A molecular breakwater-like tetrapod for organic solar cells

View Article Online

View Journal

Journal of

Materials Chemistry A

Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: J. Yang, W. He, K.

Denman, Y. Jiang and Y. Qin, J. Mater. Chem. A, 2014, DOI: 10.1039/C4TA05405A.

This is an Accepted Manuscript, which has been through the

Royal Society of Chemistry peer review process and has been

accepted for publication.

Accepted Manuscripts are published online shortly after

acceptance, before technical editing, formatting and proof reading.

Using this free service, authors can make their results available

to the community, in citable form, before we publish the edited

article. We will replace this Accepted Manuscript with the edited

and formatted Advance Article as soon as it is available.

You can find more information about Accepted Manuscripts in the

Information for Authors .

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s

standard T erms & Conditions and the Ethical guidelines still

apply. In no event shall the Royal Society of Chemistry be held

responsible for any errors or omissions in this Accepted Manuscript

or any consequences arising from the use of any information it

contains.

www.rsc.org/materialsA

Page 1 of 13

Journal of Materials Chemistry A

View Article Online

DOI: 10.1039/C4TA05405A

Journal Name

RSCPublishing

ARTICLE

A Molecular Breakwater-Like Tetrapod for Organic

Solar Cells

Cite this: DOI: 10.1039/x0xx00000x

Jianzhong Yang,^aWenhan He,^aKimberly Denman,^bYingꢀBing Jiang^cand Yang

Qin*^a

Received 00th January 2012,

Accepted 00th January 2012

We report the synthesis and characterization of a tetrapodal breakwaterꢀlike small molecule,

SO, containing a tetraphenylsilane core and four cyanoester functionalized terthiophene arms.

SO possesses a deep lying HOMO energy level of −5.2 eV and a narrow bandgap of 1.9 eV.

Absorption, Xꢀray scattering and differential scanning calorimetry (DSC) measurements

indicate crystalline nature of this compound but very slow crystallization kinetics. Solar cells

employing SO and phenylꢀC₆₁ꢀbutyric acid methyl ester (PCBM) were fabricated and

evaluated. Relatively low performance was obtained mainly due to the lack of optimal phase

separation under various processing conditions including thermal annealing, slowꢀcooling and

solvent annealing. Addition of poly(thienylene vinylene) (PTV), poly(3ꢀhexylthiophene)

(P3HT) and a platinumꢀcontaining low bandgap conjugated polymer PtꢀBODIPY, into the

SO/PCBM blend was found to induce device favorable phase separation and the polymers were

found to act as the primary hole conductor. Such ternary blend devices showed cooperatively

improved performances over binary devices employing either SO or the individual conjugated

polymer alone.

DOI: 10.1039/x0xx00000x

www.rsc.org/

charge mobilities, and at the same time have discrete and

reproducible molecular structures.^14ꢀ22These features have

1. Introduction

Organic solar cells (OSCs) are considered a promising lowꢀcost

renewable energy source.^1ꢀ2Research efforts in OSCs have

been exclusively focused on conjugated polymers (CPs) owing

to the device favorable processability and thin film forming

ability, as well as the versatility in structure/property variations

through wellꢀestablished chemical transformations.^3ꢀ5As a

result, power conversion efficiencies (PCEs) of polymer solar

cell (PSC) devices have been steadily improved to approach

10% in recent years.^6ꢀ13CPs are typically synthesized through

crossꢀcoupling reactions in stepꢀgrowth fashions that

unavoidably generate materials with large distributions of

molecular weights and structural defects are frequently

encountered. Control over this type of polymerization is poor,

which commonly leads to batchꢀtoꢀbatch and labꢀtoꢀlab

variations in polymer structures and properties. Most CPs

applied in efficient PSCs have been found to be amorphous and

thus possess relatively low charge mobilities. These aspects can

potentially limit materials mass production and impede further

device improvement. On the other hand, conjugated small

molecules can be highly crystalline and thus have superior

attracted increasing attention and bulk heterojunction (BHJ)

OSC devices employing conjugated small molecules and

fullerene derivatives have been constantly improved to rival

their CP counterparts, showing great promises in solar cell

research.^23ꢀ30

Most small molecules applied in solar cells have linear

structures containing multiple aromatic groups connected in

series. These molecules are typically highly crystalline and

conductive along the πꢀstacking direction. However, charge

migration along both long and short molecular axes are

relatively limited due to the oneꢀdimensional (1ꢀD) nature of

these molecules. Unfavorable film forming ability and grain

boundaries both originated from high crystallinity of linear

molecules are also detrimental to device performances. As a

result, significant attention has been paid to conjugated small

molecules having conjugation extended in three dimensions (3ꢀ

D).²²Such molecular design can increase absorption crossꢀ

sections and provide more extensively percolating pathways for

charge transport. Among the many 3ꢀD structures, breakwaterꢀ

like tetrapods are especially interesting owing to their unique

This journal is © The Royal Society of Chemistry 2013

J. Name., 2013, 00, 1-3 | 1

Journal of Materials Chemistry A

Page 2 of 13

^VJ^ieo^wu^Ar^rtn^ica^lel^ON^nlaⁱⁿm^ee

ARTICLE

DOI: 10.1039/C4TA05405A

ability to mutually interlock, which prevents dislodging and

provides high morphological stabilities. This concept has been

frequently applied in inorganic nanocrystal synthesis and the

size scales of resulting tetrapods are on the order of hundred

nm.^31ꢀ33On the other hand, organic molecular tetrapods are less

common in OSC research.

2. Results and Discussion

2.1 Synthesis and Characterization.

Detailed synthetic procedures for the tetrapodal molecule SO

and a model compound MO that represents a single arm of the

tetrapod, are shown in Scheme 1. Compound was prepared

,

1

Roncali et al. reported the synthesis of two tetrapodal

molecules, each containing a silicon core and four terthiophene

arms bearing alkyl and thioalkyl side chains, respectively.³⁴The

overall solar cell performances were significantly limited by the

large bandgaps of these tetrapods, but still outꢀperformed the

devices employing corresponding linear counterparts having

structures of one of the four arms. Köse et al. recently reported

low bandgap tetrapodal molecules composed of a silicon core

and four arms composed of thiophene and benzothiadiazole

units.³⁵Favorable impact of high dimensionality of these

molecules on charge mobilities in disordered media was

discovered. In both of these examples, the molecular tetrapods

all contain silicon atoms as the cores that are connected to four

thiophene rings. Owing to the electron rich character and

relatively small sizes of thiophene rings, the silicon centers are

more exposed and the siliconꢀthiophene bonds are relatively

weak. This instability can potentially complicate compound

synthesis and characterization, as well as reduce device

operation lifetimes.³⁶

from commercial 1,4ꢀdibromobenzene through lithium halogen

exchange followed by reaction with 0.20 equivalents of SiCl₄.

Compound

1 can be conveniently applied as a common core for

grafting with different arms toward 3ꢀD tetrapodal molecules.

After Stille coupling reaction with 10 followed by acetal

deprotection, the tetraꢀaldehyde compound 11 was obtained.

The aldehyde groups can be transformed to several strongly

electron withdrawing substituents, e.g., dicyanovinyl and

cyanoester groups. However, 11 was found to have very limited

solubility in common OSC processing solvents including

chlorobenzene and dichlorobenzene. We thus chose

nꢀoctyl

cyanoacetate (12) to impart the electron deficient moiety as

well as sufficient solubility. Indeed, after the simple

Knoevenegalꢀtype condensation reaction, compound SO was

obtained in high yields and has good solubility in a wide range

of organic solvents including CHCl₃, THF and chlorobenzene.

1

All compounds are fully characterized by H and ¹³C NMR

spectroscopy, which agree well with proposed structures

(Electronic Supplementary Information, ESI^†). High resolution

mass spectrometry (HRꢀMS) was attempted to confirm the

tetraꢀarm structures of the newly synthesized compounds.

Unfortunately, SO could not be ionized under current

experimental conditions and no meaningful mass signals could

be observed. Instead, HRꢀMS was performed on the precursor

11. The measured molar masses (1768.3419 [M⁺] and

1791.3317 [M+Na⁺]) match perfectly with the calculated values

(1768.3414 [M⁺] and 1791.3312 [M+Na⁺]), confirming the

On the other hand, tetraphenylsilane is a commercially

available compound and known to be robust under various

reaction conditions, and is thus a better starting point to

construct the molecular tetrapods. Herein, we report the

synthesis, characterization and application in OSCs of a stable

low bandgap breakwaterꢀlike tetrapod containing

a

tetraphenylsilane core and four cyanoester functionalized

terthiophene arms.

proposed tetrapodal structures of 11 and consequently of SO

.

Scheme 1. Synthesis of SO and MO

.

2 | J. Name., 2012, 00, 1-3

This journal is © The Royal Society of Chemistry 2012

Page 3 of 13

Journal Name

Journal of Materials Chemistry A

^{View Artic}A^leR^OnT^liIⁿC^eLE

DOI: 10.1039/C4TA05405A

fluorescence peaked at ca. 650 nm can be observed. From the

absorption edges, the optical bandgaps of SO and MO are

estimated to be both ca. 2.2 eV in solutions and 1.9 eV and 2.0

eV in thin films, respectively.

In order to quantify the frontier energy levels and bandgaps,

cyclic voltammetry (CV) measurements were performed on SO

and MO in dichloromethane solutions (1 mM). A glassy carbon

working electrode, a Ag/AgCl reference electrode and a Pt wire

counter

electrode

were

used.

Tetrabutylammonium

hexafluorophosphate (Bu₄NPF₆, 0.1 M) was used as the

supporting electrolytes. The recorded CV curves were

externally referenced to ferrocene/ferrocenium (Fc/Fc⁺) redox

couple (4.80 eV below vacuum). Therefore, the HOMO and

LUMO energy levels can be estimated using the empirical

onset

formula E_HOMO= − (E_ox^onset+ 4.80) eV and E_LUMO= − (E_red

+ 4.80) eV, respectively. As shown in Figure 2, both SO and

MO displayed nearly identical redox behaviors. Two quasiꢀ

reversible oxidation peaks and one irreversible reduction peak,

at onsets of ca. 0.4 V, 1.0 V and −1.5 V, respectively, were

observed. As a result, the HOMO and LUMO levels of SO and

MO were estimated to be –5.2 eV and –3.3 eV. This leads to an

electrochemical bandgap of ca. 1.90 eV for both SO and MO

,

agreeing well with the results of optical measurements.

Figure 1. Normalized UVꢀvis absorption and emission spectra of (A)

SO and (B) MO in chlorobenzene solutions (10⁻⁵M, solid lines) and as

thin films (dashed lines).

The electronic properties of SO and MO were investigated by

UVꢀvis absorption and fluorescence spectroscopies in both

dilute solutions and as thin films. As shown in Figure 1, both

SO and MO in chlorobenzene solutions (10⁻⁵M) display nearly

identical structureless absorption profiles with λ_max’s at ca. 470

nm. The solution fluorescence spectra of both compounds are

also indistinguishable with λ_em’s at ca. 595 nm. Such similarity

in absorption and emission spectra indicates that there is no

electronic communication among the four conjugated arms of

SO when intermolecular interactions are negligible in dilute

solutions. This is understandable since these four arms are

stretched out away from one another in a tetrahedral geometry,

which are connected through a nonꢀconjugated silicon core.

However, owing to the differences in molecular geometries, SO

and MO show very different spectra in thin films. The asꢀcast

thin film of SO displays a λ_maxat ca. 570 nm, which is redꢀ

shifted from that of the solution profile by 100 nm. This redꢀ

shift in absorption is commonly observed in conjugated systems

due to structural planarization and intermolecular interactions

in the solid state. The fluorescence of SO films is very weak,

having a λ_emat ca. 625 nm when excited at the λ_max, which

gives a relatively small Stoke’s shift of 55 nm. On the contrary,

the main absorption peak of MO is slightly blueꢀshifted to ca.

445 nm in thin films, indicating Hꢀtype aggregation of the

molecules.³⁷A low energy shoulder peak at ca. 530 nm is also

observed, which is likely originated from new electronic

species resulted from aggregation. Emission of MO films is

expectedly quenched to a large extent and only a weak

Figure 2. Cyclic voltammograms of MO and SO in CH₂Cl₂solutions

(1 mM) containing Bu₄NPF₆as the supporting electrolytes (0.1 M). The

voltages are referenced externally to ferrocene (Fc) redox couple. Scan

rate: 100 mV/s.

Thermal properties of SO and MO were studied by using

differential scanning calorimatry (DSC) measurements and the

results are displayed in Figure 3. At a typical scanning rate of

10 °C/min, SO shows an exothermic crystallization transition

peaked at 115 °C and two closely spaced melting transitions at

180 °C and 188 °C. No crystallization event is observed in the

cooling curve. At a slower scanning rate of 1 °C/min, the

crystallization transition is only observed during the cooling

event at ca. 105 °C. Such behavior indicates slow

crystallization kinetics of the compound. Similar behaviors

were previously reported from a linear conjugated small

molecule having two intentionally designed structural twists.³⁸

We originally thought that the tetrapodal structure of SO, which

can be considered as having four structural twists in the

molecule, is the leading cause for the slow crystallization

This journal is © The Royal Society of Chemistry 2012

J. Name., 2012, 00, 1-3 | 3

Journal of Materials Chemistry A

Page 4 of 13

^VJ^ieo^wu^Ar^rtn^ica^lel^ON^nlaⁱⁿm^ee

ARTICLE

DOI: 10.1039/C4TA05405A

behavior. However, the same trend is observed for MO. At a We have so far not been able to obtain high quality single

scanning rate of 10 °C/min, MO displays both a crystallization crystals for SO in order for detailed Xꢀray analysis. DSC

(35 °C) and a melting transitions (68 °C) in the heating event studies indicate certain crystallinity of the compound, which is

while the crystallization transition (24 °C) is only observed further confirmed by thin film wideꢀangle Xꢀray scattering

upon cooling when the scanning rate is reduced to 1 °C/min. experiments as shown in Figure S2†. Multiple scattering peaks

This indicates that the slow crystallization behavior of SO may are observed at 2θ values of ca. 4.0°, 5.7°, 8.1° and 12.3°,

be intrinsically resulted from the structure of each of its arms as which correspond to

demonstrated in the case of MO

dꢀspacings of ca. 2.2, 1.5, 1.1 and 0.7 nm,

.

respectively. Assignments of these scattering peaks are still not

certain at present. The SO thin films were further studied by

transmission electron microscopy (TEM) and selected area

electron diffraction (SAED) measurements, and the results are

summarized in Figure S3†. Micronꢀsized ellipsoidal aggregates

are randomly distributed throughout the film as seen in the

TEM image. From the SAED image, only a few scattering rings

can be observed and azimuthal integration of the major

scattering ring gives a dꢀspacing of ca. 1.3 nm. The SAED

pattern indicates that although there exists periodic structures in

the SO films, the compound itself does not form conventional

crystals ordered in three dimensions.

2.2 Binary Blend Devices.

Solar cell devices were fabricated using conventional

structures: ITO glass/MoO₃(10 nm)/active layer (100 nm)/Al

(100 nm). Mixtures of SO or MO and phenylꢀC₆₁ꢀbutyric acid

methyl ester (PCBM, American Dye Source, Inc.) at various

weight ratios in chlorobenzene were spun cast to form the

active layer. Thermal annealing at various temperatures was

employed to optimize the device performances. The best PCE

s

were found in asꢀcast devices employing SO/PCBM at a weight

ratio of 1/3, and MO/PCBM at a weight ratio of 1/2. Devices

employing SO generally outꢀperformed those using MO. The

current density−voltage (IꢀV) curves of the best performing

devices are shown in Figure S4†. Both SO and MO devices

show relatively high open circuit voltage (V_OC) values at 0.84 V

and 0.74 V, respectively, which is consistent with the deep

lying HOMO levels of these molecules. However both devices

suffer greatly from low short circuit current (J_SC) and fill factor

Figure 3. Differential scanning calorimetry (DSC) histograms of SO at

a scanning rate of 10 °C/min (solid line) and at 1 °C/min (dashed line);

and of MO at 10 °C/min (dotted line) and at 1 °C/min (dash dotted

line). Second heating (lower segments) and cooling (upper segments)

curves are shown.

In order to gain a deeper insight on the structural origin of these

thermal behaviors, we performed density functional theory

(DFT) calculations (B3LYP/6ꢀ31G*) on MO and the optimized

geometries are shown in Figure S1†. The calculated minimum

energy structure of MO is not completely planar. Dihedral

angles of 26° and 15° are found between the phenyl and

alkylthienyl groups and between the alkylthienyl and the

adjacent thienyl groups, respectively. These structural twists are

possibly responsible for inefficient molecular packing and thus

slow crystallization kinetics. Also seen from the calculation

results, the HOMO orbital of MO is delocalized throughout the

entire conjugation and the LUMO orbital is positioned toward

the cyanoester side. Electronic transitions from HOMO to

LUMO thus possess charge transfer characteristics as expected.

The HOMO level is calculated to be –5.3 eV, matching that

from CV measurements, while the LUMO is over estimated to

be −2.7 eV.

Hole mobilities in SO and MO films were estimated using

space charge limited current (SCLC) method³⁹in hole only

devices having ITO/MoO₃(10 nm)/organic/MoO₃(10 nm)/Al

(100 nm) geometries. The results are averaged from three

devices with different organic layer thickness for each

compound. The hole mobility of SO film is calculated to be ca.

1.8×10⁻⁴cm²/Vs, which is more than 100 times higher than that

found for the MO film at 1.1×10⁻⁶cm²/Vs. The tetrahedral

shape of SO likely leads to enhanced percolating pathways and

thus improved charge mobilities, which is consistent with the

previous report.³⁵

(

FF) values. For instance, the MO device gives a J_SCof 0.64

mA/cm²and a FF of 30%, leading to PCE of 0.14%. While the

SO device displays a J_SCof 1.01 mA/cm², a FF of 26% and the

PCE of 0.22%. Steep increases in current densities at reverse

bias are observed for both devices under light, which indicates

significant charge recombination at short circuit conditions.

Similar behaviors were observed by Köse et al. for their

tetrapodal molecules, which they ascribed to inferior blend

morphologies.³⁵We thus studied the thin film morphologies of

SO, MO and corresponding PCBM blends at optimal weight

ratios for device operation by using optical microscopy and the

photographs are included in Figure 4. Micrographs of asꢀcast

and annealed SO and MO thin films under crossꢀpolarized light

are included in Figure S5.

The asꢀcast SO thin films have smooth morphologies that are

free of any visible aggregates or crystallites (Figure 4A).

Annealing at 150 °C for 10 min leads to crystallization of SO

,

resulting in heavily textured morphology and dark crystallites

as seen in Figure 4B. This observation is consistent with those

from DSC studies and confirmed by crossꢀpolarized light

microscopy. As shown in Figure S5A, no features can be

observed in the asꢀcast SO thin films while annealing leads to

heavily textured morphologies with relatively small feature

sizes (Figure S5B). However, thin films containing SO/PCBM

mixture at a 1/3 weight ratio do not display any phase

separation (Figure 4C) even after annealing at 150 °C for 10

min (Figure 4D). It is likely that PCBM molecules are

4 | J. Name., 2012, 00, 1-3

This journal is © The Royal Society of Chemistry 2012

Page 5 of 13

Journal Name

Journal of Materials Chemistry A

^{View Artic}A^leR^OnT^liIⁿC^eLE

DOI: 10.1039/C4TA05405A

intercalated among the arms of SO, which effectively prevents

crystallization of both SO and PCBM, and thus suppresses any

appreciable phase separation within the blends. This lack of

phase separation into pure domains of electron donors and

acceptors can significantly limit charge transport and lead to

large rates of charge recombination. On the other hand, MO

thin films show slightly different behaviors. Clear and dense

crystallites are observed in the asꢀcast films of MO (Figure 4E),

which is not surprising since MO has a crystallization transition

temperature around 25 °C. Annealing at 150 °C leads to still

textured morphology with apparently less crystallites (Figure

4F). This trend is also confirmed by crossꢀpolarized light

microscopy. Crystals of MO can be cleared observed under

crossꢀpolarized light in asꢀcast films (Figure S5C), which

become larger and less densely packed after thermal annealing

(Figure S5D). Asꢀcast thin films of MO/PCBM blends show

overall smooth morphologies having a few sparsely located

crystallites (Figure 4G) that become more populated in

annealed films (Figure 4H). We suspect that these crystallites

are those of PCBM molecules. In short summary, addition of

PCBM seems to suppress crystallization of both MO and SO

,

while SO has a more pronounced effects on suppression of

PCBM aggregation than MO does, owing to its tetapodal

structure that can potential interact with PCBM molecules more

strongly. On the other hand, because of the breakꢀwater like

structure of SO, the molecularly mixed state seems to result in

thermally robust morphologies, which is preferred for OSC

operations.

Since DSC measurements showed slow crystallization

behaviors of the SO compound, we attempted slowꢀcooling

experiments on the SO/PCBM (1/3) devices in hope to induce

phase separation of these two components and thus to improve

device efficiencies. The typical procedure is to anneal the

devices on a hotplate with preset temperatures for 10 min and

the hotplate is turned off without removing the devices. The

hotplate temperature then slowly drops to r.t. within 1.5 to 2 h

and the devices are then tested. Four different preset hotplate

temperatures at 100 °C, 150 °C, 200 °C and 250 °C were

selected. However, the devices slowꢀcooled from 100 °C and

150 °C showed similar performances as the asꢀcast devices,

Figure 4. Optical micrographs (400× magnificantion) of thin films of

SO, MO and corresponding PCBM blends under different annealing

conditions. Obvious artifacts are circled out in white. Scale bars in all:

20 ꢁm.

while the devices slowꢀcooled from 200 °C and 250 °C Alternatively, solvent annealing has been shown as an effective

displayed reduced PCE to ca. 0.04% and no photovoltaic method for inducing phase separation in blend films.^40ꢀ41We

behaviors, respectively. Optical micrographs of these slowꢀ thus tested the SO/PCBM (1/3) devices under solvent annealing

cooled devices are shown in Figure S6†. When slowꢀcooled conditions by placing the devices in a sealed container saturated

from 100 °C and 150 °C, identically smooth, aggregation free with chlorobenzene vapor for up to 20h. Devices were then

morphologies are observed (Figure S6A and S6B), indicating tested after predetermined annealing times and subjected to

the ineffectiveness for inducing appreciable phase separation optical microscopy measurements as summarized in Figure 5.

between SO and PCBM under these conditions. When the No performance differences were observed for devices

temperature was raised to 150 °C, dark needle like crystals of annealed up to 1 h, although spheruliteꢀlike crystals started to

tens of microns in lengths emerged (Figure S6C), which appear and became denser and bigger with time (Figures 5A to

became significantly bigger when slowꢀcooled from 250 °C 5D). The devices annealed for 1.5 h, 3 h and 6 h all showed

(Figure S6D). Noticeably, there appear to be deꢀwetting slightly improved PCE values to ca. 0.25ꢀ0.27%, which come

processes taking place around these large crystallites, which from an improvement in FF to ca. 44% accompanied by a

possibly account for the reduced and eventually diminished decrease in J_SCto ca. 0.62 mA/cm². However, devices annealed

device performances.

for 20 h lost almost all of the photovoltaic effects and no diode

behaviors could be observed. As seen in Figures 5E to 5G, at

longer annealing times, the spheruliteꢀlike crystals seem to

grow and merge into large platelets having straight boundaries.

These platelets have lighter smoother central regions and darker

needleꢀlike peripherals, accompanied by randomly dispersed

black spheres. These two regions eventually became clearly

differentiated after annealing for 20 h (Figure 5H). We suspect

that the darker crystals are those of PCBM and the lighter

regions consist of SO or SO/PCBM complexes. Such solvent

This journal is © The Royal Society of Chemistry 2012

J. Name., 2012, 00, 1-3 | 5

Journal of Materials Chemistry A

Page 6 of 13

^VJ^ieo^wu^Ar^rtn^ica^lel^ON^nlaⁱⁿm^ee

ARTICLE

DOI: 10.1039/C4TA05405A

annealing induced crystallization, phase separation and which can potentially lead to improved photocurrents. Thus we

eventual macroꢀphase separation likely explain the initial have fabricated OSC devices using SO/PTV/PCBM ternary

device improvement and final breakdown as observed. The blends at different weight ratios. An inꢀhouse made poly(3ꢀ

exact identity of the needleꢀlike crystals and mechanisms of decylthienylene vinylene) (P3DTV, M_n= 21.3 kDa, PDI

such phase separation are currently under more detailed 2.1)⁴⁶was used in the studies, as shown in Scheme 2.

investigation.

=

Scheme 2. Structures of conjugated polymers applied in current

studies.

The binary P3DTV/PCBM devices were first fabricated and

optimized, from which a weight ratio of 1/1 and a thermal

annealing temperature of 80 °C for 10 min were found to be

optimal, leading to a PCE of 0.49%. This relatively low

efficiency is comparable to previously reported examples and is

possibly caused by the short lifetimes of excitons in PTVs.^47ꢀ48

Thus, in the ternary blends, we kept the weight ratios of

SO/PCBM and P3DTV/PCBM to be consistent at 1/3 and 1/1,

respectively. Thermal annealing was found to slightly enhance

ternary device performance and the optimal temperature was

found to be 80 °C, beyond which device deterioration occurred.

Table 1 summarizes detailed device parameters involving

binary and ternary blends at different weight ratios and the

corresponding IꢀV curves are included in Figure 6.

Table 1. Binary and Ternary Device Performances.^a

J_SC

SO/P3DTV/PCBM^b

V

_OC(V)

FF (%)

PCE (%)

(mA/cm²)

1.01

A. 2.5 / 0.0 / 7.5^c

B. 2.5 / 0.5 / 8.0

C. 2.5 / 1.5 / 9.0

D. 2.5 / 2.5 /10.0

E. 2.0 / 2.5 / 8.5

F. 1.5 / 2.5 / 7.0

G. 1.0 / 2.5 / 5.5

H. 0.5 / 2.5 / 4.0

I. 0.0 / 2.5 / 2.5

0.84

0.55

0.52

0.51

0.48

0.45

26

37

52

54

55

52

49

0.22

0.44

1.00

1.22

1.45

1.54

1.19

0.87

0.49

2.17

3.67

4.34

5.18

5.52

4.27

3.48

2.22

Figure 5. Optical micrographs (100× magnificantion) of thin films of

SO and PCBM (1/3, wt./wt.) blends after solvent annealing using

chlorobenzene for various times. Scale bars in all: 100 ꢁm.

^aAll devices are thermally annealed at 80 °C for 10 min; results are reported

as averages of five individual cells. ^bAll ratios by weight. ^cAsꢀcast device.

2.3 Ternary Blend Devices.

As discussed above, SO molecules tend to molecularly mix

with PCBM molecules and thermal annealing is ineffective to

drive appreciable phase separation. Slowꢀcooling from high

temperatures and solvent annealing do induce macroscopic

phase separation but are not able to improve the device

performances significantly. We suspect that addition of a third

component, which has stronger interactions with either SO or

PCBM, can potentially break up the SO/PCBM interactions and

lead to one of the components to crystallize and phase separate

into domains better for charge separation and extraction. To test

this hypothesis, we chose poly(thienylene vinylene) (PTV) to

be the third component. PTVs are a wellꢀknown class of

conjugated polymers possessing narrow bandgaps and high

crystallinity.^42ꢀ45The πꢀπ interactions among aromatic rings

may lead to stronger SO/PTV interactions and the absorption

windows of SO and PTV are complementary to each other,

6 | J. Name., 2012, 00, 1-3

This journal is © The Royal Society of Chemistry 2012

Page 7 of 13

Journal Name

Journal of Materials Chemistry A

^{View Artic}A^leR^OnT^liIⁿC^eLE

DOI: 10.1039/C4TA05405A

Figure 6. Current densityꢀvoltage (IꢀV) curves of devices employing

SO, P3DTV and PCBM at various weight ratios under simulated white

light (100 mW/cm²).

Several trends are clearly observed going from binary blends to

ternary blends. The V_OCvalues of ternary devices are all

between those of the binary devices employing P3DTV and

SO. These values decrease with increasing P3DTV contents

and are all much closer to the side of P3DTV. For instance, in

devices B (Table 1), addition of only 20 wt.% P3DTV (relative

to SO) reduces the V_OCfrom 0.84 V in device A to 0.55 V. It is

known that the V_OCof a BHJ solar cell is closely related to the

energy difference between the HOMO level of electron donor

and the LUMO level of the electron acceptor.⁴⁹Owing to the

high lying HOMO level of P3DTV at ca. −4.9 eV (from CV

measurements), devices containing P3DTV are expected to

have reduced V_OCvalues. Interestingly, previous examples

showed close to linear relationships between V_OCand

composition changes.^50ꢀ51In our case, we plot the V_OCvalues

against P3DTV contents in the donor blends (Figure S7†) and

find that the relationship is far from linearity. All devices

containing P3DTV display comparable V_OCvalues that are

significantly smaller than that of the SO only device. This

implies that in our ternary devices, P3DTV is the primary hole

conducting material despite of its contents, which is likely

caused by high crystallinity and longꢀchain structures of the

polymer.

Figure 7. Tappingꢀmode atomic force microscopy (AFM) height

images (left column) and phase images (right column) of binary device

A (SO/PCBM 1/3; A and B), binary device I (P3DTV/PCBM 1/1; C

and D) and ternary device F (SO/P3DTV/PCBM 1.5/2.5/7.0; E and F).

All images are 2×2 ꢁm in size. (G) Transmission electron microscopy

(TEM) image of device F; scale bar: 50 nm.

Another obvious trend is the enhancement of J_scin the ternary

blend devices that all, except B, show higher J_scvalues than the

binary devices. Device D can be considered a linear

combination of the contents of binary devices A and I, but gives

a higher J_scof 4.34 mA/cm²than the sum of J_scvalues of

devices A and I (3.23 mA/cm²). This indicates that there are

cooperative effects on photocurrent generation by mixing SO

and P3DTV, which enhance charge extraction within the

ternary blends. The best device tested is F that displays the

highest J_scof 5.52 mA/cm²and a PCE of 1.54%, which is ca. 3

and 7 times higher than those of binary devices employing

P3DTV and SO alone, respectively.

Noticeably, all ternary devices show significantly higher FF

s

than the SO binary device, which can be related to improved

morphologies by adding P3DTV. Thus, we have studied the

best ternary device F by optical microscopy (Figure S8†),

atomic force microscopy (AFM) and transmission electron

This journal is © The Royal Society of Chemistry 2012

J. Name., 2012, 00, 1-3 | 7

Journal of Materials Chemistry A

Page 8 of 13

ARTICLE

^VJ^ieo^wu^Ar^rtn^ica^lel^ON^nlaⁱⁿm^ee

DOI: 10.1039/C4TA05405A

microscopy. The binary devices are also studied by AFM for

comparison and all AFM and TEM images are included in

Figure 7. As shown in Figure S8, the ternary film under both

asꢀcast and annealing at 80 °C conditions do not display any

appreciable aggregation or crystallization. However, when

annealed at 150 °C for 10 min, large needle like PCBM

crystallites are observed. This macroꢀphase separation indicates

the effectiveness of adding P3DTV on reducing the strong

SO/PCBM interactions. The AFM measurements probe the thin

film morphologies on the nanometer scale. As displayed in

Figure 7A and 7B, the SO/PCBM binary blends show very

smooth morphologies that lack appreciable phase separation,

confirming the intermixed nature of these blends. On the other

hand, much rougher topography is observed in the

P3DTV/PCBM binary blends and larger aggregates are clearly

present (Figure 7C and 7D). As a result, the SO/P3DTV/PCBM

ternary blends display morphologies somewhat in between

those of the binary blends (Figure 7E and 7F). The nanometer

scale phase separation observed in device F by AFM is further

confirmed by TEM. As shown in Figure 7G, interꢀpenetrating

brighter (SO and P3DTV rich) and darker (PCBM rich) phases

improved morphologies in this ternary blend over the well

studied P3HT/PCBM binary blends, which implies an effective

way to optimize conjugated polymer/fullerene blend

morphologies and is currently under more detailed studies. Not

surprisingly, the highly crystalline P3HT acts as the major hole

conductor in the ternary blends similar to the case for P3DTV,

since the addition of a minority amount of P3HT drastically

decreases the V_OCvalues. On the other hand in device O,

addition of 20 wt.% SO into the PtꢀBODIPY/PCBM binary

device also improves the performance, which is however

resulted from an increase in J_SC. Since PtꢀBODIPY has a lower

bandgap than SO, such increase in J_SCis likely a result of

complementary absorption. The relatively small FFs in all

ternary devices are supposedly explained by the amorphous

nature of both SO and PtꢀBODIPY, mixtures of which are thus

expected to result in hardly improved morphologies. Both

compounds have similar deep lying HOMO levels, leading to

high V_OCvalues in all ternary devices. Interestingly, devices O

are observed, corresponding to domain sizes on the order of 10 and P have V_OCvalues slightly higher than those of

nm. Such more pronounced phase separation on the nanometer

scale creates more pure donor and acceptor domains that lead to

better charge separation and collection, and enhanced

performance of the ternary devices over that of the binary

devices.

corresponding binary devices, the reasons for which are

currently under investigation.

3. Conclusions

In addition to P3DTV, which is a low bandgap, crystalline and

less efficient polymer, we have started investigating ternary

solar cells containing SO/PCBM and other types of conjugated

polymers including regioꢀregular poly(3ꢀhexylthiophene)

(P3HT) and a platinum containing polymer (PtꢀBODIPY) made

inꢀhouse,⁵²as shown in Scheme 2. Compared with P3DTV,

P3HT possesses similarly high crystallinity and comparable

HOMO energy level, but a slightly larger bandgap at ca. 1.9

eV, and is one of the most studied high performing conjugated

polymers in OSC research. On the other hand, the PtꢀBODIPY

polymer has a small bandgap at ca. 1.7 eV, a deep lying HOMO

level at ca. −5.3 eV and is an amorphous material. Table 2

summarizes performance parameters of ternary devices

employing these materials at various weight ratios.

We have successfully prepared a molecular tetrapod possessing

a tetraphenylsilance core. The molecule has a relatively small

bandgap and a deep lying HOMO energy level. The tetrapod

was also found to be very hard to crystallize, which led to

molecularly mixed blends with PCBM and poor solar cell

device performances. This inferior morphology could be

improved by adding a conjugated polymer that can induce

appreciable phase separation in the ternary blends and result in

much enhanced device efficiencies, the effects of which depend

on the conjugated polymers' bandgaps, HOMO levels and

crystallinity. Our findings can give useful insights on the

structureꢀproperty relationships of such 3ꢀD small molecules

and provide a unique way to control device morphologies by

using ternary blends employing both a 3ꢀD small molecule and

a linear conjugated polymer. We are currently investigating the

possibilities of increasing the crystallization kinetics of the

tetrapods and of controlling the phase separations in multiꢀ

component blend films by variations of the molecular structures

and processing conditions.

Table 2. Ternary Devices Involving P3HT and PtꢀBODIPY.^a

V_OC

(V)

0.57

0.59

0.62

0.71

J_SC

(mA/cm²)

10.70

10.31

4.23

FF

(%)

59

69

62

SO/P3HT/PCBM^b

PCE (%)

J. 0.0 / 5.0 / 5.0

K. 1.0 / 5.0 / 5.0

L. 2.5 / 2.5 / 7.5

3.63

4.22

1.63

0.83

4. Experimental Section

M. 4.0 / 1.0 /12.0

SO/PtꢀBODIPY/PCBM^c

N. 0.0 / 2.5 / 7.5

O. 0.5 / 2.5 / 7.5

P. 2.5 / 2.5 / 7.5

2.46

48

4.1 Materials and General Methods.

0.86

0.91

0.93

0.86

2.23

3.39

2.49

2.05

48

45

33

38

0.91

1.39

0.77

0.68

All reagents and solvents were used as received from Sigmaꢀ

Aldrich or Alfa Aesar unless otherwise noted. THF was

distilled from Na/benzophenone prior to use. 2ꢀ

Q. 2.5 / 0.5 / 7.5

a

Tributylstannylthiophene

(

3

),⁵³

2,2’ꢀbithiopheneꢀ5ꢀ

Results are reported as averages of five individual cells; All ratios by

b

c

weight. All devices are thermally annealed at 150 °C for 10 min. Asꢀcast

devices.

carbaldehyde

carbaldehyde (

(

4

),⁵⁴

5’ꢀbromoꢀ(2,2’ꢀbithiophene)ꢀ5ꢀ

5

),⁵⁴2ꢀbromoꢀ3ꢀhexylthiophene ( ),⁵⁵3ꢀhexylꢀ2ꢀ

6

56

trimethylstannylthiophene (

7

)

were prepared according to

Consistently, cooperative effects are observed in both types of

ternary devices, as both devices K and O outperform the

corresponding optimized binary devices. In device K, addition

of 20 wt.% SO increases the optimized P3HT/PCBM binary

device efficiency from 3.63% to 4.22%, which mainly comes

from the enhancement in FF up to 69%. This indicates

literature procedures. 300.13 MHz ¹H and 75.48 MHz ¹³C

NMR spectra were recorded on a Bruker Avance III Solution

300 spectrometer. All solution H and ¹³C NMR spectra were

1

referenced internally to solvent signals. Ultraviolet−visible

(UV−vis) absorption spectra were recorded on a Shimadzu UVꢀ

2401 PC spectrometer over a wavelength range of 240−900 nm.

8 | J. Name., 2012, 00, 1-3

This journal is © The Royal Society of Chemistry 2012

Page 9 of 13

Journal Name

Journal of Materials Chemistry A

^{View Artic}A^leR^OnT^liIⁿC^eLE

DOI: 10.1039/C4TA05405A

Fluorescence emission spectra were obtained using a Varian

Cary Eclipse fluorimeter. Timeꢀofꢀflight mass spectrometry

(TOF MS) was performed on a Waters/Micromass LCT

Premier system operating under atmospheric pressure

photoionization (APPI+) mode. Cyclic voltammetry was

4.3 Synthetic Details.

. 1,4ꢀDibromobenzene (5.00 g, 21.2 mmol) was weighed into a

1

dry 100 mL Schlenk flask and 50 mL anhydrous THF was

added through cannular. The flask was cooled to −78 °C and

n

8.1 mL BuLi (2.5 M in THF, 20.2 mmol) solution was added

performed at 25 °C on

a CH Instrument CHI604xD

dropwise. The reaction mixture was stirred at −78 °C for 2

hours and a 5 mL THF solution of SiCl₄(0.742 g, 4.36 mmol)

was added dropwise through syringe. The reaction mixture was

first kept stirring at −78 °C for 1 hour and then warmed up to

room temperature and stirred overnight. The reaction mixture

was extracted by ethyl ether, followed by washing with DI H₂O

and saturated brine solution. After the organic layer was dried

with anhydrous Na₂SO₄, the solvent was removed under

reduced pressure. The crude product was purified by silica gel

electrochemical analyzer using a glassy carbon working

electrode, a platinum wire counter electrode, and a Ag/AgCl

reference electrode calibrated using ferrocene redox couple (4.8

eV below vacuum). Optical Micrographs were taken on a Carl

Zesis Axio Imager

2 microscope at 400X or 100X

magnification. Optical micrographs under cross polarized light

were obtained on a LEICA DM EP microscope at 400X

magnification. Atomic force microscopy (AFM) measurements

were performed on an Asylum MFP3D AFM instrument

operated under tapping mode. Powder Xꢀray diffraction

chromatography using hexanes to yield compound

1

as a white

1

powder (2.24 g, 78.9%). H NMR (300.13 MHz, CDCl₃):

δ

measurements were performed on

a Rigaku SmartLab

(ppm) = 7.33 (Ph-H, d, 8H, J³= 8.4 Hz), 7.53 (Ph-H, d, 8H,

HH

instrument. Differential scanning calorimetry (DSC)

measurements were performed on a Mettler Toledo DSC

STAR^esystem with ca. 10 mg sample and at scan rates of 10 °C

/ min and 1 °C / min. Thin film thickness was measured using a

KLAꢀTencor AlphaStep Dꢀ100 profiler. Electron scattering

J³= 8.1 Hz). ¹³C NMR (75.48 MHz, CDCl₃):

δ (ppm) =

HH

125.4, 131.4, 131.5, 137.6.

2

. 2ꢀBromothiophene (21.6 mL, 0.223 mol) was injected via

syringe into a 1 L 3ꢀneck round bottom flask equipped with an

addition funnel and a stir bar under positive N₂pressure.

Anhydrous THF (ca. 400 mL) was transferred into the flask

through cannular. Lithium diisopropylamide solution (2M in

THF/heptane/ethylbenzene, 123.0 mL, 0.246 mol) was

transferred into the addition funnel and added dropwise at −78

°C. The reaction mixture was kept stirring at −78 °C for 30 min

and then warmed up to room temperature. Anhydrous N,Nꢀ

dimethylformamide (25.8 mL, 0.335 mol) was added slowly

through a degassed syringe. The reaction mixture was further

stirred at room temperature overnight. After standard aqueous

experiments were carried out on

a JEOL JEMꢀ2010

transmission electron microscope (TEM) in a transmission

geometry through the thickness of the SO film (perpendicular

to the surface). The accelerating voltage of the TEM was

maintained at 200 keV. Reported images were recorded at a

nominal camera length of 300 mm on a Gatan Orius CCD

camera (1336 x 2004 pixels). TEM specimens were produced

by first spin coating a solution of SO compound onto an ITO

slide coated with PEDOT: PSS supporting layer (ca. 30 nm).

The SO films were then floated in DI water and transferred

onto carbon coated TEM grids.

workup, compound

vacuum distillation. (37.0 g, 86.8%). H NMR (300.13 MHz,

CDCl₃):

(ppm) = 7.19 (Th-H, d, 1H, J³_HH= 3.9 Hz), 7.52 (Th-

, d, 1H, J³_HH= 4.2 Hz), 9.78 (-CHO, s, 1H).

. Compound (1.09 g, 4.00 mmol) and compound

2 was obtained as a colorless liquid by

1

δ

4.2 Solar Cell Fabrication and Testing.

H

A conventional structure of ITO/MoO₃(10 nm)/active layer

(100 nm)/Al (100 nm) was adopted for the solar cells studied.

Devices were fabricated according to the following procedures.

SO and PCBM (American Dye Source, Inc.) at predetermined

weight ratios were dissolved in chlorobenzene (CB) and stirred

at 80 °C for 10 h in a nitrogen glovebox (Innovative

Technology, model PLꢀHeꢀ2GB, O₂< 0.1 ppm, H₂O < 0.1

ppm). ITOꢀcoated glass substrates (China Shenzhen Southern

8

7

5 (1.59 g,

4.80 mmol) were dissolved in 30 mL anhydrous DMF in a

pressure vessel containing a magnetic stir bar inside an argon

filled glovebox. Pd(PPh₃)₄(69.3 mg, 1.5 mol%) was then added

to the reaction mixture. The pressure vessel was sealed and

taken out of the glovebox. The reaction was carried out at 80 °C

for 24 hours and then cooled to room temperature. After

standard aqueous workup, compound

8 was further purified by

Glass Display. Ltd, 8 ꢂ/ ) were cleaned by ultrasonication

☐

silica gel chromatography with hexane/ethyl acetate (1.28 g,

88.9%). ¹H NMR (300.13 MHz, CDCl₃):

t, 3H, J³= 6.9 Hz), 1.28ꢀ1.41 (-CH₂-, m, 6H), 1.65 (-CH₂-

δ

(ppm) = 0.89 (-CH₃

,

sequentially in detergent, DI water, acetone and isopropyl

alcohol, each for 15 min. These ITOꢀcoated glass substrates

were further treated by UVꢀozone (PSD Series, Novascan) for

45 min before transferred into a nitrogen glovebox (Innovative

Technology, model PLꢀHeꢀ4GBꢀ1800, O₂< 0.1 ppm, H₂O <

0.1 ppm) for MoO₃deposition. MoO₃(10 nm) was deposited

using an Angstrom Engineering Åmod deposition system at a

base vacuum level < 7 × 10^ꢀ8Torr. The blend solution was first

filtered through a 0.45 ꢁm PTFE filter and spinꢀcoated on top

of the MoO₃layer at preset speeds for 30s. Typical thickness of

organic layers was ca. 100 nm. Al (100 nm) was finally

thermally evaporated through patterned shadow masks as

anodes. Current−voltage (I−V) characteristics were measured

by a Keithley 2400 sourceꢀmeasuring unit under simulated

AM1.5G irradiation (100 mW/cm²) generated by a Xe arcꢀlamp

based Newport 67005 150ꢀW solar simulator equipped with an

AM1.5G filter. The light intensity was calibrated by using a

Newport thermopile detector (model 818Pꢀ010ꢀ12) equipped

with a Newport 1916ꢀC Optical Power Meter.

HH

m, 2H), 2.78 (-CH₂-, t, 2H, J³= 7.8 Hz), 6.96 (Th-H, d, 1H,

HH

J³= 6.6 Hz), 7.07 (Th-H, d, 1H, J³= 3.6 Hz), 7.22 (Th-H

,

HH

d, 1H, J³= 4.8 Hz), 7.25 (Th-H, d, 1H, J³= 4.2 Hz), 7.32

HH

(

Th-H, d, 1H, J³= 3.6 Hz), 7.68 (Th-H, d, 1H, J³= 3.9

HH HH

Hz), 9.87 (-CHO, s, 1H). ¹³C NMR (75.48 MHz, CDCl₃):

δ

(ppm) = 13.9, 22.4, 29.0, 29.1, 30.2, 31.4, 123.6, 124.2, 126.2,

126.3, 129.5, 130.0, 134.9, 137.1, 138.0, 140.1, 141.2, 146.5,

182.0.

9

. Compound

8

(1.28 g, 3.55 mmol), ethylene glycol (2.0 mL,

ꢀtoluenesulfonic acid

35.8 mmol) and a catalytic amount of

p

were dissolved in 50 mL benzene in a 100 mL round bottom

flask equipped with a DeanꢀStark apparatus. The reaction

mixture was refluxed at 150 °C for 24 hours. The resulted

reaction mixture was extracted by ethyl ether and followed by

washing with saturated NaHCO₃, DI H₂O and saturated brine

solution. After the organic layer was dried with anhydrous

Na₂SO₄, the solvent was removed under reduced pressure. The

crude compound was further dried under vacuum and used for

J. Name., 2012, 00, 1-3 | 9

Journal of Materials Chemistry A

Page 10 of 13

^VJ^ieo^wu^Ar^rtn^ica^lel^ON^nlaⁱⁿm^ee

ARTICLE

DOI: 10.1039/C4TA05405A

1

next step without further purification (1.24 g, 86.4%). H NMR removal under vacuum, 12 was purified by vacuum distillation

1

as a colorless liquid (9.40 g, 81.1%). H NMR (300.13 MHz,

(300.13 MHz, CDCl₃):

δ

(ppm) = 0.88 (-CH₃, t, 3H, J³= 6.6

HH

Hz), 1.26ꢀ1.39 (-CH₂-, m, 6H), 1.64 (-CH₂-, m, 2H), 2.76 (

-

CDCl₃): δ

(ppm) = 0.89 (-CH₃, t, 3H, J³= 6.9 Hz), 1.28ꢀ1.38

-CH₂-, m, 10H), 1.68 (-CH₂-, m, 2H), 3.45 (-CH₂-, s, 1H), 4.20

-OCH₂-, t, 2H, J³_HH= 6.9 Hz). ¹³C NMR (75.48 MHz, CDCl₃):

(ppm) = 14.0, 22.6, 24.7, 25.7, 28.3, 29.1, 31.7, 67.1, 113.0,

HH

CH₂-, t, 2H, J³= 7.8 Hz), 4.04 (-OCH₂-, m, 2H), 4.14 (

(

δ

HH

OCH₂-, m, 2H), 6.09 (-OCHO-, s, 1H), 6.93 (Th-H, d, 1H, J³

HH

= 5.1 Hz), 7.01 (Th-H, d, 1H, J³= 3.6 Hz), 7.05 (Th-H, d,

HH

1H, J³= 3.6 Hz), 7.07 (Th-H, d, 1H, J³= 3.6 Hz), 7.11

162.9.

13. Compound

HH

(

Th-H, d, 1H, J³= 3.6 Hz), 7.17 (Th-H, d, 1H, J³= 5.1

8 (0.600 g, 1.66 mmol) was dissolved in 25 mL

HH HH

Hz). ¹³C NMR (75.48 MHz, CDCl₃):

δ

(ppm) = 14.1, 22.6, DMF in a 50 mL round bottom flask and immersed in an ice

29.2, 29.3, 30.6, 31.6, 65.2, 100.2, 123.0, 123.9, 124.2, 126.4, bath. Nꢀbromosuccinimide (NBS, 0.356 g, 2.00 mmol) powder

127.0, 130.1, 130.2, 135.6, 136.8, 138.2, 139.9, 140.6. was added with vigorous stirring. The reaction was warmed up

10. Compound (0.505 g, 1.25 mmol) was weighed into a dry to room temperature and stirred for overnight. The resulting

9

100 mL Schlenk flask under nitrogen, and 50 mL anhydrous reaction mixture was first extracted with CHCl₃, followed by

THF was transferred through a cannular. The flask was cooled washing with saturated Na₂SO₃, 1 M HCl solution, DI H₂O and

n

to −78 °C and 0.55 mL BuLi (2.5 M in THF, 1.37 mmol) brine solution. Solvents were removed under reduced pressure

solution was added dropwise through a degassed syringe. The and the resulted crude product was further purified by column

reaction mixture was stirred at −78 °C for 30 min and warmed chromatography to get 13 as a yellow solid (0.655 g, 89.5%).

up to room temperature. A Me₃SnCl solution (1 M in THF, 1.5

δ (ppm) = 0.89 (-CH₃, t, 3H,

¹H NMR (300.13 MHz, CDCl₃):

J³= 6.9 Hz), 1.28ꢀ1.41 (-CH₂-, m, 6H), 1.62 (-CH₂-, m, 2H),

mL, 1.5 mmol) was then added dropwise. The reaction mixture

was kept stirring at room temperature overnight. The resulting

reaction mixture was extracted by ethyl ether, followed by

washing with DI H₂O and saturated brine solution. After the

organic layer was dried with anhydrous Na₂SO₄, the solvent

was removed under reduced pressure. The crude compound was

further dried under high vacuum and used for next step without

HH

2.71 (-CH₂-, t, 2H, J³_HH= 7.8 Hz), 6.92 (Th-H, s, 1H), 7.01 (Th-

H

, d, 1H, J³= 3.9 Hz), 7.25 (Th-H, d, 1H, J³= 4.8 Hz),

HH HH

7.30 (Th-H, d, 1H, J³= 3.6 Hz), 7.68 (Th-H, d, 1H, J³

=

HH

4.2 Hz), 9.87 (-CHO, s, 1H). ¹³C NMR (75.48 MHz, CDCl₃):

δ

(ppm) = 13.9, 22.4, 28.9, 29.1, 30.1, 31.4, 111.1, 123.9, 126.2,

126.7, 131.0, 132.7, 135.5, 136.5, 137.1, 140.7, 141.5, 146.2,

182.1.

1

further purification (0.709 g, 100%). H NMR (300.13 MHz,

CDCl₃):

δ

(ppm) = 0.38 (-CH₃, s, 9H), 0.87 (-CH₃, t, 3H, J³

=

-

14

.

Compound 13 (0.655 g, 1.50 mmol) and

HH

6.9 Hz), 1.26ꢀ1.39 (-CH₂-, m, 6H), 1.64 (-CH₂-, m, 2H), 2.78 (

trimethylstannylbenzene (0.542 g, 2.25 mmol) were dissolved

in 30 mL anhydrous DMF in a pressure vessel containing a

magnetic stir bar under argon. Pd(PPh₃)₄(43.4 mg, 2.5 mol%),

CuI (28.5 mg, 10.0 mol%) and CsF (0.342 g, 2.25 mmol) were

added to the pressure vessel in an argon filled glovebox. The

CH₂-, t, 2H, J³= 7.8 Hz), 4.02 (-OCH₂-, m, 2H), 4.15 (

HH

OCH₂-, m, 2H), 6.08 (-OCHO-, s, 1H), 6.99 (Th-H, d, 1H, J³

HH

= 3.9 Hz), 7.00 (Th-H, s, 1H), 7.04 (Th-H, d, 1H, J³= 3.6

HH

Hz), 7.06 (Th-H, d, 1H, J³_HH= 3.9 Hz), 7.10 (Th-H, d, 1H, J³

HH

= 3.6 Hz). ¹³C NMR (75.48 MHz, CDCl₃):

δ (ppm) = −8.2, pressure vessel was sealed and taken out of the glovebox. The

14.1, 22.6, 29.2, 29.3, 30.7, 31.6, 65.2, 100.2, 122.9, 124.2, reaction was stirred at 80 °C for 24 hours. After cooling to

126.0, 127.0, 135.9, 136.0, 136.5, 136.8, 138.3, 138.4, 140.4, room temperature, the reaction mixture was first extracted with

141.0.

CHCl₃and followed by washing with 1 M HCl solution,

11. Compound

1

(0.170 g, 0.26 mmol), compound 10 (0.709 g, saturated NaHCO₃, DI H₂O and brine solution. Solvents were

1.25 mmol) and Pd(PPh₃)₄(9 mg, 3 mol%) were dissolved in removed under reduced pressure and the crude product was

20 mL anhydrous DMF in a pressure vessel containing a further purified by silica gel column chromatography with

1

magnetic stir bar under argon. The pressure vessel was sealed hexane/ethyl acetate (0.46 g, 70.7%) H NMR (300.13 MHz,

and stirred at 90 °C for 24 hours. After cooling to room

CDCl₃):

(

δ

(ppm) = 0.90 (-CH₃, t, 3H, J³= 6.9 Hz), 1.30ꢀ1.42

HH

-CH₂-, m, 6H), 1.62 (-CH₂-, m, 2H), 2.79 (-CH₂-, t, 2H, J³

=

temperature, the reaction mixture was extracted with CHCl₃,

followed by washing with 1 M HCl, saturated NaHCO₃, DI

H₂O and brine. After the organic layer was dried with

anhydrous Na₂SO₄, the solvent was removed under reduced

pressure. The crude product was further purified by silica gel

chromatography with dichloromethane/ethyl acetate to yield 11

HH

7.8 Hz), 7.11 (Th-H, d, 1H, J³= 3.9 Hz), 7.17 (Th-H, s, 1H),

HH

7.25 (Th-H, d, 1H, J³= 3.6 Hz), 7.30 (Ph-H, d, 1H, J³

=

HH

6.6 Hz), 7.32 (Th-H, d, 1H, J³= 3.6 Hz), 7.39 (Ph-H-, m,

HH

2H), 7.60 (Ph-H-, d, 2H, J³= 8.4 Hz), 7.67 (Th-H, d, 1H,

HH

J³= 3.9 Hz), 9.87 (-CHO, s, 1H). ¹³C NMR (75.48 MHz,

HH

1

as a brown solid (200 mg, 43.3%). H NMR (300.13 MHz,

CDCl₃):

δ (ppm) = 14.1, 22.6, 29.2, 29.7, 30.4, 31.6, 123.9,

CDCl₃):

1.43 (-CH₂-, m, 24H), 1.71 (-CH₂-, m, 8H), 2.81 (-CH₂-, t, 8H,

J³= 4.8 Hz), 7.13 (Th-H, d, 4H, J³= 3.6 Hz), 7.24 (Th-H

δ

(ppm) = 0.90 (ꢀCH₃, t, 12H, J³= 6.9 Hz), 1.32ꢀ

HH

125.6, 126.2, 126.4, 126.5, 127.8, 128.9, 129.2, 133.8, 135.1,

137.3, 138.3, 141.4, 141.5, 142.8, 146.9, 182.3.

MO. Compounds 14 (0.131 g, 0.300 mmol) and 12 (0.118 g,

0.60 mmol) were dissolved 15 mL CHCl₃containing 0.5 mL

triethylamine. The reaction mixture was first bubbled with N₂

for 30 min and stirred at room temperature for 2 days. Solvents

,

HH

s, 4H), 7.27(Th-H, d, 4H, J³= 3.6 Hz), 7.33 (Th-H, d, 4H,

HH

J³= 3.9 Hz), 7.64(Ph-H, m, 16H), 7.68 (Th-H, d, 4H, J³

=

δ

HH

3.9 Hz), 9.87 (-CHO, s, 4H). ¹³C NMR (75.48 MHz, CDCl₃):

(ppm) = 14.1, 22.6, 29.2, 29.7, 30.4, 31.6, 124.0, 125.0, 126.5, were removed under reduced pressure and MO was purified by

126.6, 126.8, 129.9, 133.1, 135.1, 135.3, 136.9, 137.4, 138.1, silica

gel

column

chromatography

with

141.5, 141.6, 142.2, 146.8, 182.4. TOF MS (APPI⁺): Calcd. for hexane/dichloromethane and by precipitation into methanol as a

SiC₁₀₀H₉₂O₄S₁₂N: 1768.3414 [M⁺], 1791.3312 [M+Na⁺]; found: dark red solid (160 mg, 86.7%) ¹H NMR (300.13 MHz,

1768.3419 [M⁺], 1791.3317 [M+Na⁺].

12. Cyanoacetic acid (5.00 g, 58.8 mmol), octanol (9.8 mL,

CDCl₃):

(

δ

(ppm) = 0.90 (-CH₃, t, 3H, J³= 6.9 Hz), 1.33ꢀ1.42

HH

-CH₂-, m, 16H), 1.74 (-CH₂-, m, 4H), 2.81 (-CH₂-, t, 2H, J³

HH

= 7.8 Hz), 4.30 (-CH₂-, t, 2H, J³_HH= 6.9 Hz) 7.13 (Th-H, d, 1H,

J³= 4.2 Hz), 7.18 (Th-H, s, 1H), 7.26 (Th-H, d, 1H, J³

61.8 mmol) and a catalytic amount of pꢀtoluenesulfonic acid

were dissolved in 30 mL benzene into a 100 mL round bottom

flask equipped with a DeanꢀStark apparatus. The reaction

mixture was refluxed at 120 °C for 24 hours. After solvent

=

HH

3.6 Hz), 7.31 (Ph-H-, d, 1H, J³_HH= 7.2 Hz), 7.37 (Th-H, d, 1H,

J³_HH= 3.9 Hz), 7.39 (Ph-H-, m, 2H), 7.61 (Ph-H-, d, 2H, J³

=

HH

10 | J. Name., 2012, 00, 1-3

Page 11 of 13

Journal Name

Journal of Materials Chemistry A

^{View Artic}A^leR^OnT^liIⁿC^eLE

DOI: 10.1039/C4TA05405A

7.8 Hz), 7.68 (Th-H, d, 1H, J³= 4.2 Hz), 8.26 (ꢀCH=Cꢀ, s,

HH

9 H. Zhou; L. Yang; A. C. Stuart; S. C. Price; S. Liu; W. You Angew.

Chem. Int. Ed. 2011, 50, 2995.

1H). ¹³C NMR (75.48 MHz, CDCl₃):

δ (ppm) = 14.1, 22.6,

25.6, 28.5, 29.1, 29.7, 30.4, 31.7, 31.8, 66.6, 97.6, 116.0, 124.1,

125.6, 126.3, 126.6, 126.9, 127.8, 128.9, 129.3, 133.7, 134.2,

134.8, 138.7, 139.1, 141.6, 142.9, 146.0, 147.2, 163.0.

10 Z. He; C. Zhong; S. Su; M. Xu; H. Wu; Y. Cao Nat. Photon. 2012, 6,

591.

11 T. Yang; M. Wang; C. Duan; X. Hu; L. Huang; J. Peng; F. Huang; X.

Gong Energy Environ. Sci. 2012, , 8208.

12 J. You; L. Dou; K. Yoshimura; T. Kato; K. Ohya; T. Moriarty; K.

SO. Compound 11 (88 mg, 0.05 mmol), compound 12 (98.5

mg, 0.50 mmol) and 1 mL triethylamine were dissolved in 30

mL CHCl₃. The reaction mixture was purged with N₂for 30

min and then stirred at room temperature for 48 h. After solvent

removal under vacuum, SO was purified by silica gel

chromatography with hexane/dichloromethane and then

precipitation into methanol as a dark red solid (105 mg, 84.5%).

5

Emery; C.ꢀC. Chen; J. Gao; G. Li; Y. Yang Nat. Commun. 2013, 4,

1446.

13 J. You; C.ꢀC. Chen; Z. Hong; K. Yoshimura; K. Ohya; R. Xu; S. Ye;

J. Gao; G. Li; Y. Yang Adv. Mater. 2013, 25, 3973.

¹H NMR (300.13 MHz, CDCl₃):

24H), 1.31ꢀ1.34 (-CH₂-, m, 64H), 1.69ꢀ1.77 (-CH₂-, m, 16H),

2.82 (-CH₂-, t, 8H, J³= 7.8 Hz), 4.29 (-OCH₂-, t, 8H, J³

δ (ppm) = 0.89ꢀ0.90 (-CH₃, t,

14 B. Walker; C. Kim; T.ꢀQ. Nguyen Chem. Mater. 2011, 23, 470.

15 J. E. Anthony Chem. Mater. 2011, 23, 583.

=

HH

6.9 Hz), 7.14 (Th-H, d, 4H, J³= 3.9 Hz), 7.25 (Th-H, d, 4H,

16 J. M. Szarko; J. Guo; B. S. Rolczynski; L. X. Chen Nano Rev. 2011,

7249.

2,

HH

J³= 3.9 Hz), 7.27 (Th-H, s, 4H), 7.37 (Th-H, d, 4H, J³

=

HH

3.9 Hz), 7.64 (Ph-H, m, 16H), 7.67 (Th-H, d, 4H, J³= 4.2

HH

17 Y. Lin; Y. Li; X. Zhan Chem. Soc. Rev. 2012, 41, 4245.

18 A. Mishra; P. Bäuerle Angew. Chem. Int. Ed. 2012, 51, 2020.

19 Y. Chen; X. Wan; G. Long Acc. Chem. Res. 2013, 46, 2645.

Hz), 8.26 (ꢀCH=C, s, 4H). ¹³C NMR (75.48 MHz, CDCl₃):

δ

(ppm) = 14.1, 22.6, 25.8, 28.6, 29.2, 29.7, 30.4, 31.7, 31.8,

66.6, 97.7, 116.0, 124.2, 125.0, 126.7, 126.8, 126.9, 129.9,

133.2, 134.3, 135.0, 135.1, 136.9, 138.6, 139.1, 141.7, 142.4,

146.1, 147.1, 163.0.

20 J. E. Coughlin; Z. B. Henson; G. C. Welch; G. C. Bazan Acc. Chem.

Res. 2014, 47, 257.

21 J. Roncali; P. Leriche; P. Blanchard Adv. Mater. 2014, 26, 3821.

22 T. Jarosz; M. Lapkowski; P. Ledwon Macromol. Rapid Commun.

2014, 35, 1006.

Acknowledgements

The authors would like to acknowledge University of New

Mexico and National Science Foundation (Grant No. IIAꢀ

1301346) for financial support for this research.

23 V. Steinmann; N. M. Kronenberg; M. R. Lenze; S. M. Graf; D. Hertel;

K. Meerholz; H. Bürckstümmer; E. V. Tulyakova; F. Würthner Adv.

Energy Mater. 2011,

1, 888.

Notes and references

Department of Chemistry & Chemical Biology, University of New

24 L.ꢀY. Lin; Y.ꢀH. Chen; Z.ꢀY. Huang; H.ꢀW. Lin; S.ꢀH. Chou; F. Lin;

a

C.ꢀW. Chen; Y.ꢀH. Liu; K.ꢀT. Wong J. Am. Chem. Soc. 2011, 133

,

Mexico, MSC03ꢀ2060, Albuquerque, NM 87131, USA.

15822.

b

Cottonwood Classical Preparatory School, 801 Jefferson Street NE,

25 R. Fitzner; E. Reinold; A. Mishra; E. MenaꢀOsteritz; H. Ziehlke; C.

Körner; K. Leo; M. Riede; M. Weil; O. Tsaryova; A. Weiβ; C.

Uhrich; M. Pfeiffer; P. Bäuerle Adv. Funct. Mater. 2011, 21, 897.

26 J. Zhou; X. Wan; Y. Liu; G. Long; F. Wang; Z. Li; Y. Zuo; C. Li; Y.

Chen Chem. Mater. 2011, 23, 4666.

Albuquerque, NM 87109, USA.

c

TEM & FIB Laboratories, University of New Mexico, Albuquerque,

87131, USA.

Eꢀmail: yangqin@unm.edu

†

Electronic Supplementary Information (ESI) available: DFT

27 G. Wei; X. Xiao; S. Wang; J. D. Zimmerman; K. Sun; V. V. Diev; M.

E. Thompson; S. R. Forrest Nano Lett. 2011, 11, 4261.

calculations, Xꢀray scattering profile of SO films, TEM and SAED

images, JꢀV curves, optical micrographs under cross polarized light,

optical micrographs of the SO/PCBM devices under slowꢀcooling

conditions, relation plot of V_OCvs. P3DTV contents, optical micrographs

of the ternary blends and NMR spectra of all synthesized compounds. See

DOI: 10.1039/b000000x/

28 Y. Sun; G. C. Welch; W. L. Leong; C. J. Takacs; G. C. Bazan; A. J.

Heeger Nat. Mater. 2012, 11, 44.

29 X. Liu; Y. Sun; L. A. Perez; W. Wen; M. F. Toney; A. J. Heeger; G.

C. Bazan J. Am. Chem. Soc. 2012, 134, 20609.

30 G. Chen; H. Sasabe; Z. Wang; X.ꢀF. Wang; Z. Hong; Y. Yang; J. Kido

Adv. Mater. 2012, 24, 2768.

1 H. Hoppe; N. S. Sariciftci J. Mater. Res. 2004, 19, 1924.

31 D. V. Talapin; J. H. Nelson; E. V. Shevchenko; S. Aloni; B. Sadtler;

2 H. Spanggaard; F. C. Krebs Sol. Energy Mater. Sol. Cells 2004, 83

,

A. P. Alivisatos Nano Lett. 2007, 7, 2951.

125.

32 S. Dayal; M. O. Reese; A. J. Ferguson; D. S. Ginley; G. Rumbles; N.

Kopodakis Adv. Funct. Mater. 2010, 20, 2629.

3 B. C. Thompson; J. M. J. Fréchet Angew. Chem. Int. Ed. 2008, 47, 58.

4 G. Dennler; M. C. Scharber; C. J. Brabec Adv. Mater. 2009, 21, 1323.

5 Y.ꢀJ. Cheng; S.ꢀH. Yang; C.ꢀS. Hsu Chem. Rev. 2009, 109, 5868.

33 J. Lim; D. Lee; M. Park; J. Song; S. Lee; M. S. Kang; C. Lee; K. Char

J. Phys. Chem. C 2014, 118, 3942.

6 H.ꢀY. Chen; J. Hou; S. Zhang; Y. Liang; G. Yang; Y. Yang; L. Yu; Y.

Wu; G. Li Nat. Photon. 2009, , 649.

34 S. Roquet; R. De Bettignies; P. Leriche; A. Cravino; J. Roncali J.

Mater. Chem. 2006, 16, 3040.

3

7 Y. Liang; Z. Xu; J. Xia; S.ꢀT. Tsai; Y. Wu; G. Li; C. Ray; L. Yu Adv.

Energy Mater. 2010, 22, E135.

35 Z. Lin; J. Bjorgaard; A. G. Yavuz; M. E. Köse J. Phys. Chem. C 2011,

115, 15097.

8 T.ꢀY. Chu; J. Lu; S. Beaupré; Y. Zhang; J.ꢀR. Pouliot; S. Wakim; J.

Zhou; M. Leclerc; Z. Li; J. Ding; Y. Tao J. Am. Chem. Soc. 2011,

133, 4250.

36 We have started out on tetrathienylsilane cored tetrapod similar to the

one reported herein. However, the aldehyde intermediates

decomposed during the Knoevenegal condensation reactions owing

to siliconꢀcarbon bond cleavages.

J. Name., 2012, 00, 1-3 | 11

Journal of Materials Chemistry A

Page 12 of 13

ARTICLE

^VJ^ieo^wu^Ar^rtn^ica^lel^ON^nlaⁱⁿm^ee

DOI: 10.1039/C4TA05405A

37 F. C. Spano Acc. Chem. Res. 2010, 43, 429.

38 L. F. Lai; J. A. Love; A. Sharenko; J. E. Coughlin; V. Gupta; S.

Tretiak; T.ꢀQ. Nguyen; W.ꢀY. Wong; G. C. Bazan J. Am. Chem. Soc.

2014, 136, 5591.

39 P. N. Murgatroyd J. Phys. D: Appl. Phys. 1970, 3, 151.

40 G. De Luca; E. Treossi; A. Liscio; J. M. Mativetsky; L. M. Scolaro; V.

Palermo; P. S. Samorì J. Mater. Chem. 2010, 20, 2493.

41 C. Sinturel; M. Vayer; M. Morris; M. A. Hillmyer Macromolecules

2013, 46, 5399.

42 C. Zhang; T. Matos; R. Li; S.ꢀS. Sun; J. E. Lewis; J. Zhang; X. Jiang

Polym. Chem. 2010, 1, 663.

43 Y. Qin; M. A. Hillmyer Macromolecules 2009, 42, 6429.

44 J. Y. Kim; Y. Qin; D. M. Stevens; O. Ugurlu; V. Kalihari; M. A.

Hillmyer; C. D. Frisbie J. Phys. Chem. C 2009, 113, 10790.

45 I. Jestin; P. Frère; P. Blanchard; J. Roncali Angew. Chem. Int. Ed.

1998, 37, 942.

46 G. Yang; K. Hu; Y. Qin J. Polym. Sci. A Polym. Chem. 2014, 52, 591.

47 E. Olejnik; B. Pandit; T. Basel; E. Lafalce; C.ꢀX. Sheng; C. Zhang; X.

Jiang; Z. V. Vardeny Phys. Rev. B 2012, 85, 235201.

48 A. J. Musser; M. AlꢀHashimi; M. Maiuri; D. Brida; M. Heeney; G.

Cerullo; R. H. Friend; J. Clark J. Am. Chem. Soc. 2013, 135, 12747.

49 B. Qi; J. Wang J. Mater. Chem. 2012, 22, 24315.

50 P. P. Khlyabich; B. Burkhart; B. C. Thompson J. Am. Chem. Soc.

2011, 133, 14534.

51 P. P. Khlyabich; B. Burkhart; A. E. Rudenko; B. C. Thompson

Polymer 2013, 54, 5267.

52 W. He; Y. Jiang; Y. Qin Polym. Chem. 2014, 5, 1298.

53 S. Pu; C. Zheng; Q. Sun; G. Liu; C. Fan Chem. Commun. 2013, 49

,

8036.

54 R. Grisorio; L. De Marco; G. Allegretta; R. Giannuzzi; G. P. Suranna;

M. Manca; P. Mastrorilli; G. Gigli Dyes Pigments 2013, 98, 221.

55 E. Amir; K. Sivanandan; J. E. Cochran; J. J. Cowart; S.ꢀY. Ku; J. H.

Seo; M. L. Chabinyc; C. J. Hawker J. Polym. Sci. A Polym. Chem.

2011, 49, 1933.

56 D. J. Crouch; D. Sparrowe; M. Heeney; I. McCulloch; P. J. Skabara

Macromol. Chem. Phys. 2010, 211, 2642.

12 | J. Name., 2012, 00, 1-3

Page 13 of 13

Journal of Materials Chemistry A

View Article Online

DOI: 10.1039/C4TA05405A

Graphical Abstract

The tetrahedral geometry of the conjugated small molecule displays unique thermal

properties that lead to distinct behaviors in binary and ternary blend organic solar cells.

Article Doi

A molecular breakwater-like tetrapod for organic solar cells

DOI: 10.1039/c4ta05405a

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c4ta05405a

Full text of DOI:10.1039/c4ta05405a

Products guided by the article

R&D Labs maybe for 133787-27-6

Relevant to this article

Hot Product