Design principle of conjugated polyelectrolytes to make them water-soluble and highly emissive

Design Principle of Conjugated Polyelectrolytes to Make

Them Water-Soluble and Highly Emissive

Kangwon Lee, Hyong-Jun Kim, and Jinsang Kim*

biological applications. In addition, main-

taining the emissive property of a CPE in

aqueous solution is another requirement

for many biosensor applications because

the merit of using conjugated polymers

as a sensor is their ampliﬁed ﬂuorescence

signaling property upon environmental

changes.^[35–45]

However, in this regard, CPE inher-

ently has a critical solubility limitation

in aqueous environment because the

π-conjugated polymer backbone is chemi-

cally hydrophobic and structurally rigid.

The rigid-rod and hydrophobic nature of

CPEs induces polymer aggregation via

intermolecular hydrophobic interaction

among the polymer backbones in aqueous

environment.^[37,38,46]Therefore, the solu-

bility of CPE in water is signiﬁcantly low,

consequently inducing signiﬁcant decrease

in the ﬂuorescence quantum yield due to the

aggregation-induced self-quenching.^[47–51]

Moreover, once CPEs are completely

The correlation between the molecular design of a conjugated polyelectrolyte

(CPE) and its aggregated structure and the emissive properties in water is sys-

tematically investigated by means of UV–vis spectrometry, ﬂuorescence spec-

troscopy, and scanning/transmission electron microscopy. Five different and

rationally designed CPEs having carboxylic acid side chains are synthesized.

All ﬁve conjugated polyelectrolytes are seemingly completely soluble in water

in visual observation. However, their quantum yields are dramatically different,

changing from 0.45 to 51.4%. Morphological analysis by electron microscopy

combined with ﬂuorescence spectrophotometry reveals that the CPEs form

self-assembled aggregates at the nanoscale depending on the nature of their

side chains. The feature of the self-assembled aggregates directly determines

the emissive property of the CPEs. The nature and the length of the spacer

between the carboxylic acid group and the CPE backbone have a strong inﬂu-

ence on the quantum yield of the CPEs. Our study demonstrates that bulky

and hydrophilic side chains and spacers are required to achieve complete

water-solubility and high quantum yield of CPEs in water, providing an impor-

tant molecular design principle to develop functional CPEs.

dried, it is tremendously difﬁcult to re-dissolve them in water

due to the rigid hydrophobic nature of the backbone and

resulting strong and cohesive aggregation. Besides the solu-

bility issue, there is another requirement for CPE to be useful

for biological applications. To achieve efﬁcient and convenient

bio-conjugation between the CPE and a biological molecule,

CPEs should have an efﬁcient and convenient functional group

such as a carboxylic acid group or an amine group.^[52,53]Due

to these demanding requirements, it remains a difﬁcult task

to develop highly emissive and completely water-soluble func-

tional CPEs.

1. Introduction

A conjugated polyelectrolyte (CPE)^[1,2]a conjugated polymer con-

taining a charged (anionic or cationic) group, has received con-

siderable attention for its bio-applications such as solution-based

DNA sensor,^[3–11]DNA microarray,^[12–14]protein sensor,^[15–24]

bioimaging^[25–28]as well as optoelectronic applications such

as organic semiconductors,^[29–31]light-emitting devices,^[32,33]

and actuators.^[34]The ionic side group plays an important role

to provide CPEs with water-solubility that is central to many

Several research groups have developed CPE-based func-

tional systems by utilizing the emissive property of CPEs.

Leclerc et al. developed DNA sensors using cationically charged

polythiophene derivatives. Charge–charge interaction between

the cationic CPE and a single strand DNA and subsequent

detection of the complementary DNA produces a conforma-

tion change of the CPE and consequent color change as a sen-

sory signal.^{[7,15,54–57]}Bazan and co-workers have reported signal

amplifying biosensors based on cationically charged CPEs and

ﬂuorescence resonance energy transfer (FRET).^[4]Schanze

et al. investigated CPE systems and reported ampliﬁed ﬂuores-

Dr. K. Lee,^[+]Dr. H.-J. Kim,^[++]Prof. J. Kim

Department of Materials Science and Engineering

University of Michigan

Ann Arbor, MI 48109, USA

E-mail: jinsang@umich.edu

Prof. J. Kim

Department of Chemical Engineering

Macromolecular Science and Engineering

Biomedical Engineering

University of Michigan

Ann Arbor, MI 48109, USA

cence quenching of sulfonated CPEs due to π–π aggregation of

[+] Present address: School of Engineering and Applied Science,

Harvard University, MA, USA

[37,38,40]

the rigid linear CPEs in aqueous media.

Bunz and co-

workers have investigated pH-dependent optical properties of

CPEs by modulating carboxylic acid side-chain.^[58]Furthermore,

cationic gold nanoparticles conjugated with carboxylated CPEs

[++] Present address: College of Engineering, Kongju National University

DOI: 10.1002/adfm.201102027

1076 wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2012, 22, 1076–1086

polyelectrolytes have successfully been developed for bacteria

identiﬁcation vehicles.^[59]Recently, completely water-soluble

CPEs have been also reported.^[53,60]Understanding the correla-

tion among the quenching dynamics, the chemical structure of

receptor-functionalized CPEs, and the ﬂuorescence efﬁciency

is critically important to develop biosensory materials based

on water-soluble CPEs. However, to our knowledge there has

not been any article that comprehensively and systematically

provides design principles to develop highly emissive and com-

pletely water-soluble CPEs.

We have rationally designed and prepared a series of PPE-

based CPEs and systematically investigated the effects of the side

chain structure on the solubility and the ﬂuorescence quantum

yield of the CPEs in aqueous environment. Here, we report

our comprehensive quantum yield study, scanning transmis-

sion electron microscopy and conventional/cryogenic transmis-

sion electron microscopy (TEM) study to reveal the correlation

among the chemical and structural characteristics of the side

chain of the CPEs, their molecular assembly, and their emis-

sive property. We chose carboxylic acid moiety for this study as

a pendant ionic group considering that it is the most conven-

ient functional group for bioconjugation with the ubiquitous

amine group present in biological molecules. As the molecular

design parameter, we controlled the bulkiness of the side chain,

the length of the linker molecule between the conjugated back-

bone and the carboxylic acid group, and the hydrophobic and

hydrophilic property of the linker as illustrated in Figure 1.

concentrated H₂SO₄were heated for 24 h to reﬂux, with separa-

tion of the water using a Dean–Stark trap. Reaction mixture was

cooled down and the organic layer was washed with water and

dried with MgSO₄. Further puriﬁcation was done by column

chromatography (ethyl acetate: hexane = 1:15 v/v) to get vis-

cous yellow oil (0.14 g, 30%). H-NMR (500 MHz, CDCl₃): δ/

ppm 8.26 (s, 2H, aromatic), 4.27 (d, 4 H, -OCH₂-), 1.79 (m, 2H,

-CH-), 1.55-1.30 (m, 16H, -CH₂-), 0.95 (m, 12H, CH₃). ¹³C-NMR

(125 MHz, CDCl₃): δ/ppm 165.5, 139.9, 138.1, 92.7, 68.6, 38.9,

30.6, 29.0, 24.0, 23.0, 14.1, 11.1. HRMS (Voltage EI+): calculated

m/z of [M+] 642.0691; measured m/z 642.0679.

Diethyl

(M2): To

4,4′-(2,5-diiodo-1,4-phenylene)bis(oxy)dibutanoate

a solution of 2,5-diiodo-1,4-hydroquinone (2,

1.0 g, 2.76 mmol) were added a potassium carbonate (1.615 g,

8.28 mmol), ethyl 4-bromobutyrate (1.615 g, 8.28 mmol) and

DMF (15 mL) and reaction mixture was stirred at 80 °C for

48 h. After the reaction, reaction mixture was cooled down and

ﬁltered. DMF was removed with rotary evaporator at reduced

pressure. Crude mixture was re-dissolved in chloroform and

extracted twice with deionized water. After drying over MgSO₄

and ﬁltering, chloroform was removed in vacuo. Further puri-

ﬁcation was done by column chromatography (ethyl acetate:

hexane = 1:1 v/v) and the following recrystallization in meth-

anol at –18 °C to give white waxy powder (yield: 0.65 g, 41%).

¹H-NMR (500 MHz, CDCl₃): δ/ppm 7.10 (s, 2H, aromatic),

4.20 (m, 4H, -OCH₂CH₃), 4.01 (t, 4H, -OCH₂-), 2.60 (t, 4H,

-CH₂COO-), 2.15 (m, 4H, -CH₂-), 1.27 (t, 6H, -CH₃). ¹³C-NMR

(125 MHz, CDCl₃): δ/ppm 173.1, 152.7, 122.5, 86.1, 69.9, 60.3,

30.6, 24.4, 14.1. Elemental analysis calcd; C 36.63, H: 4.10,

obsd; C: 36.71, H: 4.15.

Diethyl 7,7′-(2,5-diiodo-1,4-phenylene)bis(oxy)diheptanoate

(M3): Synthetic procedure for this compound is the same as

that for M2 except for using ethyl 7-bromoheptanoate (2 g,

8.43 mmol) as a reactant and different column eluent (ethyl

acetate: hexane = 1:4 v/v) for column puriﬁcation (yield: 0.89 g,

47%). ¹H-NMR (500 MHz, CDCl₃): δ/ppm 7.18 (s, 2H, aromatic

C-H), 4.15 (m, 4H, COO-CH₂-CH₃), 3.94 (t, 4H, O-CH₂-), 2.33

2. Results and Discussion

2.1. Synthesis of the Monomers and Polymers (P1-P5)

Synthesis of Bis(2-ethylhexyl) 2,5-diiodoterephthalate (M1):

2,5-diiodoterephthalic acid (1, 0.3 g, 0.72 mmol), 2-ethyl-1-

hexanol (0.28 g, 2.16 mmol), toluene (20 mL), and 0.1 mL of

Figure 1. The chemical structure of the CPEs (P1-P5).

Adv. Funct. Mater. 2012, 22, 1076–1086

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1077

(t, 4H, CH₂-CH₂-CO-), 1.82 (m, 4H, -CH₂-), 1.69 (m, 4H, -CH₂-),

1.54 (m, 4H, -CH₂-), 1.42 (m, 4H, -CH₂-), 1.27 (t, 6H, -CH₃).

¹³C-NMR (125 MHz, CDCl₃): δ/ppm 173.1, 152.7, 122.5, 86.1,

70.1, 60.3, 33.5, 31.3, 29.1, 25.7, 25.5, 14.1. Elemental analysis

calcd; C 42.75, H: 5.38, obsd; C: 42.85, H: 5.40.

2.51 (t, 4H, -CH₂COO-), 1.42 (s, 18H, -C(CH₃)₃), ¹³C-NMR

(125 MHz, CDCl3): δ/ppm 28.3, 36.4, 67.1, 69.8, 70.4, 70.6, 70.7,

70.8, 80.7, 86.6, 123.6, 153.3, 171.1. HRMS (C₃₂H₅₂I₂O₁₂+Na):

calculated m/z of 905.1435; measured m/z 905.1442.

1,4-diiodo-2,5-bis(11-carboxy-3,6,9-trioxaundecyloxy)benzene

(M4): To a 4.00 g (4.53 mmol) of compound 5 was added 85 mL

of triﬂuoroacetic acid (CF₃COOH). As soon as triﬂuoroacetic

acid was added, the color of reaction mixture turned red. The

mixture was stirred at room temperature for overnight. The

reaction mixture was evaporated at reduced pressure. The crude

mixture was dissolved in chloroform and washed with water

three times. Organic layer was dried over MgSO₄and ﬁltered.

The ﬁltrate was evaporated to dryness and the compound M4

was further dried in vacuo and solidiﬁed to white-yellow waxy

powder (yield: 2.57 g, 74%). ¹H-NMR (500 MHz, CDCl₃): δ/

ppm 9.8 (broad s, 2H, -COOH), 7.22 (s, 2H, aromatic), 4.15 (t,

4H, -OCH₂-), 3.83 (t. 4H, -OCH₂-), 3.81-3.50 (m, 20H, -OCH₂-),

2.60 (t, 4H, -CH₂COOH). ¹³C-NMR (125 MHz, CDCl3): δ/ppm

35.2, 66.7, 69.8, 70.5, 70.7, 70.8, 70.9, 71.32, 86.6, 123.6, 153.3,

175.4. MS (Voltage ESI-): calculated m/z of [M-H] 769.0218;

measured m/z 769.0221.

Tert-butyl 3-(2-(2-(2-hydroxyethoxy)ethoxy)ethoxy)propanoate

(3): This compound was prepared by procedure in a previous

literature through a slight modiﬁcation.^[52]In 1000 mL of

2-necked round-bottomed ﬂask, triethylene glycol (128 mL,

0.40 mol) is dissolved in 500 mL of THF. 0.34 g (14.8 mmol)

of sodium lump was sliced and added to the solution under

argon purging. The solution was vigorously stirred to dissolve

sodium completely. After no more gas or bubble, tert-butyl acr-

ylate (48 mL, 0.33 mol) was added to the solution. The reaction

solution was stirred under argon atmosphere at room tempera-

ture for 20 h. The solution was neutralized with 1 m HCl and

THF was evaporated at reduced pressure. Crude compound

was suspended to saturated brine and extracted with ethyl ace-

tate. Organic layer was washed with saturated NaCl solution

and water again and dried over anhydrous MgSO₄(yield: 58.5 g,

53%). H-NMR (500 MHz, CDCl₃): δ/ppm 3.75-3.21 (m, 14H,

-OCH₂-), 2.69 (broad s, 1H, OH), 2.51 (t, 2H, -CH₂COO-), 1.45

(s, 9H, -C(CH₃)₃). ¹³C-NMR (125 MHz, CDCl3): δ/ppm 28.2,

36.3, 61.7, 66.9, 70.4, 70.5, 70.6, 70.7, 80.6, 171.0.

Polymer synthesis P1: M1 (65 mg, 0.14 mmol) and M5

(90 mg, 0.14 mmol) were placed into a Schlenck ﬂask (50 mL).

Toluene (1.5 mL) and diisopropylamine (3 mL) were added.

After complete dissolution of two monomers, the solution was

degassed by three times of vacuum and argon purging. In a

separate Schlenck ﬂask, tetrakistriphenylphosphine palladium

(0) and copper (I) iodide were dissolved in toluene (1.5 mL)

under a nitrogen atmosphere in a glove box and degassed.

The degassed solution containing catalyst was cannulated onto

the monomer solution. After transfer of the catalysis solu-

tion to monomer solution, polymerization solution was ﬁnally

degassed again and allowed to stir under argon purging at 55 °C

for 2 days. The reaction mixture ﬁltered with 0.45 micrometer

membrane syringe. The toluene solution was precipitated in

methanol 2 times. For deprotection of ethylhexyl group of car-

boxylic group, the collected ﬂuorescent yellow precipitate was

redissolved in 100 mL of tetrahydrofuran (THF) and 1 m of

NaOH (100 mL) was added. The solution was stirred overnight

at 35 °C. THF was evaporated at the reduced pressure, ﬁltered

and the water solution was dialyzed against deionized water for

3 days (membrane MW cut off: 12 000–14 000 gmol⁻¹, 10 × 4 L

water exchanges). The polymer solution was lyophilized to yield

Tert-butyl3-(2-(2-(2-(toluenesulfonyloxy)ethoxy)ethoxy)ethoxy)-

propanoate (4): Compound 3 (58.5 g, 0.21 mol) and triethyl-

amine (171 mL) was dissolved in anhydrous dichloromethane

(290 mL) and the solution was cooled down to 4 °C using iced

bath. p-toluenesulfonyl chloride (46.82 g, 0.245 mol) in 100 mL

of dichloromethane was added dropwise. The temperature of

reaction solution was slowly increased to room temperature

and the solution was stirred overnight. After the reaction, the

solution was poured into 1300 mL of 1 M HCl and the aqueous

phase is removed. Organic phase was washed with saturated

NaCl solution and dried over MgSO₄. The compound was puri-

ﬁed by column chromatography (ethyl acetate: hexane = 1:

1 v/v) (yield: 69.9 g, 77%). H-NMR (500 MHz, CDCl₃): δ/ppm

7.60 (d, J = 5 Hz, 2H, aromatic H), 7.18 (d, J = 5 Hz, 2H, aro-

matic H), 3.97 (t, 2H, S-O-CH₂), 3.58-3.31 (m, 12H, -O-CH₂-),

2.29 (t, 2H, -CH₂-COO), 2.25 (s, 3H, Ar-CH₃), 1.25 (s, 9H,

C(CH₃)₃). ¹³C-NMR (125 MHz, CDCl3): δ/ppm 21.8, 28.2, 36.4,

67.0, 68.8, 69.4, 70.5, 70.6, 70.7, 70.8, 80.6, 128.1, 130.0, 133.1,

144.9, 171.0. HRMS (C₂₀H₃₂O₈S+Na): calculated m/z of [M+Na]

455.1705; measured m/z 455.1714.

a yellow solid (74 mg, 80%). H-NMR (500 MHz, D₂O): δ/ppm

1,4-diiodo-2,5-bis(11-(tert-butoxycarbonyl)-3,6,9-trioxaun-

decyloxy)benzene (5): Compound 4 (11.28 g, 26.08 mmol),

compound 2 (3.93 g, 10.87 mmol), potassium iodide (0.018 g,

0.11 mmol), potassium carbonate (9 g, 65.22 mmol) and 30 mL

of 2-butanone were added to a 250 mL of two neck round-

bottomed ﬂask with condenser. Reaction solution was reﬂuxed

for 38 h and 2-butanone was evaporated at reduced pressure.

The crude mixture was suspended to methylene chloride and

washed with 1 M HCl. The organic layer was again washed with

saturated NaCl and dried over MgSO₄. Further puriﬁcation was

achieved by column chromatography on sllica gel (ethyl acetate:

hexanes = 7:3 v/v). Compound was again chromatographed on

silica gel (ethyl acetate: hexanes = 1:1 v/v) (yield: 4.26 g, 44%).

¹H-NMR (500 MHz, CDCl₃): δ/ppm 7.22 (s, 2H, aromatic), 4.15

(t, 4H, -OCH₂-), 3.87 (t, 4H, -OCH₂-), 3.8-3.6 (m, 20H, -OCH₂-),

7.60 (s, 2H, aromatic), 7.11 (s, 2H, aromatic), 4.13 (broad t, 4H,

-OCH₂-), 3.90-3.30 (broad m, 2OH, -OCH₂CH₂-), 3.15 (s, 6H,

-OCH₃), Molecular weight based on PS-GPC in THF before

hydrolysis of ethylhexylgroup M_n= 163 700, M_w= 624 600,

PDI = 3.82.

P2: Except M6 (85 mg, 95.4 μmol) instead of M5, the poly-

merization step was followed by synthetic route of P1 above. After

polymerization, polymer solution was centrifuged to get the

supernatant (3500 rpm). The supernatant solution was evapo-

rated and redissolved in 10 mL tetrahydrofuran and 10 mL

of 1 m NaOH solution. The solution was stirred overnight at

35 °C and evaporated at reduced pressure. The solution was

dissolved in deionized water and centrifuged to remove the

impurity insoluble to water. The water solution was dialyzed

against deionized water for 3 days. Centrifugation was again

1078 wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2012, 22, 1076–1086

conducted to get supernatant after dialysis. The polymer solu-

tion was lyophilized to yield a yellow solid (91 mg, 87%). H-

P5: M4 (60.6 mg, 78.7 μmol) and M6 (73.6 mg, 82.6 μmol)

were placed into a 50 mL of Schlenck ﬂask and DMF (2 mL)

and diisopropylamine (1 mL) were added to the reaction vessel.

After complete dissolution of two monomers, the solution

was degassed by three times of vacuum and argon purging.

In a separate Schlenck ﬂask, tetrakistriphenylphosphine pal-

ladium(0) (1 mol% of the monomer) and copper(I) iodide

(1 mol% of the monomer) were transferred under a nitrogen

atmosphere of a glove box and argon was purged in the Sch-

lenck ﬂask for 10 min. Two catalysts were dissolved in mor-

pholine (1 mL) and degassed by three times of vacuum and

argon purging. The degassed solution containing catalyst was

cannulated onto monomer solution. After transfer of the cata-

lyst solution to monomer solution, three cycles of degassing

to a polymer solution was ﬁnally done again. The polymer

solution was allowed to stir under argon purging at 55 °C for

2 days. The solvent was evaporated to dryness. The crude

polymer was redissolved in 50 mL of 1 m sodium hydroxide

solution and stirred under argon atmosphere at room tem-

perature for 1 h. Polymer solution was centrifuged and the

supernatant was dialyzed against deionized water for 2 days

(10 × 4 L water exchanges). The polymer solution was lyophi-

lized to yield a yellow waxy solid (77 mg, 67%). ¹H-NMR

(500 MHz, D₂O): δ/ppm 8.30 (broad s, 2H, -COOH), 7.25

(s, 2H, aromatic), 7.09 (s, 2H, aromatic), 4.16 (t, 4H, -OCH₂-),

4.08 (m, 2H, -OCH-), 3.8-3.2 (broad m, 80H, -OCH₂CH₂O),

3.15 (s, 12H, -OCH₃), 2.27 (s, 4H, -CH₂COO-), molecular

weight; M_nby ¹H-NMR end analysis = 14 200.

NMR (500 MHz, D₂O): δ/ppm 7.58 (s, 2H, aromatic), 7.12 (s,

2H, aromatic), 4.13 (m, 2H, -OCH-), 3.80-3.20 (broad m, 56H,

-OCH₂-), 3.11 (s, 12H, -OCH₃). GPC-based molecular weight

before the cleavage of protection group, M_n= 32,100 gmol⁻¹,

M_w= 105,900 gmol⁻¹, PDI = 3.3.

P3: A general procedure about polymerization is identical

to the method for P1. Monomer M2 (40.8 mg, 69.1 μmol),

monomer M6 (61.6 mg, 69.1 μmol), toluene (1.0 mL), and diiso-

propylamine (2 mL) are placed into a 50 mL of Schlenck ﬂask.

After complete dissolution of two monomers, the solution was

degassed by three times of vacuum and argon purging. In a sep-

arate Schlenck ﬂask, tetrakistriphenylphosphine palladium (0)

(5 mol% of the monomer) and copper (I) iodide (5 mol% of the

monomer) were transferred under a nitrogen atmosphere of a

glove box and argon was purged in the Schlenck ﬂask for 10 min.

Two catalysts were dissolved in toluene (1.0 mL) and degassed

by three times of vacuum and argon purging. The degassed

solution containing catalyst was cannulated onto monomer solu-

tion. After transfer of the catalyst solution to monomer solution,

three cycles of degassing to a polymer solution was ﬁnally done

again. The polymer solution was allowed to stir under argon

purging at 55 °C for 2 days. The reaction mixture was ﬁltered

with 0.45 micrometer membrane syringe. The mixture solution

was concentrated at reduced pressure and precipitated in dieth-

ylether (15 mL). The crude polymer was redissolved in 15 mL of

dioxane and the solution was mixed with 10% aqueous NaOH

solution (15 mL). Solution was stirred under argon atmosphere

at room temperature for 12 h. Polymer solution was centrifuged

and dialyzed against deionized water for 2 days (10 × 4 L water

exchanges). The polymer solution was lyophilized to yield a

P5-A: End-capping reaction was conducted in-situ after

polymerization of P5 was ﬁnished. 4-ethynylbenzoic acid

(11 mg, 79 μmol) as an end-capper was dissolved in DMF

(0.5 mL) and DIPA (0.2 mL). End-capper solution was degassed

and cannulated onto polymer solution. A trace amount of pal-

ladium catalyst and cupper iodide in DMF (0.5 mL) degassed

by vacuum and argon purging recycles was also added to

polymer solutions. The polymer solution was allowed to stir

under argon purging at 55 °C for an additional 24 h. After the

reaction, a work-up procedure for polymer recovery was same

yellow solid (51 mg, 60%). H-NMR (500 MHz, D₂O): δ/ppm

7.27 (s, 2H, aromatic), 7.15 (s, 2H, aromatic), 4.03 (broad m, 6H,

-CH₂CH₂O-, -OCH-), 3.81-3.21 (broad m, 56H, -OCH₂CH₂), 3.18

(broad s, 12H, -OCH₃), 2.25 (broad t, 4H, -CH₂CH₂COO-), 1.87

(broad m, 4H, -CH₂CH₂CH₂-), GPC (THF) based M_n= 73,100

gmol⁻¹, M_w= 214,200 gmol⁻¹, PDI = 2.93.

P4: Except M3 (41.85 mg, 62 μmol) instead of M2, the poly-

merization step was conducted by synthetic route of P3 above.

After the polymerization, polymer mixture was centrifuged to

get the supernatant (3500 rpm). The supernatant solution was

concentrated at reduced pressure, precipitated in ether, and

washed with acetone. The polymer was redissolved in 10 mL

tetrahydrofuran and 10 mL of 1 m sodium hydroxide solution.

The solution was stirred overnight at 35 °C and evaporated at

reduced pressure. The solution was dissolved in DI water and

centrifuged to remove the unknown impurity. The water solu-

tion was dialyzed against deionized water for 3 days. During

the dialysis, ﬁbril type aggregations observed due to the hydro-

phobic long alkyl chain and the protonation of carboxylic

group. The polymer solution was lyophilized to yield a yellow

solid (46 mg, 57%). A solid P4, of which a carboxylic group

as P5. Two new peaks at H-NMR analysis emerged at 7.78,

7.51 ppm corresponding to the aromatic protons of the end-

capper, conﬁrming that the carboxylic group was chemically

attached.

2.2. Discussion for Synthesis of Monomers and Polymers

Synthetic routes for the preparation of all monomers and CPEs

are described in Schemes 1 and 2. All polymers were prepared

by the palladium-catalyzed Sonogashira-Hagihara copolymeri-

zation method. At ﬁrst, we tried polymerization with a diiodo-

phenyl unit having unprotected free carboxylic acid. However,

reactions were not successful because the carboxylic group

in the ortho-position caused a side reaction during the poly-

merization and resulted in a low molecular weight.^[61–63]P1 and

P2 were prepared by the copolymerization of a diiodophenyl

monomer having carboxylic groups protected with ethylhexyl

side chains. After the polymerization, the ethylhexyl group was

hydrolyzed by base treatment to give a negatively charged car-

boxylate ion as a side chain to the polymer structure. A ﬂexible

is protonated, was completely soluble in water (pH = 8). H-

NMR (500 MHz, D₂O): δ/ppm 7.01 (s, 2H, aromatic), 6.75 (s,

2H, aromatic), 4.41 (m, 2H, -OCH-), 3.95 (t, 4H, -OCH₂-), 3.80-

3.23 (broad m, 56H, -OCH₂-), 3.15 (s, 12H, -OCH₃), 2.05 (t, 4H,

-CH₂COO-), 1.78-1.10 (broad m, 16H, -CH₂-), GPC (THF) based

M_n= 19 200 gmol⁻¹, M_w= 57 800 gmol⁻¹, PDI = 3.01.

Adv. Funct. Mater. 2012, 22, 1076–1086

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1079

showed a blue-shifted and much weaker

emission. Cleavage of the protecting group

from the carboxlic moiety (electron-with-

drawing group) will affect overall dipole

moment and charge density of the polymer

backbone (electron-donating group). The sol-

vent polarity also should play a role in the

blue shift.

The quantum yield of P1 was only 0.45%

in water. The solubility of P1 in water was

high enough to dissolve more than 1 mg of

P1 in 1 mL of deionized water. P1 solution

in water looked yellow and transparent in

visual observation. However, even though the

aqueous solution of P1 looked to be trans-

parent to the naked eye, our co-solvent study

and surfactant study strongly imply that P1

was aggregated in water. We examined the

photoluminescence properties of P1 in the

water/methanol co-solvent system. As shown

in Figure 3, the emission intensity of P1

increased as the volume fraction of methanol

was increased in the water/methanol mixture

because methanol is a better solvent than

water, suggesting P1 aggregation in water. To

further investigate the aggregation feature we

conducted a surfactant study by using sodium

dodecylsulfate (SDS, anionic), Tween20 (non-

ionic), and dodecyltrimethylammonium bro-

mide (DTAB, cationic) and investigated their

de-aggregation capability for P1 in water

(Figure 4).^[64–68]Fluorescence emission inten-

sity of P1 was enhanced as the surfactant concentration was

increased in all three cases likely due to de-aggregation of the

polymer aggregates induced by the surfactants. The difference

in the emission enhancement of P1 at a given concentration of

each surfactant indicates that the cationic surfactant DTAB most

Scheme 1. Synthesis of Monomers M1 to M4.

and hydrophilic ethylene oxide unit was also introduced in order

to give good water solubility to the hydrophobic polymer back-

bone by suppressing the hydrophobic aggregation. P3 and P4

were also prepared by the polymerization of a diiodo monomer

having ethyl-protected carboxylic groups to avoid the solubility

problem of the free carboxylic acid group in

organic solvents. Representative physical and

photophysical data of the CPEs described in

this contribution are summarized in Table 1.

All CPEs were dissolved in water and showed

blue-green emission having the emission

λ_maxof about 460 nm.

2.3. P1: Conventional CPE Design with Ionic

and Non-Ionic Water-Soluble Side Chains

P1 is designed to have alternating ionic and

non-ionic water-soluble side chains. This

molecular design strategy has been conven-

tionally used to make water-soluble CPEs

in the literature. The precursor polymer of

P1 before the deprotection of the carboxylic

acid group showed a well-deﬁned 0–0 emis-

sion band at 487 nm and the quantum yield

of 45% in chloroform (Figure 2). However,

emission of P1 in water after the deprotection Scheme 2. Polymer synthesis (P1–P5).

1080 wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2012, 22, 1076–1086

Table 1. Physical properties of all polymers used in this study.

poly

DP^b)

194

E_g/eV^d)

M_n/gmol^−1a)

λ_max,abs/nmλ_max,em/nm

Stokes shift cm^−1c)

4110

Φ_F(%, D₂O)^e)

163,700

384

456

368

460

421

463

406

464

412

457

2.14

0.45 ± 0.49

32,100

73,100

19,200

14,200

5430

2150

3080

2390

2.18

2.54

2.52

2.64

0.09 ± 0.02

31.6 ± 5.50

5.3 ± 0.55

P5 (or P5-A)

51.4 ± 9.55 (36.6 ± 4.37)

^a)Molecular weight of all polymers except P5 was measured by GPC before hydrolysis of the ethylhexyl protection. M_nfor P5 was done by ¹H-NMR end analysis in

D₂O; ^b)Degree of polymerization (DP) was calculated from the M_nand the molar mass of the repeat unit; ^c)The magnitude of the Stokes shift was calculated by Δ =

_max,em–λ_max,abs;

^d)The optical HOMO-LUMO energy gap is based on the low-energy onset in the solution-state UV/vis spectra; ^e)Quantum yield is absolute quantum value

measured by using an integrating sphere, polymer concentration: 1 mg/L.

the increase in the ﬂuorescence intensity of P1 with increasing

concentration of added DTAB was most signiﬁcant between

0.1 wt% of DTAB and 0.5 wt% of DTAB. Interestingly, our cal-

culation showed that 0.4 wt% of DTAB is required to make 1:1

charge complex with carboxylic acid groups of P1 as schemati-

cally illustrated in Figure 5.^[69]Distinct 0–0 and 0–1 emission

bands are observed in Figure 4(c) indicating that DTAB effec-

tively dissembles the P1 aggregates.^[48,53]

effectively dissembles P1 aggregates. Considering the fact that

P1 is a negatively charged CPE, cationic surfactants should be

more effective than nonionic and anionic surfactants. Note that

We investigated CPE aggregation in an aqueous environment

by means of various electron microscopic techniques.^[70–74]

Conventional TEM microscopy images of P1 shown in

Figure 6a revealed tree-like, fractal aggregation suggesting that

P1 was completely aggregated. The magniﬁed transition region

shows that rigid rod-like P1 chains aggregated to form cylin-

drical aggregates. A few single P1 chains could aggregate into

a ﬁbril by hydrophobic π–π stacking and several ﬁbrils could

agglomerate to form few tens of nanometers wide ﬁbers. How-

ever, this aggregated feature might be developed during the

drying process for the conventional TEM sample preparation

rather than representing the packing of P1 in water. Therefore,

we additionally conducted cryogenic TEM to observe an actual

morphology of P1 in aqueous environment. A P1 solution in

water was rapidly quenched in liquid ethane using cryo-plunge

and images were obtained as a vitriﬁed state below −170 °C.

Interestingly, we observed a self-assembled sheet-like structure

of P1 from the cryo-images (Figure 6b). Hydrophobic inter-

action between the rigid-rod polymer chains likely induces

the molecular packing and thin layer formation, followed by

aggregation formation as schematically suggested in Figure 6c.

We believed that the initially charged carboxylic acid group of

P1 was gradually protonated during the prolonged dialysis to

remove oligomers and excess ions, accelerating the polymer

aggregation as well. Therefore, even though P1 was modiﬁed

with water-soluble ionic and non-ionic side chains, P1 mole-

cules aggregate due to the rigidity and the hydrophobic nature

of the main chain, resulting in the extremely low quantum yield

of 0.45 in water.

Figure 2. UV and Photoluminescence spectra of P1 before (A, in chlo-

roform) and after (B, in water) the cleavage of the ethylhexyl protecting

group (P1 conc. = 5 mgl⁻¹).

Figure 3. Photoluminescence spectra of P1 in various water/methanol

mixture solvents (P1 conc. = 0.7 mgml⁻¹, excitation wavelength: 365 nm).

Adv. Funct. Mater. 2012, 22, 1076–1086

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1081

Figure 4. Photoluminescence proﬁle of P1 in water by adding different types of surfactants; a) SDS, negative, b) tween20, neutral, c) DTAB, positive

(P1 conc. = 5 mgml⁻¹).

2.4. P2: Preventing Aggregation by the Bulky Bifurcated

Ethylene Oxide Side Chains

while the P2 derivative having the ethyl protected carboxylic acid

side chains has 55.0% quantum yield in chloroform. Ionic-

pendant groups having the sodium counter ions directly

attached to the CPE backbone is hypothesized to be closely

related with photoluminescence quenching of CPE in water. As

supporting evidence, we found enhancement of emission inten-

sity of P1 and P2 in an acidic condition where the carboxylic

group should be protonated. It is believed that the observed

quenching is attributed to the photoinduced electron transfer

quenching mechanism as a result of the formation of electron

donor/acceptor charge transfer complex between the polymer

backbone and the directly connected ionic side chain.^[75,76]

We replaced the single strand ethylene oxide side chains of P1

with the bulky bifurcated ethylene oxide chain and prepared P2

to efﬁciently sheath the rigid hydrophobic CPE backbone and

minimize the π–π stacking (Figure 1).^[53,60]Initially, we meas-

ured the molecular weight of P2 by DMF-based GPC after the

cleavage of the carboxy-protecting group. However, the molecular

weight of P2 was measured up to a few millions (gmol⁻¹) likely

due to an incorrectly exaggerated hydrodynamic volume of the

polymer caused by the limited solubility and aggregation of P2

in DMF. The molecular weight of P2 before the cleavage of the

protection group measured by GPC was 32,100. The absorption

and emission spectra of P2 in water are presented in Figure 7.

We examined the effect of the bulky ethylene oxide side chains

in molecular aggregation by TEM and found that there was no

large aggregation like the one found from P1 even though P2

solution was dried during the conventional TEM sample prepa-

ration. Instead, spherical particles of a few tens of nanometers

in size were observed, suggesting that aggregation of CPE was

efﬁciently suppressed by the bulky nonionic ethylene oxide side

chains.^[53,60]However, surprisingly even the non-aggregated P2

in aqueous solution showed very low quantum yield of 0.9%

2.5. P3 and P4: Spacer Between the CPE Backbone and the Ionic

Moiety

We put an alkyl spacer between the CPE backbone and the

ionic pendant groups and prepared P3 to exam the hypoth-

esis that the direct connection of the ionic pendant group to

the CPE backbone causes the emission quenching. Figure 8

illustrates the absorption and the emission spectra of P3. The

emission spectrum of P3 shows a well-deﬁned 0–0 band at

the λ_maxof 463 nm. As we expected, the aqueous solution of

P3 has the quantum yield of 31.6% (1 mgL⁻¹) that is dramati-

cally improved from the 0.9% of the aqueous solution of P2,

strongly supporting that the directly attached carboxylic acid

groups to P2 backbone cause the quenching. We previously

prepared similar PPEs having sulfonate ionic groups via a

propyloxy linkage and bifurcated ethylene oxide side chains.

The sulfonated version of P3 had a high quantum efﬁciency

of 0.53 showing a good agreement with P3.^[11]Conventional

TEM microscopy images of 1 wt% P3 aqueous solution was

essentially identical to that of P2 showing spherical nanoparti-

cles of a few tens of nanometer.

We increased the length of the alkyl spacer from propyl (C₃)

to hexyl (C₆) and prepared P4 to test whether a long hydro-

phobic spacer would cause aggregation of CPE. Accordingly

the quantum yield of the aqueous solution of P4 signiﬁcantly

dropped down to 5.3%. Dialysis puriﬁcation of P4 also indi-

cated that the longer hexyl hydrophobic chain lowers the sol-

ubility of P4 in water. Conventional TEM microscopy images

(data not shown) show that P4 having the hexyl spacers are

more aggregated in water than P3 due to the long hydrophobic

Figure 5. Schematic illustration of surfactant effect on P1 in water.

1082 wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2012, 22, 1076–1086

Figure 6. TEM microscopy images of P1 in a) dried state and b) 1 w% water (arrows represent lacey carbons), c) proposed models for the aggregated

structures of P1 in air or water.

Figure 7. Absorption and Emission spectra of P2 (10 mgL⁻¹) in water

(excitation at 365 nm), excitation spectra of P2 was obtained corre-

sponding to the emission at 460 nm.

Figure 8. Normalized absorption (dotted) and emission (solid) spectra

of P3 (7 mgL⁻¹) in water (excitation at 365 nm).

alkyl spacers.^[77–84]We hardly observed any signiﬁcant aggre-

gation during the dialysis of P1, P2 and P3 in water. Under

visual observation, they remained soluble in deionized water

and the solubility exceeded approximately 1 mg mL⁻¹. How-

ever, protonation of carboxylic group of P4 during the dialysis

induced precipitation of P4, indicating that the long alkyl

spacers reduced the water-solubility of P4 compared to other

CPEs. After producing negative charges to P4

water (>10 mg/mL) and its solubility was independent to pH of

the aqueous solution. P5 was also soluble in a polar solvent such

as dimethyl sulfoxide and methanol and partially soluble in tet-

rahydrofuran. Photoluminescence spectrum of P5 in Figure 9(a)

is narrow with a well-deﬁned 0–0 band at 457 nm. P5 has the

highest ﬂuorescent emission quantum yield of 51.4% (1 mgL⁻¹)

among other CPEs and is over 110 times more emissive than P1.

in phosphate buffer (pH = 8) or in a slightly

basic aqueous solution, the solubility of P4 in

water was signiﬁcantly enhanced.

2.5. P5: Bulky Anionic Side Chain and Water-

Soluble Spacer for the Ionic Side Chain

We replaced the hydrophobic alkyl linker with

a hydrophilic ethylene oxide linker when we

synthesized P5 to prevent aggregation of CPE

Figure 9. a) UV and PL spectra (5 mgL⁻¹) of P5 and P5-A (Polymers were excited at 365 nm),

induced by the hydrophobic nature of the alkyl

linker unit. P5 completely dissolved in pure b) ﬂuorescence quantum yield of all polymers (P1 to P5) at various concentrations.

Adv. Funct. Mater. 2012, 22, 1076–1086

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1083

and as the concentration increased above

that regime the quantum yield decreased

showing the concentration dependent self-

quenching.

Finally, we prepared P5-A which has

two directly-attached carboxylic acid units

at the chain ends to investigate further the

inﬂuence of carboxylic acid group to the

emissive property of CPEs. In situ end-

capping reaction to P5 during polymeriza-

tion was undertaken by adding 4-ethynyl-

benzoic acid with additional palladium

catalyst (Scheme 3).^[11,85–87]As shown in

Figure 9a, the chain end modiﬁcation essen-

tially did not cause any spectral broadening

nor bathochromic shift, indicating minimal

inﬂuence on the solubility of P5-A and neg-

ligible aggregation of P5-A in water. P5-A

indeed showed a similar solubility in water

and identical TEM images as P5. How-

ever, the quantum yield of P5-A was largely

reduced to 31.6% that is 38.5% drop from

51.4% of P5, clearly demonstrating that a

directly-connected ionic group to the conju-

gated backbone of CPEs has a detrimental

Figure 10. Electron microscopy images of P5; a) SEM (SE2 mode, 30 kV) and b) transmission

mode (STEM) and c) conventional TEM images (200 kV). All samples were prepared with

1 wt% solution in water on a lacey carbon ﬁlm (indicated by the arrows in the ﬁgures) and

dried in air.

Electron microscopic analysis for P5 did not show any signif-

icant aggregation, indicating that the hydrophilic nature of the

side chain is necessary to prevent CPE aggregation in water. As

shown using the scanning transmission electron microscope

(STEM) and the conventional TEM (Figure 10), agglomera-

tion of P5 was noticeably suppressed while P1 or P4 showed a

micrometer-sized massive aggregation. From the conventional

TEM images one can apparently observe ﬁbrils of approxi-

mately 25 nm (Figure 10c). Molecular mechanics calculations

carried out using MM2 force ﬁeld predicted that a hydrophobic

backbone of P5 would be wrapped by the hydrophillic side

chains and the resulting thickness of the polymer chain would

be 2.4 nm. Therefore, the observed ﬁbrils are likely composed

of only a few of polymer chains even after water evaporated

slowly during the TEM sample preparation. These results lead

to a reasonable conclusion that hardly any π–π stacking among

P5 forms in water. We also additionally carried out cryogenic

TEM of P5 in aqueous environment to further investigate the

correlation between the nature of the side chain and the aggre-

gation behavior of CPE. A spider web-like entangled structure

formed by the rigid rod-like P5 chains shown in Figure 11

clearly indicates discrete polymer chains without agglomera-

tion throughout the whole area. The thickness of the polymer

chain in the cyro-TEM image is about 7 nm, that is, thicker

than 2.4 nm predicted by molecular modeling. The clumps

found along the polymer chain in the middle ﬁgure are ice

crystals formed during the cryo-TEM sample preparation. The

cryo-TEM analysis convincingly reveals the non-aggregated

morphology of P5 in water.

effect to the emission property of CPEs.^[88]

3. Conclusion

We systematically investigated the effect of the chemical nature,

shape, and the length of the ionic and nonionic side chains

on the water solubility and quantum yield of conjugated poly-

electrolytes by means of electron microscopy and spectroscopic

analysis. Simple ionic and anionic decoration of CPE did not

warrant good water-solubility because of the rigid and hydro-

phobic nature of the conjugated backbone of CPEs. TEM

analysis combined with quantum yield measurement reveals

that unless CPEs are modiﬁed with bulky hydrophilic ethylene

oxide side chains, CPEs form aggregates in water and conse-

quent ﬂuorescence quenching. Carboxylic acid group, which

is a commonly used and convenient functional group for

We additionally investigated the concentration dependence

of the quantum yield of the CPEs. As shown in Figure 9b, the

quantum yield of CPEs in water has the same trend of other

ﬂuorophores in organic solvents. They mostly show the highest

ﬂuorescence efﬁciency at submicromolar concentration regime

Figure 11. Cryo-TEM images of P5-A (1 wt% in water, applied voltage:

120 kV). The sample was prepared with 1 wt% solution in water on a

formvar-coated grid, stabilized with evaporated carbon ﬁlm. Massive

lumps (arrows) in the left ﬁgure are ice crystals.

1084 wileyonlinelibrary.com

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Adv. Funct. Mater. 2012, 22, 1076–1086

to enhance contrast. Sample preparation for Figure

10 was done in a similar way and SEM and TEM

images were respectively obtained using Ultra55 Field

Emission Scanning Electron Microscope (FESEM,

Zeiss, 30 kV) and JEOL 2100 TEM (acceleration

voltage: 200 kV). For cryo-TEM images, a carbon

coated ﬁlm on a broken pattern consists of woven-

mesh-like holes (300 mesh) or a formvar coated grid,

stabilized with evaporated carbon ﬁlm were purchased

from Electron Microscopy Sciences, Inc (PA, USA)

and a cryoplunge (Gatan, Inc, CA, USA) was used for

polymer sample preparation. The specimen grid was

clamped between the tips of plunging tweezers and

3 μL of polymer solution (1 w% or 0.1 wt% in water)

was blotted to grid for production of thin, aqueous

ﬁlm. It is then plunged into a temperature controlled

ethane bath (<−170 °C) and polymer solution is

rapidly frozen as vitriﬁed ice. Sample was stored in liquid nitrogen until

use. The grid was transferred to the TEM for unstained, in situ observation.

Images were taken in the bright-ﬁeld mode with JEOL 2100 transmission

electron microscope at 120 kV accelerating voltage.

Scheme 3. In situ end-capping reaction for P5-A.

bioconjugation, turned out to have a detrimental inﬂuence on

the emissive property of CPE when it is connected directly to

the CPE backbone. Placing a spacer linker between carboxylic

acid and the CPE backbone solved the quenching problem.

However, the nature and the length of the spacer group also

largely inﬂuence on the water-solubility of CPEs. When the alkyl

linker was long, the hydrophobic nature of the linker induced

self-assembled aggregates. The presented results reveal the

effects of the side chain design on the water-solubility and the

consequent emissive property of CPEs and provide a molecular

design principle to achieve highly emissive, completely water-

soluble, and conveniently functionalized CPEs.

Acknowledgements

This work is supported by the National Science Foundation (Career DMR

064486). We greatly thank Professor Darrin Pochan and Tuna Yucel for

valuable provisions and discussions of preliminary TEM images.

Received: August 26, 2011

Revised: October 11, 2011

Published online: January 9, 2012

4. Experimental Section

[1] M. Pinto, K. S. Schanze, Synthesis 2002, 9, 1293–1309.

[2] H. Jiang, P. Taranekar, J. R. Reynolds, K. S. Schanze, Angew. Chem.

Int. Ed. 2009, 48, 4300–4316.

[3] F. Feng, H. Wang, L. Han, S. Wang, J. Am. Chem. Soc. 2008, 130,

11338–11343.

[4] H. Li, R. Yang, G. C. Bazan, Macromolecules 2008, 41, 1531–1536.

[5] M. Kang, O. K. Nag, R. R. Nayak, S. Hwang, H. Suh, H. Y. Woo,

Macromolecules 2009, 42, 2708–2714.

[6] G. C. Bazan, S. Wang, L. Liu, L. An, Angew, Chem. Int. Ed. 2009, 48,

4372–4375.

[7] H.-A. Ho, M. Béra-Abérem, M. Leclerc, Chem. Eur. J. 2005, 11,

1718–1724 and references therein.

[8] K. P. R. Nilsson, O. Inganäs, Nat. Mater. 2003, 2, 419–424.

[9] B. Liu, G. C. Bazan, Chem. Mater. 2004, 16, 4467–4476 and references therein.

[10] C. J. Yang, M. Pinto, K. Schanze, W. Tan, Angew. Chem. Int. Ed. 2005,

44, 2572–2576.

[11] K. Lee, L. K. Povlich, J. Kim, Adv. Funct. Mater. 2007, 17, 2580–2587.

[12] B. Liu, G. C. Bazan, Proc. Natl. Acad. Sci. USA. 2005, 102, 589–593.

[13] K. Lee, J.-M. Rouillard, T. Pham, E. Gulari, J. Kim, Angew. Chem. Int.

Ed. 2007, 46, 4667–4670.

[14] C.-C. Pun, K. Lee, H.-J. Kim, J. Kim, Macromolecules. 2006, 39, 7461.

[15] M. B. Abérem, A. Najari, H.-A. Ho, J.-F. Gravel, P. Nobert,

D. Boudreau, M. Leclerc, Adv. Mater. 2006, 18, 2703–2707.

[16] D. Yu, Y. Zhang, B. Liu, Macromolecules 2008, 41, 4003–4011.

[17] K.-Y. Pu, B. Liu, Macromolecules 2008, 41, 6636–6640.

[18] F. Feng, Y. Tang, S. Wang, Y. Li, D. Zhu, Angew. Chem. Int. Ed. 2007,

46, 7882–7886.

Materials and Methods: All chemicals were purchased from Sigma-

Aldrich, Inc. or Acros Organics, Inc. and used without further puriﬁcation.

Compounds 1,^[89,90]2,^[91,92]M5^[50]and M6^[50,60,93]in Schemes 1 and 2 are

prepared according to the literature published previously. All polymers

(P1 to P5 and P5-A) were puriﬁed by dialysis against deionized water

(molecular weight cut-off: 12,000–14,000 gmol⁻¹) for 3 days, lyophilized

to dry the polymer, and stored in the dried state at 4 °C. The molecular

weight of all CPEs except P5 was determined by GPC with the polystyrene

references in THF before the cleavage of the ethylhexyl protecting group

of carboxylic acid. Due to the limited solubility of P5 in THF, its number-

averaged molecular weight (M_n) was calculated by ¹H NMR end-

group analysis. NMR characterization was done with Varian Inova 500

(500 MHz, 11.7 T, Tin)

Photophysical Experiments: UV/vis absorption spectra of the CPE

solutions were obtained on a Cary UV50 UV/Vis spectrometer (Varian,

Inc.). Steady-state ﬂuorescence spectra of the CPEs were recorded on

a PTI QuantaMaster spectroﬂuorometer™. The molar concentration

of the CPE solutions was determined based on the repeat unit of the

CPEs. Corrected ﬂuorescence spectra were obtained for variations in

photomultiplier response over wavelength using correction curves

generated on the instrument. The ﬂuorescence spectra were normalized

by the optical density corresponding to the highest ﬂuorescence intensity.

The quantum yields of the CPEs in various concentrations were measured

by exciting the CPEs at 365 nm in deionized water using an integrating

sphere attached to the PTI QuantaMaster spectroﬂuorometer.

Electron Microscopy Analysis: A copper TEM grid coated with a 20–30 nm

ﬁlm of pure carbon (purchased from Electron Microscopy Sciences, PA,

USA) was held at the tip of a tweezer. A small drop of aqueous CPE

solution was placed on the grid to form a bead. An excess samples in

water were blotted off by touching the grid with a ﬁlter paper and the

sample was left for drying. Images were taken in the bright-ﬁeld mode

with a Tecnai G2 12 Twin transmission electron microscope at 120 kV

accelerating voltage. Structures were imaged at slight underfocus in order

[19] C. Fan, K. W. Plaxco, A. J. Heeger, J. Am. Chem. Soc. 2002, 124, 5642–5643.

[20] L. Chen, D. W. McBranch, H.-L. Wang, R. Helgeson, F. Wudl,

D. G. Whitten, Proc. Natl. Acad. Sci. USA. 1999, 96, 12287–12292.

[21] K. P. R. Nilsson, A. Herland, P. Hammerström, O. Inganäs, Bio-

chemistry 2005, 44, 3718–3724.

Adv. Funct. Mater. 2012, 22, 1076–1086

2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1085