD, and

BBT. When comparing the closely related BTD and BTz

oligomers, the computations show that BTz has the higher

HOMO level. This suggests that the nitrogen atom has a larger

electron-donating effect when compared to sulfur. A trend can

be established when extending this comparison to BTzTD and

BBT, which, similar to BTz and BTD, differ only by one atom.

BTzTD has the higher HOMO level when compared to BBT

in vacuo and in dichloromethane, although the difference is

much smaller in dichloromethane. Again, this strongly suggests

that nitrogen has a stronger electron-donating effect that raises

the HOMO level of BTzTD when compared to sulfur in BBT.

This is not unexpected, however, because the lone pair on the

second period nitrogen is likely to have better orbital overlap

with the acceptor π-electron system than the lone pair on the

larger third period sulfur atom.

Using bond length and bond index data, we can assemble a

resonance picture for oligomers 1-C and 5, shown in Figure 8.

X-ray Crystallographic Analysis. With computational and

experimental data showing that BTzTD has increased electron-

accepting ability relative to BTz, we turned our attention to

characterizing the structures of 2-H, 2, 2-E, and 4. To identify

the structural features that allow BTzTD to function as a strong

acceptor and to allow comparison with computationally

predicted structural features, single crystals of 2 were grown

by slow evaporation of an acetone/hexanes solution and then

examined by X-ray diffraction techniques. The crystals obtained

were small, dark blue needles with oligomer 2 crystallizing in

the monoclinic C2/c space group with eight molecules in the

unit cell. The molecular structure and packing are shown in

Figure 9 with selected atoms labeled. A full table with bond

lengths and angles is supplied in the Supporting Information. It

Figure 8. Selected resonance contributors for oligomers 1-C (top) and

5 (bottom).

The bond lengths and indices suggest that there is some

intramolecular charge separation present in both 1-C and 5,

although the charge separation is greater in 5. Structures 1-

C(b) and 1-C(c) allow visualization of this charge separation in

the BTz oligomer, while 5b and 5c demonstrate this charge

separation in the BTD oligomer. Resonance forms 1-C(b) and

1-C(c) seem favorable when considering that the acceptor

exists as the annulation of two Clar sextets in the form of a

benzene ring and a zwitterionic triazole ring. While the same is

true for 5b and 5c, the bond length and bond index data

suggest that these two resonance forms contribute less than

they do in oligomer 1-C. Because of the enhanced aromatic

character in the BTz core, breaking that aromaticity requires

greater energy. In addition, the conjugation pathway that would

allow the two thiophenes to interact via conjugation is partly

disrupted by the presence of the aromatic benzene ring. This

effect is similar to that seen in a family of oligomers that contain

either naphthyl or methano[10]annulene cores.³⁵The naphthyl

oligomers show shorter wavelength absorption due to the

localization of electrons in a benzene structure according to

Clar’s rule, while the methano[10]annulene oligomers were

afforded full delocalization and longer wavelength absorption.

Consequently, we believe that because a symmetrically N-

alkylated BTz acceptor will tend toward resonance forms 1-

C(b) and 1-C(c), higher energies will be required to break

aromaticity. Thus, electron-accepting ability will be lower than

in the BTD analogues, which have smaller contributions from

resonance forms 5b and 5c.

should be noted that, similar to other D−A−D oligomers,^17,36

2

shows some disorder with respect to the orientation of the

thiophene rings relative to the core of the molecule. The

molecules pack in extended slip stacks with no obvious edge to

face interactions and intermolecular π−π distances of 3.359 Å,

similar to that observed for BBT oligomer 6, which has

intermolecular π−π distances of 3.42 Å.¹⁷To our knowledge,

there have been no crystal structures of D−A−D BTz

oligomers reported. In our own lab, we have had difficulty

obtaining single crystals of 1 and are thus unable to make a

direct structural comparison between BTz and BTzTD

containing oligomers at present.

The structure of oligomer 2 is highly planar with a rms

deviation from plane of 0.002 Å for the BTzTD acceptor core

and 0.173 Å for the thiophene−BTzTD−thiophene triad. The

larger deviation from plane in the triad reflects that the

thiophene rings are twisted slightly out of plane. The dihedral

angles defined by C12′−C11−C6−-C5 and C8−C7−C3−C4

are small at −0.29° and 5.30°, respectively, showing that there

is effective π-overlap between the thiophene donors and the

BTzTD core. This molecular planarity is driven partly by

nonbonded N···S interactions between the thiophene sulfur

atoms and the imino nitrogen atoms in the acceptor, a feature

that is well-known for other thiophene−acceptor−thiophene

and similar triads.^17,37We observe N···S distances of 2.804−

2.963 Å, close to values observed for similar systems.^17,37In

oligomers where such interactions are not possible, the

Also worthy of mention is the fact that the calculated

HOMO levels show a surprising amount of variation (Figure 2)

given that the HOMO energy should be dominated by

contributions from the thiophene donor units (shown in

Figure 1). In fact, to our knowledge, there has been only a small

discussion²⁵regarding the electron-donating ability moieties

generally viewed as electron acceptors, but it appears that

across the analogous series we examine, a non-negligible

amount of donor character arises from the central N or S

2607

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

imply the need for hypervalent sulfur in the presented

molecular schemes.⁴¹Thus, it is the hypervalent sulfur that is

thought to help impart strong electron-accepting ability by

allowing the π-electron system to easily adopt a quinoidal

structure through D−A interactions.⁴²However, as discussed

by Leusser and co-workers,⁴³it is incorrect to invoke expanded

octets in describing sulfur diimide type compounds, and,

moreover, the energetics of the sulfur d-orbitals prohibit their

participation in bonding, thus precluding an expanded octet. In

the case of oligomer 2, the N(1)−S(1) and N(2)−S(1) bond

lengths are longer than would be expected for a NS double

bond,^39,43indicating significant single bond character. There-

fore, we believe that the −N−S−N− linkage is best described

as a three-center-four-electron bond. For such a situation, the

sulfur atom is expected to bear a formal positive charge, thus

making it electron deficient and imparting electron-accepting

character on BTzTD. This electron deficiency in the ground

state would likely drive strong D−A interactions.

Given the bond lengths observed and without invoking

hypervalent sulfur, we propose several resonance contributors

to the overall observed structure for 2. These resonance forms

are depicted in Figure 10. Structures A and B depict the three-

Figure 9. X-ray crystal structure of oligomer 2 (top) and packing

diagram showing intermolecular π-stacking (bottom). Thermal

ellipsoids are shown at 50% probability. Hydrogen atoms have been

omitted for clarity; some molecules have been removed for clarity in

the packing diagram. S3', C12', S2', and C8' are the minor parts of the

disordered thiophene groups. The rest of the thiophene group atoms

are not disordered.

molecules are shown to adopt a structure that is much less

planar.³⁸The bond lengths of C11−C6 and C7−C3 are

1.453(2) and 1.4564(19) Å, respectively, surprisingly shorter

than the value of 1.471(6) Å for the analogous bonds in a

related BBT oligomer,¹⁷which indicates that there is a strong

donor−acceptor interaction between BTzTD and thiophene in

the present case. The C−O bond lengths in the ester group

(C16−O2 = 1.2043(17) Å and C16−O1 = 1.3306(17) Å) are

similar to values expected for an aliphatic ester system,³⁹

showing this group is electronically isolated from the oligomer

π-system due to the presence of the conjugation breaking

methylene group.

The bond lengths in the central six-membered ring do not

show the bond length pattern (short−long−short−long−long−

long) that is expected to result from the benzannulation of the

thiadiazole and triazole rings. The two longest bonds in the

BTzTD six-membered ring are C1−C2 at 1.4579(19) Å and

C4−C5 1.4357(19) Å with C1−C2 being longer, which fits

with the bond length trend predicted by DFT computations in

the BTz and BTD oligomers. This suggests that there is some

local aromatic character in the triazole ring leading to a

shortening of C4−C5 relative to C1−C2.

Figure 10. Selected resonance contributors for oligomer 2.

center-four-electron bond at the −N−S−N− linkage. In both

structures, sulfur is electron deficient and bears a formal

positive charge, while the more electronegative nitrogen bears a

formal negative charge. Structures C, D, E, and F account for

electron donation from thiophene into either the thiadiazole (C

and D) or triazole (E and F) rings. In these structures, the

thiophene sulfur atoms bear a positive charge, and the imine

nitrogens bear a negative charge. Resonance forms H and I take

into account donation of the triazole central nitrogen to the

triazole imine nitrogens. In H and I, a benzene structure is

realized in the central six-membered ring as well as an aromatic

and zwitterionic triazole ring similar to what was proposed for

BTz oligomer 1-C in Figure 8.

The electron-accepting ability of BTzTD can be partly

explained by examining the nature of the −N−S−N−

framework, as the triazole ring is not expected to play an

important role. In the literature, compounds containing BBT

are usually explicitly described as containing hypervalent sulfur

atoms^17,40or shown containing a −NSN− linkage that

implies a hypervalent sulfur. Recent reports on polymers

containing −N−S−N− linkages along the main chain also

To support the resonance forms above, we used natural

population analysis (NPA),⁴⁴similar to the work by Leusser,⁴³

to examine the charge distribution on individual atoms in

oligomer 2-C. The NPA data were obtained as part of our DFT

calculations. In particular, we note that sulfur has a computed

2608

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

Table 4. Compilation of Photophysical Data for Compounds 1−4

parameter

λ_max(nm)

solvent

1

3

2-H

330, 540

659

0.87

2

4

CH₂Cl₂

278, 387

456

320, 450

516, 546

336, 630

723

0.57

338, 409, 729

870

λ_em(nm)

a

d

b

c

Φ_F

τ_F(ns)

0.44

0.32

0.06

2.08 (94%)

4.12 (6%)

1.01

0.73 (12%)

11.1 (88%)

0.44 (4%)

1.16 (73.7%)

10.3 (96%)

e

5.06 (26.3%)

d

e

Φ_Δ

CDCl₃

THF

0.53

0.67

−

f

TT_Abs(λ/nm)

TT_Abs(ΔA, t = 0)

τ_triplet(μs)

476

0.18

46.0

640

0.40

31.1

−

f

e

−

f

−

Measured using coumarin 480 in ethanol (Φ_F= 0.76) as the actinometer.²⁸Measured using 3,3′-diethyloxadicarbocyanine iodide in ethanol (Φ_F=

a

b

0.49) as the actinometer.⁴⁹Measured using ZnTPP (Φ_F= 0.04 in toluene) as the actinometer.²⁹Measured using benzophenone in oxygen-

c

d

saturated CDCl₃(Φ_Δ= 0.55) as actinometer.⁵⁰Not measured. Transient absorption not observed.

e

f

formal charge of +0.912 in vacuo (+0.939 in dichloromethane),

while the thiadiazole nitrogens have computed formal charges

of −0.627 in vacuo (−0.636 in dichloromethane). Because

sulfur is computed to have a formal charge of near unity, it is

electron deficient. In the excited state, electron density is

transferred from the donor groups to the thiadiazole moiety to

alleviate sulfur’s electron deficiency; thus, the electron-deficient

sulfur is a significant feature that imparts strong electron-

accepting ability on BTzTD. In contrast, the N−CH₃nitrogen

has a computed formal charge of 0.066 (0.076 in dichloro-

methane), while the imine nitrogens have computed formal

charges of −0.272 (−0.277 in dichloromethane) in BTz

oligomer. The N−CH₃unit has a formal charge of 0.344 (0.371

in dichloromethane). In contrast, BTz oligomer 2-C has smaller

formal charges, suggesting that excited-state charge transfer will

not occur as easily. A complete list of atom charges is available

in the Supporting Information.

the series of BTz and BTzTD oligomers that are the focus of

the present investigation, nanosecond−microsecond transient

absorption spectroscopy was carried out for samples in

degassed THF solution at room temperature. As shown in

Figure 11, BTz containing trimer 1 and pentamer 3 exhibit

Photophysics of Oligomers. To further characterize the

optical properties of the oligomers, we explored the solution-

phase excited-state dynamics for the BTz and BTzTD

oligomers. All of the oligomers, as discussed previously, show

strong fluorescence with moderate to high quantum yields.

Fluorescence lifetime measurements reveal that BTz containing

compounds 1 and 3 have lifetimes (τ_f) in the range of 1−2 ns

(Table 4). The radiative decay rates for 1 and 3 calculated by

the expression k_r= Φ_F/τ_fare in the range (2−3) × 10⁸s⁻¹. The

lifetimes and radiative decay rates for the 1 and 3 are typical for

π-conjugated oligomers with strongly allowed long-axis

polarized π,π* singlet excited states.⁴⁵By contrast, the

BTzTD compounds 2-H and 2 have fluorescence lifetimes

that are 1 order of magnitude larger than the BTz series

compounds (τ ≈ 10 ns), and the corresponding radiative rates

are significantly less (8 and 6 × 10⁷s⁻¹for 2-H and 2,

respectively). The significant difference in fluorescence lifetime

and radiative rates for 2-H and 2 likely reflects the fact that the

singlet state in the BTzTD oligomers has a significant degree of

charge transfer character, and it adds support to the notion that

the BTzTD unit is a strong acceptor. Also notable is the fact

that the fluorescence quantum yield for 4 is markedly less than

for the other oligomers. This effect is likely due to the energy

gap law, where the nonradiative decay rate is increased due to

the relatively low energy of the singlet excited state in this

oligomer.⁴⁶

Figure 11. Transient absorption spectroscopy of compounds 1 (−)

and 3 (− − −) in THF at room temperature. Spectra obtained at 100

ns delay following a 355 nm, 8 mJ excitation pulse.

strong transient absorption in the mid-visible region following

355 nm excitation. The transient absorption is quenched

rapidly by oxygen when the experiment is carried out in air-

saturated solution, leading us to assign it to the triplet−triplet

absorption of the lowest triplet state. For both oligomers, the

triplet state absorption decays with first-order kinetics, and in

argon outgassed solution the triplet lifetime is in the 30−50 μs

range (Table 4). Although absolute triplet yields were not

determined, the significant intensity of the triplet−triplet

absorption leads us to conclude that the triplet yield from

intersystem from the singlet to the lowest triplet is greater than

0.1, likely due to the significant spin−orbit coupling induced by

the thienyl sulfur.^47bNote that increasing the number of

thiophene rings in the oligomer system leads to significant

changes in the transient T₁→ T_nabsorption spectra.

Specifically, as compared to trimer 1, the triplet−triplet

absorption of pentamer 3 is red-shifted from 476 to 640 nm

and is more intense. The red-shift in the triplet−triplet

absorption suggests that the triplet state is more delocalized

in the longer oligomer system, and the increase in the

absorption implies that the triplet yield is larger in 3. Note

Thiophene containing oligomers typically exhibit long-lived

triplet excited states that are produced in moderate yield

following intersystem crossing from the singlet state that is

produced by excitation.⁴⁷To characterize the triplet state for

2609

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

that the possibility that the triplet yield is larger in 3 as

compared to 1 is consistent with the observation that the

fluorescence quantum efficiency is smaller for the longer

oligomer (note that fluorescence and intersystem crossing are

competing processes, so an increase in one pathway would be

reflected by a reduction in the other).

the solvent/electrolyte system employed and show UV and

high energy visible transitions that are not indicative of D−A

interactions. In contrast, a new acceptor, BTzTD, which is a

hybrid of BTz and BTD, proves to be both an excellent

electron acceptor as well as a good electron-donating group.

While BTz containing oligomers absorb mostly in the UV

region, the analogous BTzTD containing oligomers feature

absorption bands that extend into the long wavelength region

of the visible spectrum. The BTzTD oligomer 2 examined

electrochemically shows reduction at a potential slightly

negative to that of the conduction band of nanocrystalline

TiO₂. The X-ray crystal structure of a BTzTD trimer shows a

high level of planarity with short D−A bond lengths, suggesting

strong coupling across the D−A−D framework. The bond

lengths support the idea that the triazole moiety has greater

local aromatic character than the thiadiazole moiety. Transient

absorption and singlet oxygen emission spectroscopy were used

to probe the triplet states of the oligomers. These studies reveal

that BTz containing oligomers examined show strong triplet−

triplet absorption and the triplets can efficiently sensitize the

formation of singlet oxygen, implying that the triplet yields are

large for both oligomers studied. By contrast, these studies

show that the BTzTD containing oligomers do not undergo

efficient intersystem crossing to produce a triplet excited state.

In conclusion, we have demonstrated the ability to construct

donor−acceptor conjugated oligomers, which feature the

hybrid BTzTD unit in the core. These oligomers exhibit

strong visible light absorption as a consequence of the donor−

acceptor electronic structure. In preliminary work, we have

shown that the acid forms produced by hydrolysis of the tert-

butyl esters in 1 and 2 readily adsorb to nanocrystalline TiO₂

films. Work in progress seeks to characterize these systems as

visible light sensitizers for dye-sensitized solar cell application.

Interestingly, in contrast to the to the BTz oligomers,

transient absorption spectroscopy of the BTzTD containing

oligomers 2-H, 2, and 4 reveals no observable transient

absorption on the nanosecond−microsecond time scale due to

a triplet excited state, despite the fact that these two trimers

contain thiophene rings as do 1 and 3. From this observation,

we conclude that intersystem crossing in the BTzTD oligomers

is inefficient, which is surprising given that their fluorescence

quantum yields are only slightly higher than those for the BTz

oligomers. We suspect that the very low intersystem crossing

yield in the BTzTD oligomers is a consequence of the fact that

the singlet excited state has a significant degree of charge-

transfer character. Although a previous study explored triplet

energies in donor−acceptor oligomers,⁴⁸we are not aware of

any previous studies that have correlated the intersystem

crossing yield with the degree of donor−acceptor interaction in

π-conjugated oligomers or polymers. The interesting observa-

tion here that the triplet yield is low for the BTzTD oligomers

suggests that a systematic study of triplet yields in donor−

acceptor oligomers is warranted.

Finally, to further probe the triplet state characteristics of

BTz, we examined their ability to sensitize singlet oxygen in

solution, which should be facile given the strong transient

absorption of both 1 and 3 with τ_triplet> 10 s. Singlet oxygen

(¹O₂(¹Δ_g)) is produced via energy transfer from excited triplet

3

states to molecular oxygen according to the reaction, S* +

O₂(³∑_g⁻) → S + ¹O₂(¹Δ_g), where S is the (oligomer)

sensitizer. The ¹O₂that is produced by sensitization is detected

by monitoring its emission in the near-infrared region at 1270

nm;⁵⁰moreover, by measuring the singlet oxygen quantum

yield (Φ_Δ), it is possible to establish a lower limit for the

intersystem crossing yield in the sensitizer.⁵¹As anticipated,

BTz oligomers 1 and 3 efficiently sensitize ¹O₂(¹Δ_g), as

evidenced by the observation of strong emission at 1270 nm

from air-saturated CDCl₃solutions of the oligomers. The

quantum yield of sensitized singlet oxygen formation (Φ_Δ) was

determined for the two oligomers, and the values for 1 (Φ_Δ=

0.53) and 3 (Φ_Δ= 0.67) are comparatively large. Note that in

many cases the singlet oxygen yield is a direct measure of the

intersystem crossing efficiency of the sensitizer.⁵¹Thus, these

comparatively large values underscore the fact that intersystem

crossing is rather efficient in the BTz oligomers. Furthermore,

it is interesting to note that for both oligomers the sum of the

singlet oxygen yield and the fluorescence yield is close to unity

(i.e., Φ_F+ Φ_Δ≈ 1, Table 4). This fact indicates that for these

oligomers the predominant decay pathways for the singlet state

are fluorescence and intersystem crossing to the triplet excited

state.

EXPERIMENTAL SECTION

Complete synthetic procedures and NMR characterization for all new

compounds as well as specifics on the instrumentation and methods

employed are detailed in the Supporting Information.

■

ASSOCIATED CONTENT

* Supporting Information

■

S

X-ray crystallographic data for oligomer 2. Synthetic procedures

and NMR characterization for 1−4, 1-E, 2-E, 2-H, 12a, 13, 19,

and 20. Complete list of authors for references. Computational

parameters (table of energy values, atomic coordinates, NPA

analysis) for oligomers 1-C, 2-C, and 5−7. This material is

available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION

Corresponding Author

kschanze@chem.ufl.edu; reynolds@chemistry.gatech.edu

■

Present Addresses

^†Department of Chemistry, University of Illinois at

Urbana−Champaign, 600 South Mathews Avenue, Urbana,

Illinois 61801, United States.

SUMMARY AND CONCLUSION

■

^#School of Chemistry and Biochemistry, Georgia Institute of

Technology, Atlanta, GA 30332, United States.

We have shown that the BTz moiety, considered an electron

acceptor in polymeric systems, does not function as a good

electron acceptor in oligomeric systems, which we believe is a

consequence of the larger aromatic character of BTz relative to

other electron acceptors. BTz containing molecules show

reversible oxidation when examined by cyclic voltammetry, but

do not show reductive processes within the useful window of

ACKNOWLEDGMENTS

■

Dr. Khalil A. Abboud wishes to acknowledge the National

Science Foundation and the University of Florida for funding of

the purchase of the X-ray equipment. D.G.P. would like to

2610

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

(19) Wang, E. R.; Hou, L. T.; Wang, Z. Q.; Hellstrom, S.; Mammo,

W.; Zhang, F. L.; Inganas, O.; Andersson, M. R. Org. Lett. 2010, 12,

4470−4473.

(20) Perzon, E.; Zhang, F. L.; Andersson, M.; Mammo, W.; Inganas,

O.; Andersson, M. R. Adv. Mater. 2007, 19, 3308−3311.

(21) Salhi, F.; Lee, B.; Metz, C.; Bottomley, L. A.; Collard, D. M. Org.

Lett. 2002, 4, 3195−3198.

(22) Cramer, C. J. Essentials of Computational Chemistry: Theories and

Models, 2nd ed.; Wiley: Chichester, West Sussex, England; Hoboken,

NJ, 2004.

thank David Liu for help with aquiring DPV data. D.G.P., F.F.,

J.R.R., and K.S.S. gratefully acknowledge financial support from

the Department of Energy (Grant Number DE-FG-02-

96ER14617) for the synthesis and experimental character-

ization effort. S.H. would like to acknowledge the Camille and

Henry Dreyfus Foundation, and Y.-y.O. gratefully acknowl-

edges the National Science Foundation (NSF OCI 1102418)

for funding the computational effort.

(23) Audebert, P.; Sadki, S.; Miomandre, F.; Clavier, G.; Vernieres,

M. C.; Saoud, M.; Hapiot, P. New J. Chem. 2004, 28, 387−392.

(24) Zhang, G.; Musgrave, C. B. J. Phys. Chem. A 2007, 111, 1554−

1561.

REFERENCES

■

(1) (a) Nazeeruddin, M. K.; Kay, A.; Rodicio, I.; Humphry-Baker, R.;

Muller, E.; Liska, P.; Vlachopoulos, N.; Gratzel, M. J. Am. Chem. Soc.

̈

1993, 115, 6382−6390. (b) Nazeeruddin, M. K.; Klein, C.; Liska, P.;

(25) Price, S. C.; Stuart, A. C.; Yang, L.; Zhou, H.; You, W. J. Am.

Chem. Soc. 2011, 133, 8057−8057.

(26) Fraind, A. M.; Tovar, J. D. J. Phys. Chem. B 2010, 114, 3104−

3116.

Gratzel, M. Coord. Chem. Rev. 2005, 249, 1460−1467. (c) Nazeeruddin,

̈

M. K.; Pec

́

hy, P.; Renouard, T.; Zyss, J.; Humphry-Baker, R.; Comte,

P.; Liska, P.; Cevey, L.; Costa, E.; Shklover, V.; Spicca, L.; Deacon, G.

(27) Pham, C. V.; Burkhardt, A.; Shabana, R.; Cunningham, D. D.;

Mark, H. B.; Zimmer, H. Phosphorus, Sulfur Silicon Relat. Elem. 1989,

46, 153−168.

B.; Bignozzi, C. A.; Gratzel, M. J. Am. Chem. Soc. 2001, 123, 1613−

̈

1624.

(2) (a) Amb, C. M.; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C.

E.; So, F.; Reynolds, J. R. J. Am. Chem. Soc. 2011, 133, 10062−10065.

(b) Chen, H. Y.; Hou, J. H.; Zhang, S. Q.; Liang, Y. Y.; Yang, G. W.;

Yang, Y.; Yu, L. P.; Wu, Y.; Li, G. Nat. Photonics 2009, 3, 649−653.

(c) Wang, E.; Ma, Z.; Zhang, Z.; Vandewal, K.; Henriksson, P.;

(28) Rurack, K.; Spieles, M. Anal. Chem. 2011, 83, 1232−1242.

(29) Guldi, D. M.; Luo, C. P.; Prato, M.; Dietel, E.; Hirsch, A. Chem.

Commun. 2000, 5, 373−374.

(30) Jayakannan, M.; Van Hal, P. A.; Janssen, R. A. J. J. Polym. Sci.,

Part A: Polym. Chem. 2002, 40, 251−261.

Inganas, O.; Zhang, F.; Andersson, M. R. J. Am. Chem. Soc. 2011, 133,

̈

(31) Kono, T.; Kumaki, D.; Nishida, J.; Tokito, S.; Yamashita, Y.

Chem. Commun. 2010, 46, 3265−3267.

14244−14247.

(3) Havinga, E. E.; Tenhoeve, W.; Wynberg, H. Synth. Met. 1993, 55,

299−306.

(32) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:

Fundamentals and Applications, 2nd ed.; Wiley: New York, 2001; p

xxi, 833 p.

(4) Roncali, J. Macromol. Rapid Commun. 2007, 28, 1761−1775.

(5) Mishra, A.; Fischer, M. K. R.; Bauerle, P. Angew. Chem., Int. Ed.

2009, 48, 2474−2499.

(33) (a) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan,

G. C. Adv. Mater. 2011, 23, 2367−2371. (b) Thompson, B. Variable

Band Gap Poly(3,4-alkylenedioxythiophene)-Based Polymers for

Photovoltaic and Electrochromic Applications. Ph.D. Dissertation,

University of Florida, Gainesvlle, FL, 2005.

(34) Casado, J.; Zgierski, M. Z.; Ewbank, P. C.; Burand, M. W.;

Janzen, D. E.; Mann, K. R.; Pappenfus, T. M.; Berlin, A.; Perez-

Inestrosa, E.; Ortiz, R. P.; Navarrete, J. T. L. J. Am. Chem. Soc. 2006,

128, 10134−10144.

(35) Peart, P. A.; Tovar, J. D. Org. Lett. 2007, 9, 3041−3044.

(36) Kitamura, C.; Tanaka, S.; Yamashita, Y. Chem. Mater. 1996, 8,

570−578.

(37) Akhtaruzzaman, M.; Kamata, N.; Nishida, J.; Ando, S.; Tada, H.;

Tomura, M.; Yamashita, Y. Chem. Commun. 2005, 25, 3183−3185.

(38) Akhtaruzzaman, M.; Tomura, M.; Nishida, J.; Yamashita, Y. J.

Org. Chem. 2004, 69, 2953−2958.

(39) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,

A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1−S19.

(40) Ono, K.; Tanaka, S.; Yamashita, Y. Angew. Chem., Int. Ed. Engl.

1994, 33, 1977−1979.

(41) (a) Wang, M. F.; Sun, Y. M.; Tong, M. H.; Chesnut, E. S.; Seo,

J. H.; Kumar, R.; Wudl, F. J. Polym. Sci., Part A: Polym. Chem. 2011, 49,

441−451. (b) Wang, M. F.; Wudl, F. J. Mater. Chem. 2010, 20, 5659−

5663.

(6) Patel, D. G.; Bastianon, N. M.; Tongwa, P.; Leger, J. M.;

Timofeeva, T. V.; Bartholomew, G. P. J. Mater. Chem. 2011, 21, 4242−

4250.

(7) Kira, A.; Matsubara, Y.; Iijima, H.; Umeyama, T.; Matano, Y.; Ito,

S.; Niemi, M.; Tkachenko, N.; Lemmetyinen, H.; Imahori, H. J. Phys.

Chem. C 2010, 114, 11293−11304.

(8) Velusamy, M.; Thomas, K. R. J.; Lin, J. T.; Hsu, Y.-C.; Ho, K.-C.

Org. Lett. 2005, 7, 1899−1902.

(9) (a) Balan, A.; Baran, D.; Gunbas, G.; Durmus, A.; Ozyurt, F.;

Toppare, L. Chem. Commun. 2009, 44, 6768−6770. (b) Balan, A.;

Gunbas, G.; Durmus, A.; Toppare, L. Chem. Mater. 2008, 20, 7510−

7513. (c) Baran, D.; Balan, A.; Celebi, S.; Esteban, B. M.; Neugebauer,

H.; Sariciftci, N. S.; Toppare, L. Chem. Mater. 2010, 22, 2978−2987.

(d) Fang, Q.; Tanimoto, A.; Yamamoto, T. Synth. Met. 2005, 150, 73−

78. (e) Peng, B.; Najari, A.; Liu, B.; Berrouard, P.; Gendron, D.; He,

Y.; Zhou, K.; Leclerc, M.; Zou, Y. Macromol. Chem. Phys. 2010, 211,

2026−2033. (f) Tanimoto, A.; Yamamoto, T. Adv. Synth. Catal. 2004,

346, 1818−1823. (g) Tanimoto, A.; Yamamoto, T. Macromolecules

2006, 39, 3546−3552.

(10) Tam, T. L.; Li, H.; Lam, Y. M.; Mhaisalkar, S. G.; Grimsdale, A.

C. Org. Lett. 2011, 13, 4612−4615.

(11) Cozzolino, A. F.; Gruhn, N. E.; Lichtenberger, D. L.; Vargas-

Baca, I. Inorg. Chem. 2008, 47, 6220−6226.

(42) Salzner, U.; Karalti, O.; Durdagi, S. J. Mol. Model. 2006, 12,

(12) Akbasoglu, N.; Balan, A.; Baran, D.; Cirpan, A.; Toppare, L. J.

Polym. Sci., Part A: Polym. Chem. 2010, 48, 5603−5610.

(13) Liu, L.; Hong, D.-J.; Lee, M. Langmuir 2009, 25, 5061−5067.

(14) Liu, B. K.; Wu, Q.; Qian, X. Q.; Lv, D. S.; Lin, X. F. Synthesis

2007, 17, 2653−2659.

687−701.

(43) Leusser, D.; Henn, J.; Kocher, N.; Engels, B.; Stalke, D. J. Am.

Chem. Soc. 2004, 126, 1781−1793.

(44) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,

899−926.

(15) Katritzky, A. R.; Pastor, A.; Voronkov, M.; Tymoshenko, D. J.

Comb. Chem. 2001, 3, 167−170.

(45) Walters, K. A.; Ley, K. D.; Cavalaheiro, C. S. P.; Miller, S. E.;

Gosztola, D.; Wasielewski, M. R.; Bussandri, A. P.; van Willigen, H.;

Schanze, K. S. J. Am. Chem. Soc. 2001, 123, 8329−8342.

(46) (a) Caspar, J. V.; Meyer, T. J. J. Phys. Chem. 1983, 87, 952−957.

(b) Henry, B. R.; Siebrand, W. In Organic Molecular Photophysics;

Birks, J. B., Ed.; J. Wiley: New York, 1973.

(16) Seitz, D. E.; Lee, S. H.; Hanson, R. N.; Bottaro, J. C. Synth.

Commun. 1983, 13, 121−128.

(17) Karikomi, M.; Kitamura, C.; Tanaka, S.; Yamashita, Y. J. Am.

Chem. Soc. 1995, 117, 6791−6792.

(18) Steckler, T. T.; Abboud, K. A.; Craps, M.; Rinzler, A. G.;

Reynolds, J. R. Chem. Commun. 2007, 46, 4904−4906.

(47) (a) de Melo, J. S.; Silva, L. M.; Arnaut, L. G.; Becker, R. S. J.

Chem. Phys. 1999, 111, 5427−5433. (b) Zhou, Z.; Corbitt, T. S.;

2611

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Journal of the American Chemical Society

Article

Parthasarathy, A.; Tang, Y.; Ista, L. F.; Schanze, K. S.; Whitten, D. G. J.

Phys. Chem. Lett. 2010, 1, 3207−3212.

(48) Karsten, B. P.; Bijleveld, J. C.; Viani, L.; Cornil, J.; Gierschner, J.;

Janssen, R. A. J. J. Mater. Chem. 2009, 19, 5343−5350.

(49) Dempster, D. N.; Thompson, G. F.; Morrow, T.; Rankin, R. J.

Chem. Soc., Faraday Trans. 1972, 68, 1479−1496.

(50) Wilkinson, F.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref.

Data 1993, 22, 113−262.

(51) Burrows, H. D.; de Melo, J. S.; Serpa, C.; Arnaut, L. G.;

Monkman, A. P.; Hamblett, I.; Navaratnam, S. J. Chem. Phys. 2001,

115, 9601−9606.

2612

dx.doi.org/10.1021/ja207978v | J. Am. Chem.Soc. 2012, 134, 2599−2612

Article Doi

It takes more than an imine: The role of the central atom on the electron-accepting ability of benzotriazole and benzothiadiazole oligomers

DOI: 10.1021/ja207978v

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja207978v

Full text of DOI:10.1021/ja207978v

Products guided by the article

R&D Labs maybe for 15155-41-6

Relevant to this article

Hot Product