PolymerChemistry

PAPER

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article OnlineView Journal | View Issue

Synthesis and pro

Beijing National Laboratory for Molecula

Chemistry and Physics of Ministry of Educa

Engineering, Peking University, Beijing 100

zshen@pku.edu.cn

† Electronic supplementary informationsynthesis of PPLG-g-PMPCS bottlebrushespectra, tting to the SAXS curves,10.1039/c4py00033a

Cite this: Polym. Chem., 2014, 5, 4948

Received 10th January 2014Accepted 20th April 2014

DOI: 10.1039/c4py00033a

www.rsc.org/polymers

4948 | Polym. Chem., 2014, 5, 4948–4

perties of a rod-g-rod bottlebrushwith a semirigid mesogen-jacketed polymer as theside chain†

Hai-Jian Tian, Wei Qu, Yu-Feng Zhu, Zhihao Shen* and Xing-He Fan*

A series of bottlebrushes, poly(g-3-azidopropanyl-L-glutamate)-g-poly{2,5-bis[(4-methoxyphenyl)-

oxycarbonyl]styrene (PPLG-g-PMPCS), with an a-helical rigid backbone and mesogen-jacketed rigid side

chains were synthesized by the “grafting onto” method from yne-terminated PMPCS (side chain) and

poly(g-azide propanyl-L-glutamate) (backbone) by a Cu(I)-catalyzed 1,3-dipolar cycloaddition click

reaction. The chemical structures of the bottlebrushes were confirmed by 1H NMR, gel permeation

chromatography (GPC), GPC coupled with multi-angle laser light scattering and differential refractive

index detection, and Fourier-transform infrared spectroscopy. The bottlebrushes were obtained with high

grafting densities (>85%). The properties of the bottlebrushes were studied by small-angle X-ray scattering

and atomic force microscopy. And the results show that the conformation of the bottlebrush changes

from a cylinder to an ellipsoid as the degree of polymerization of the side chain increases from 14 to 65.

Introduction

Bottlebrushes are densely graed polymers, with propertiesdistinctly different from those of sparsely graed or linearpolymers.1–4 Bulky side chains are packed around the backbone,and the steric effect between the neighboring side chains forcesthe backbone to adopt a stretched conformation. Even with aexible backbone and exible side chains, a bottlebrush showsa rigid or worm-like conformation.5,6 When their structures areprecisely tailored, bottlebrushes can be used to form nano-wires7,8 and nanotubes.9–11 Because of their complex structures,they were once difficult to synthesize. With the development ofliving and controlled polymerization and the “click” strategy,now there are mainly three ways to synthesize a bottlebrush:“graing onto”, “graing through”, and “graing from”.1–4

With these methods, different bottlebrushes with a exiblebackbone and exible side chains (coil-g-coil),5,6 with a rigidbackbone and exible side chains (rod-g-coil),12–14 and with aexible backbone and rigid side chains (coil-g-rod)15 have beensynthesized. However, bottlebrushes with a rigid backbone andrigid side chains are rare16 because it is difficult to synthesizerigid chains by controlled and living polymerization. With arigid backbone, the bottlebrush will be rigid even if the side

r Sciences, Key Laboratory of Polymer

tion, College of Chemistry and Molecular

871, China. E-mail: fanxh@pku.edu.cn;

(ESI) available: The details for thes, 1H NMR results, GPC curves, FT-IRand derived parameters. See DOI:

956

chains are not so bulky; and with rigid side chains, the rigidityof the side chains will affect the main chain, making it stiffer.15

Therefore, rod-g-rod bottlebrushes may have the advantage offorming ordered nanostructures with large dimensions, servingas potential candidates for nano-porous materials, photoniccrystals, and so on.

Polypeptides are rigid polymers owing to their ability to formsecondary structures (e.g., a-helix and b-sheet).17 In addition,they can be prepared by controlled ring-opening polymerization(ROP) of a-amino acid N-carboxyanhydrides (NCAs).18,19 Thesetwo factors make polypeptides attractive for building bottle-brushes. Bottlebrushes with polypeptides as backbones aresynthesized by “graing onto”12,20–23 and “graing from”

methods.13,24,25 In those studies, side-chain degree of polymer-ization (DP)-dependent polymer properties, such as secondarystructures23 and thermo-responsive and self-assemblingbehaviors,24 were studied. In addition to polypeptides, otherrigid polymers like poly(p-phenylene terephthalamide)26 orpolyphenylene27 can also be used as bottlebrush backbones.However, these polymers cannot be obtained by controlledpolymerization, which limits their studies.

On the other hand, mesogen-jacketed liquid crystallinepolymers (MJLCPs) are also rigid polymers which have beenextensively studied by our group.28,29 The origin of their rigidityis similar to that of bottlebrushes. Because in general bulky sidegroups (oen mesogenic ones) are laterally attached directly tothe backbone, the repulsion between neighboring side groupsalso forces the backbone to be extended.30,31 Dimensions ofMJLCPs can be readily tuned by varying the mesogenic groupsand molecular weights (MWs). Compared with other rigidpolymers, MJLCPs can be easily prepared by controlled radical

This journal is © The Royal Society of Chemistry 2014

Paper Polymer Chemistry

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

polymerization reactions. Therefore, they are excellent rod-likecandidates for preparing rigid chains. A coil-g-rod bottlebrushwith an MJLCP as the side chain and polystyrene as the back-bone has been previously reported by our group.32

Herein, we report a rod-g-rod bottlebrush with poly(g-azidepropyl-L-glutamic acid) (PPLG) as the backbone and poly{2,5-bis[(4-methoxyphenyl)oxycarbonyl]styrene} (PMPCS) as the sidechain. By combining atom transfer radical polymerization(ATRP), ROP of NCA, and a Cu(I)-catalyzed click reaction, wesuccessfully synthesized the bottlebrush PPLG-g-PMPCS withdifferent side-chain lengths via the “graing onto” strategy.Chemical structures of the bottlebrushes were characterized by1H NMR, Fourier-transform infrared spectroscopy (FT-IR), gelpermeation chromatography (GPC), GPC coupled with multi-angle laser light scattering and differential refractive indexdetection (GPC-MALLS-DRI), and the physical properties wereexamined by small-angle X-ray scattering (SAXS) and atomicforce microscopy (AFM).

ExperimentalMaterials

2-Bromo-2-methylpropionyl bromide (98%, Acros), propargylalcohol (99%, Aldrich), 1-chloro-3-hydroxypropane (98%, TCI),hexamethyldisilazane (HMDS, 98%, Acros), N,N,N0,N0,N0 0-pentamethyldiethylenetriamine (PMDETA, 98%, TCI), and tri-ethylamine (98%, Acros) were used as received. N,N-Dime-thylformamide (DMF), tetrahydrofuran (THF), andchlorobenzene were used aer passing through activatedcolumns under an argon atmosphere (M.Braun, Inc.). All otherreagents were purchased from Beijing Chemical Reagents andused as received.

Methods1H NMR spectra were recorded on a Bruker AV-400 spectrom-eter. Matrix-assisted laser desorption/ionization-time of ightmass spectrometry (MALDI-TOF MS) measurements were per-formed on a Bruker Autoex high-resolution tandem massspectrometer. GPC experiments were conducted on a Waters2410 instrument equipped with a Waters 2410 refractive index(RI) detector. THF with a ow rate of 1.0 mL min�1 was used asthe eluent. The absolute MW of poly(g-3-chloropropanyl-L-glutamate) (PCPLG) was determined by GPC-MALLS-DRI at 50�C on an SSI pump connected to Wyatt Optilab DSP and WyattDAWN EOS light scattering detectors. DMF containing 0.02 MLiBr with a ow rate of 1.0 mL min�1 was used as the eluent.Absolute MWs of linear PMPCS and the bottlebrushes weredetermined by GPC-MALLS-DRI at 35 �C. THF with a ow rate of1.0 mL min�1 was used as the eluent. Refractive index (RI)increments (dn/dc, using 1.0, 2.0, 3.0, 4.0, and 5.0 mg mL�1)were measured off-line at ambient temperature at 658 nm usingthe Optilab Rex interferometeric refractometer. For PCPLG in0.02 M LiBr–DMF, dn/dc¼ 0.0704 mL g�1 (which was 0.0685 mLg�1 in 0.01 M LiBr–DMF22). For the bottlebrush in THF, dn/dc ¼0.177 mL g�1, slightly larger than that of linear PMPCS (dn/dc ¼0.172 mL g�1).33 FT-IR analysis was carried out using a Thermo

This journal is © The Royal Society of Chemistry 2014

Scientic Nicolet iN10-MX spectrophotometer between 4000and 400 cm�1 at a spectral resolution of 4 cm�1, and thenumber of scans was 16. SAXS measurements were performedon the beamline BL16B1 using a three-slit system at theShanghai Synchrotron Radiation Facility. The wavelength of theX-ray was 0.124 nm. The two-dimensional (2D) SAXS patternswere recorded by using a Mar165 CCD at a resolution of 2048 �2048 pixels and a pixel size of 80 mm � 80 mm for quantitativeanalysis. The sample-to-detector distance was 5248 mm forSAXS experiments (calibrated by beef tendon standard). AFMimages were obtained under ambient conditions with a Multi-mode Nanoscope IIIa atomic force microscope (VeecoMetrology Group) in tapping mode. The polymers were trans-ferred onto a PMPCS-modied silicon wafer32 by the Langmuir–Blodgett technology.

Synthesis of the bottlebrushes

The PPLG-g-PMPCS bottlebrushes were successfully synthesizedby combining ATRP, ROP of NCA, and Cu(I)-catalyzed clickreaction (Scheme 1) through the “graing onto” strategy. Thedetailed synthesis routes are described as follows.

Synthesis of azide-substituted poly(L-glutamate) (PAPLG, 5)

Azide-substituted poly(L-glutamate) was synthesized fromg-chloropropanyl-L-glutamate by the method described in theliterature.22 Its synthesis method and the polymer properties aredescribed in the ESI.†

Synthesis of a-alkyne PMPCS (7)

PMPCS is a typical MJLCP, which has been systematicallystudied by our group.34,35 With an alkynyl-substituted ATRPinitiator (6), a-alkyne PMPCS was easily obtained.32,36 Thesynthesis method and the polymer properties are described inthe ESI.†

Synthesis of poly(g-3-azidopropanyl-L-glutamate)-g-poly{2,5-bis[(4-methoxyphenyl)oxycarbonyl]styrene} (PPLG-g-PMPCS)by the click reaction

A typical click procedure is described as follows. a-Yne-termi-nated PMPCS (7, 500 mg, Mn ¼ 9.8 � 103 g mol�1, 51 mmol ynegroups), PAPLG (5, 7.2 mg, 34 mmol azide groups), and CuBr (4.8mg, 34 mmol) were introduced into a Schlenk ask and dis-solved in DMF (30 mL). The Schlenk ask was degassed by threefreeze–pump–thaw cycles, sealed under vacuum, and thenimmerged into an oil bath thermostated at 80 �C for 2 days. Thesolution was then cooled to ambient temperature and passedthrough a neutral alumina column to remove the Cu(I) catalyst.The raw product was precipitated in methanol.

Reprecipitation

In a general procedure, 500 mg of the raw bottlebrush wasdissolved in 20 mL of CH2Cl2. An equal volume of acetone wasadded to the solution. Methanol (20 mL) was added dropwise tothe solution until it was turbid. The suspension was collected by

Polym. Chem., 2014, 5, 4948–4956 | 4949

Scheme 1 Synthesis pathways of the backbone, side chains, and the bottlebrushes.

Polymer Chemistry Paper

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

centrifugation. This procedure was repeated twice to remove theunreacted PMPCS (52% yield).

Fig. 1 GPC traces of a-alkyne PMPCS: PMPCS14 (1), PMPCS24 (2),PMPCS36 (3), PMPCS55 (4), and PMPCS64 (5).

Results and discussionSynthesis of the bottlebrushes via the “graing onto” strategy

There are mainly three ways to synthesize a bottlebrush:“graing onto”, “graing through”, and “graing from”.Compared to the other two strategies, the “graing onto”strategy has the advantages over the control of the backbonelength, side-chain length, and polydispersity. The only problemis the control of the graing density. Following the developmentof the click reaction, this strategy leading to high graingdensities becomes popular.

The backbone (PAPLG) and side chains (PMPCS) weresynthesized separately. Azide and yne groups are needed in theCu(I)-catalyzed 1,3-dipolar cycloaddition click reaction.Compared with yne-ended g-propargyl-L-glutamate which is aliquid,12 g-chloropropanyl-L-glutamate (2) is a solid at ambienttemperature, which makes purication and the next step (ROPof NCA) easier. HMDS was used as the initiator in the controlledROP of the N-carboxyanhydride based on g-3-chloropropanyl-L-glutamic acid (CPLG-NCA). PCPLG with an absolute MW of25.9 � 103 g mol�1 and a low polydispersity index (PDI) of 1.04

4950 | Polym. Chem., 2014, 5, 4948–4956

was obtained (Fig. S3 in the ESI†). Treated with NaN3 in DMF at60 �C for two days, PCPLG was converted to PAPLG. The 1HNMR results demonstrate a quantitative conversion (Fig. S4 inthe ESI,† peak a). The FT-IR spectrum of PAPLG (Fig. S5 in the

This journal is © The Royal Society of Chemistry 2014

Table 1 Polymerization results and MW characteristics of a-alkyne PMPCS samples

Entry [M]0/[I]0 Temperature (�C) Time (h) Mna (103 g mol�1) PDIa Mw

b (103 g mol�1) DP

1 20 80 2 3.8 1.05 5.7 142 30 80 6 6.5 1.07 9.8 243 50 80 6 9.2 1.10 14.1 364 75 90 12 13.2 1.09 21.9 555 90 100 12 15.0 1.10 26.3 64

a Determined by GPC (calibrated with polystyrene standards). b Determined by GPC-MALLS-DRI (THF) with dn/dc ¼ 0.172, revised according to theresults from MALDI-TOF.

Paper Polymer Chemistry

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

ESI,† nN]N]N ¼ 2095 cm�1) reveals the presence of the azidegroup.

a-Alkyne PMPCS was synthesized by ATRP. With precisecontrol of the feeding ratio, time, and temperature, a series ofa-alkyne PMPCS samples with different MWs were synthesized.GPC traces are shown in Fig. 1. Table 1 summarizes the poly-merization results. As in our previous work,36 the absolute MWof linear PMPCS is approximately 1.52 times the value ofnumber-average MW (Mn) from GPC. In this study, when theMW is relatively small, this relationship is true (entries 1–3).However, as the MW becomes larger, this relationship may nothold (entries 4 and 5). Therefore, we chose GPC-MALLS-DRI todetermine the absolute MWs.

In the synthesis of bottlebrushes, the same PCPLG backbone(absolute Mw ¼ 25.9 � 103 g mol�1, PDI ¼ 1.04) was used. a-Alkyne PMPCS was attached to the backbone by copper-cata-lyzed alkyne–azide [2 + 3] 1,3-dipolar cycloaddition. In order toobtain a high graing density, the ratio of alkyne to azidegroups was set to 1.5 : 1. Bottlebrushes were more difficult todissolve in acetone compared to linear PMPCS, which madepurication easier. The successful synthesis of the bottle-brushes was veried by GPC (Fig. 2). 1H NMR result shows thatthe signal of the alkyne group disappears aer the click reaction(Fig. S6 in the ESI†). The molecular weighs were measured byGPC-MALLS-DRI (Table 2).

In our previous work,32 bottlebrushes with a polystyrenebackbone showed a downward trend of graing density (from

Fig. 2 Representative GPC traces of PMPCS14 (1), the blend after theclick reaction (2), and the bottlebrush PPLG126-g-PMPCS14 (3).Because PAPLG forms aggregates in THF, the GPC curve of PAPLG isnot shown.

This journal is © The Royal Society of Chemistry 2014

92% to 75%) as the MW of the backbone is increased, probablybecause the azide groups are embedded in the coil PS chainsand it is difficult for them to be accessed for the click reactionwhen the graing density is increased. As for a helical poly-peptide, the azide groups outside the helix are more accessiblefor yne-endcapped side chains to react. In this work, the gra-ing density is in the range of 88–95%.

Because the content of the backbone is quite low in thebottlebrush, traditional characterization methods such as 1HNMR (Fig. S6 in the ESI†) and FT-IR (Fig. S7 in the ESI†) showlittle or no difference between the side-chain polymers andbottlebrushes, which makes the characterization challenging.

Structural studies by SAXS

Scattering studies, including laser light scattering,15,37,38

SAXS,15,38,39 and small-angle neutron scattering (SANS),37,40–42

and AFM23,43 are usually used to study the nanostructures ofbottlebrushes. In order to study the sizes and shapes of thebottlebrushes, SAXS measurements were carried out on solu-tions of the bottlebrushes in THF at a concentration of 1 mgmL�1 at 25 �C. As shown in Fig. 3, SAXS analysis was performedover a q (scattering vector) range of 0.08 to approximately0.8 nm�1, which provides information on the overall size(q ¼ 0.08–0.2 nm�1) as well as the cross-sectional size andparticle shape (q ¼ 0.2–0.8 nm�1).

We rst utilized model-independent methods to analyze theSAXS data. According to Guinier,44 the size parameter, radius ofgyration (Rg), can be extracted directly from the Guinier plot(Fig. S8 in the ESI†) at small q values following eqn (1):

ln½DIðqÞ� ¼ ln½a0� � 1

3Rg

2q2 (1)

And for bottlebrushes, which are cylindrical particles, thecross-sectional radius of gyration (Rg,c) can also be determinedby eqn (2):45

ln½qDIðqÞ� ¼ ln½b0� � 1

2Rg;c

2q2 (2)

here, DI(q) is the experimental excess scattering intensity. Dataare calculated using primus in the ATSAS soware package.46

The results are summarized in Table 3.Guinier plots can give information about the overall or

cross-sectional size. As the side-chain DP increases, Rg and

Polym. Chem., 2014, 5, 4948–4956 | 4951

Table 2 Compositional parameters of PPLG-g-PMCPS bottlebrushes

Bottlebrusha Mw,side chainb (103 g mol�1) PDIc DPside chain Mw,bottlebrush

d (106 g mol�1) PDId Ygrae (%)

PPLG126-g-PMPCS14 5.7 1.05 14 0.67 1.06 90PPLG126-g-PMPCS24 9.8 1.07 24 1.20 1.02 95PPLG126-g-PMPCS36 14.1 1.10 36 1.60 1.01 87PPLG126-g-PMPCS55 21.9 1.09 55 2.49 1.03 88PPLG126-g-PMPCS64 26.3 1.10 64 2.99 1.03 90

a Synthesized with PAPLG126 and PMPCS of different lengths. b Determined by GPC-MALLS-DRI (THF) with dn/dc ¼ 0.172, revised according toresults from MALDI-TOF MS. c Determined by GPC (calibrated with polystyrene standards). d Determined by GPC-MALLS-DRI (THF) with dn/dc¼ 0.177. e Graing density Ygra ¼ [(Mw,bottlebrush � Mw,backbone)/Mw,side chain]/DPbackbone.

Fig. 3 Double-logarithmic plots of the SAXS profiles of the bottle-brushes with side chains of different DPs.

Table 3 Parameters of the bottlebrushes obtained from Guinier plots

Bottlebrush Rg (nm) Rg,c (nm) Rca (nm)

PPLG126-g-PMPCS14 7.96 3.73 5.27PPLG126-g-PMPCS24 8.56 4.71 6.66PPLG126-g-PMPCS36 9.49 5.98 8.46PPLG126-g-PMPCS55 10.60 7.60 10.75PPLG126-g-PMPCS64 11.20 7.94 11.23

a Rc is cross-sectional radius of the bottlebrush. For a circularhomogeneous cylinder, Rc ¼

ffiffiffi2

pRg;c:

Fig. 4 Calculated PDDFs for the bottlebrushes with side chains ofdifferent DPs.

Fig. 5 Rg values calculated from various methods for the bottle-brushes with side chains of different DPs.

Polymer Chemistry Paper

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

Rg,c increase. However, from the Guinier plots, little orno information about the particle structure can beobtained. Another model-independent analysis, indirectFourier transform (IFT), gives the pair-distance-distributionfunctions (PDDFs, p(r)'s). The PDDF proles giveindications about the particle shape: spherical, prolate, oroblate. In addition, radius of gyration, Rg, can also be givenby eqn (3):

Rg;IFT ¼ 1

2

ðDmax

0

pðrÞr2drðDmax

0

pðrÞdr(3)

4952 | Polym. Chem., 2014, 5, 4948–4956

PDDFs are calculated using GNOM in the ATASA sowarepackage47 (Fig. 4 and Table S1 in the ESI†). The obtainedasymmetrical PDDF plots indicate elongated shapes. Withincreasing side-chain DP, the largest dimension increasesslightly (Table S1 in the ESI†), and the asymmetry of the plotdecreases, which means that the particle becomes less elon-gated. Rg,IFT values calculated from PDDFs are also presented inTable S1 in the ESI.† Rg,IFT increases as the side-chain DPincreases.

This journal is © The Royal Society of Chemistry 2014

Fig. 6 Rc values calculated from various methods for the bottle-brushes with side chains of different DPs, with the solid lines repre-senting a power-law fit to the data.

Fig. 7 Rc values from the cylinder model and the WLC model for thebottlebrushes with side chains of different DPs. Data are comparedwith those from bottlebrushes having flexible side chains, PNb-g-PS,from the literature.41

Fig. 8 Length of the cylinder, Lcyl, according to the cylinder model forthe bottlebrushes with side chains of different DPs.

Fig. 9 Cross-sectional radius (Rc), the ratio between the two ellip-soidal axes (v), and asphericity parameter (as) of particles for thebottlebrushes with side chains of different DPs according to theellipsoid model.

Paper Polymer Chemistry

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

To further study the shapes of the bottlebrushes, model-dependent methods are used to analyze the form factors. Threedifferent models, cylinder, ellipsoid, and wormlike chain (WLC)models, are used to t the SAXS proles and obtain the formfactors. The ttings were done by using the SASt sowarepackage.40,42 From the tting results, particle sizes, shapes, andchain stiffness can be estimated (see details in the ESI†). Theresults are comparable with those from the Guinier plots andthe PDDFs.

Fig. 5 shows the Rg values obtained from differentmethods. There is no distinct difference between thesemethods. As the side-chain DP increases, the size of thebottlebrush increases. The cross-sectional radius Rc is asso-ciated with the length of the side chain. With the increasingside-chain DP, Ns, a power-law dependence, Rc z Ns

x, isfound (Fig. 6), where x ¼ 0.44–0.52 for different methods. Thevalue of x is consistent with those in the literature,37,41 whichsuggests a three-dimensional self-avoiding walk conforma-tion of the bottlebrush side chains.48

This journal is © The Royal Society of Chemistry 2014

Compared with those from bottlebrushes having exible PSside chains41 (Fig. 7), Rc values of bottlebrushes withMJLCP sidechains are almost two times larger at the same side-chain DP.For short side chains, the linkage between side chains and thebackbone may play a large role. However, when the side-chainDP is large enough (Ns > 50), the linkage can be ignored (a 9-atom spacer for PNb-g-PS, and 14-atom for PPLG-g-PMPCS).This dramatic difference in the cross-sectional radius is causedby the rigid structure of PMPCS, which has an extended-chainconformation because of the bulky mesogenic group.

From the cylinder model, we can evaluate the particle length,which can be used to estimate the length of the backbone.37 Asthe side-chain DP increases, the length of the cylinder (Lcyl)increases slightly (Fig. 8). Such an increasemay be caused by theside chains near the ends or the increased steric effect caused byneighboring side chains of larger DPs. The average length of thecylinder is Lm ¼ 27.1 nm. Because the number-average DP forthe polypeptide backbone is Nm ¼ 126, the average length permonomeric unit lm¼ Lm/Nm¼ 27.1/126 nm¼ 0.215 nm. For the

Polym. Chem., 2014, 5, 4948–4956 | 4953

Fig. 10 Values of persistence length, lp, according to the wormlikechain model for the bottlebrushes with side chains of different DPs.

Polymer Chemistry Paper

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

polypeptide, this value suggests a stretched helicalconformation.23

The ellipsoid model gives the information about the cross-sectional radius Rc and the ratio between the two ellipsoidalaxes (v). As the side-chain DP increases, Rc increases, and vdecreases (Fig. 9). We also calculated the asphericity parameter(as), which is used to quantify the deviation from the sphericalshape (as can be any value between 0 and 1, with 0 representinga perfect sphere and 1 representing an innitely long rod).49 Inthis work, 0.2 < as < 0.7, indicating that the bottlebrushes arenot perfect spheres, nor are they perfect long cylinders. As theside-chain DP increases, as decreases, which means that theparticle looks more like a sphere, in other words, less elongated.

Fig. 11 AFM images of the bottlebrushes with side chains of differentPMPCS36; d: PPLG126-g-PMPCS55; e: PPLG126-g-PMPCS64) and height p

4954 | Polym. Chem., 2014, 5, 4948–4956

This is consistent with the conclusion from the PDDF resultsand in good agreement with the work by Cheng andcoworkers.40 They used SANS to study the structure of poly-styrene graed with oligo(ethylene glycol). When the DP of thebackbone increases, polymer conformation changes from anellipsoid to a rigid cylinder, then to a wormlike chain. In thiswork, with increasing side-chain DP, conformational change ofnanorods is opposite. A similar result was also reported recentlyby Pesek and coworkers.41

As the side-chain DP increases, the stiffness characterized bythe persistence length (lp) of the backbone increases (Fig. 10).The lp values studied here are rather small compared withresults in the literature,37,48 smaller than that of poly(g-benzyl-a-L-glutamate),50 which has the same backbone as the bottle-brushes in this work. However, there are also several studiesreporting small lp values of bottlebrushes in the literature.41,42

Here, the bottlebrushes were synthesized from the “graingonto” method, and the graing density cannot reach 100%.Therefore, the steric effect caused by the neighboring sidechains may not be large enough. And the linkage between sidechains and the backbonemay also relieve the steric constraint.41

In addition, the bottlebrushes studied here have a backbone ofa modest length (Lc ¼ 45 nm) with a rather long contour length(Rc ¼ 5–10 nm). Thus, whether the WLC model is suitable hereis still doubtful.

Morphological studies by AFM

Besides SAXS, AFM is also frequently used to study bottle-brushes, especially for imaging.43 AFM was also used to imagethe rod-g-rod bottlebrushes. PMPCS-graed bottlebrushes

DPs (a: PPLG126-g-PMPCS14; b: PPLG126-g-PMPCS24; c: PPLG126-g-rofile of PPLG126-g-PMPCS24 (f).

This journal is © The Royal Society of Chemistry 2014

Table 4 Sizes of the bottlebrushes obtained from AFM results

Bottlebrush ha (nm) Eb (nm) Lsc (nm) ls

d (nm) Lme (nm) lm

f (nm)

PPLG126-g-PMPCS14 2.1 5.4 — — — —PPLG126-g-PMPCS24 2.4 5.7 10 0.21 29 0.23PPLG126-g-PMPCS36 2.5 5.8 15 0.21 29 0.23PPLG126-g-PMPCS55 2.6 5.9 24 0.22 30 0.24PPLG126-g-PMPCS64 3.0 6.2 28 0.22 30 0.24

a Heights of the bottlebrushes. b Errors caused by the tip size, determined from E ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiHð2Rc �HÞp

; 51 where Rc is the tip radius, Rc¼ 8 nm, andH isthe height of the sample. c Length of the side chains, calculated using Ls¼ L� 2E, L is the value directly measured from the AFM images. d ls ¼ L/Ns.e Lengths of the bottlebrushes. f lm ¼ L/Nm, Nm ¼ 126.

Paper Polymer Chemistry

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

prefer to aggregate on the unmodied silicon wafer.32 In orderto obtain a unimolecular image, all the silicon wafers used weregraed with PMPCS homopolymers. Sample images (Fig. 11)show a monolayer lm on silicon wafers obtained by theLangmuir–Blodgett technology. Uniform nanoscaled rods witha length of about 25 nm are packed densely on the surface. Asthe backbone length is xed, the nanorods show a similarlength. From the AFM images, we can obtain the length of thecylinder, shown in Table 4. The average length per monomericunit (lm) can be calculated, and lm ¼ 0.23–0.24 nm, consistentwith values in the literature.23 However, lm is larger than that insolution. For PPLG126-g-PMPCS24, the cross-sectional radiusfrom the cylinder model, Rc, is 6.46 nm in the THF solution.When the bottlebrush is adsorbed on the silicon wafer, theheight is almost 2.5 nm (Fig. 11f), much smaller than 2Rc(12.9 nm), whichmeans that the side chains are attened on thesurface. This adds to the steric effect and makes the backbonemore stretched, and the average length per monomeric unitbecomes larger. Furthermore, the structure of the bottlebrushchanges from a thin rod to a round ellipsoid as the side-chainDP increases (from Fig. 11a–e).

Conclusions

In summary, a series of bottlebrushes containing a rod backboneand rod side chains, PPLG-g-PMPCS, with different side-chain DPswere successfully synthesized by the combination of ATRP, ROP,and click reaction. Structures of the bottlebrushes were studied bySAXS and AFM. The results demonstrate a transformation from arod to an ellipsoid as the side-chain DP increases. Although thestiffness of this kind of bottlebrushes is not as high as one wouldexpect because the graing density is not 100% and there is aexible linkage between the side chains and the backbone, thesebottlebrushes with rigid PMPCS side chains have longer side-chain lengths compared to those having exible polystyrene sidechains with the same side-chain DP, which may be used to buildnano-porous materials with large pore sizes.

Acknowledgements

This research was supported by the National Natural ScienceFoundation of China (Grants 21134001 and 20990232). Theauthors also thank the Shanghai Synchrotron Radiation Facilityfor assistance during the SAXS experiments.

This journal is © The Royal Society of Chemistry 2014

Notes and references

1 M. F. Zhang and A. H. E. Muller, J. Polym. Sci., Part A: Polym.Chem., 2005, 43, 3461–3481.

2 S. S. Sheiko, B. S. Sumerlin and K. Matyjaszewski, Prog.Polym. Sci., 2008, 33, 759–785.

3 I. I. Potemkin and V. V. Palyulin, Polym. Sci. Ser. A, 2009, 51,123–149.

4 C. Feng, Y. Li, D. Yang, J. Hu, X. Zhang and X. Huang, Chem.Soc. Rev., 2011, 40, 1282–1295.

5 K. Terao, Y. Takeo, M. Tazaki, Y. Nakamura and T. Norisuye,Polym. J., 1999, 31, 193–198.

6 K. Terao, Y. Nakamura and T. Norisuye, Macromolecules,1999, 32, 711–716.

7 M. F. Zhang, C. Estournes, W. Bietsch and A. H. E. Muller,Adv. Funct. Mater., 2004, 14, 871–882.

8 J. Y. Yuan, Y. Y. Xu, A. Walther, S. Bolisetty, M. Schumacher,H. Schmalz, M. Ballauff and A. H. E. Muller, Nat. Mater.,2008, 7, 718–722.

9 K. Huang and J. Rzayev, J. Am. Chem. Soc., 2009, 131, 6880–6885.

10 K. Huang, D. P. Canterbury and J. Rzayev, Macromolecules,2010, 43, 6632–6638.

11 K. Huang and J. Rzayev, J. Am. Chem. Soc., 2011, 133, 16726–16729.

12 A. C. Engler, H.-i. Lee and P. T. Hammond, Angew. Chem., Int.Ed., 2009, 48, 9334–9338.

13 Y. Liu, P. Chen and Z. B. Li,Macromol. Rapid Commun., 2012,33, 287–295.

14 H. Y. Tang and D. H. Zhang, Polym. Chem., 2011, 2, 1542–1551.

15 Y. Saito, L. T. N. Lien, Y. J. Jinbo, J. Kumaki, A. Narumi andS. Kawaguchi, Polym. J., 2013, 45, 193–201.

16 H. Enomoto, B. Nottelet, S. A. Halifa, C. Enjalbal, M. Dupre,J. Tailhades, J. Coudane, G. Subra, J. Martinez andM. Amblard, Chem. Commun., 2013, 49, 409–411.

17 P. Papadopoulos, G. Floudas, H. A. Klok, I. Schnell andT. Pakula, Biomacromolecules, 2004, 5, 81–91.

18 H. Lu and J. J. Cheng, J. Am. Chem. Soc., 2007, 129, 14114–14115.

19 H. Lu and J. J. Cheng, J. Am. Chem. Soc., 2008, 130, 12562–12563.

20 A. C. Engler, A. Shukla, S. Puranam, H. G. Buss, N. Jreige andP. T. Hammond, Biomacromolecules, 2011, 12, 1666–1674.

Polym. Chem., 2014, 5, 4948–4956 | 4955

Polymer Chemistry Paper

Publ

ishe

d on

22

Apr

il 20

14. D

ownl

oade

d by

Bro

wn

Uni

vers

ity o

n 31

/10/

2014

01:

16:2

9.

View Article Online

21 C. Xiao, C. Zhao, P. He, Z. Tang, X. Chen and X. Jing,Macromol. Rapid Commun., 2010, 31, 991–997.

22 H. Y. Tang and D. H. Zhang, Biomacromolecules, 2010, 11,1585–1592.

23 H. Tang, Y. Li, S. H. Lahasky, S. S. Sheiko and D. Zhang,Macromolecules, 2011, 44, 1491–1499.

24 J. X. Ding, C. S. Xiao, L. Zhao, Y. L. Cheng, L. L. Ma,Z. H. Tang, X. L. Zhuang and X. S. Chen, J. Polym. Sci., PartA: Polym. Chem., 2011, 49, 2665–2676.

25 J. Ding, C. Xiao, Z. Tang, X. Zhuang and X. Chen, Macromol.Biosci., 2011, 11, 192–198.

26 S. Du, J. Zhang, Y. Guan and X. Wan, Aust. J. Chem., 2014, 67,39–48.

27 S. Zhou, Y. Zhou, H. Tian, Y. Zhu, Y. Pan, F. Zhou, Q. Zhang,Z. Shen and X. Fan, J. Polym. Sci., Part A: Polym. Chem., 2014,52, 1519–1524.

28 Q. F. Zhou, H. M. Li and X. D. Feng, Macromolecules, 1987,20, 233–234.

29 X.-F. Chen, Z. Shen, X.-H. Wan, X.-H. Fan, E.-Q. Chen, Y. Maand Q.-F. Zhou, Chem. Soc. Rev., 2010, 39, 3072–3101.

30 Q. F. Zhou, X. L. Zhu and Z. Q. Wen, Macromolecules, 1989,22, 491–493.

31 X. H. Wan, F. Zhang, P. Q. Wu, D. Zhang, X. D. Feng andQ. F. Zhou, Macromol. Symp., 1995, 96, 207–218.

32 Z. Ma, C. Zheng, Z. Shen, D. Liang and X. Fan, J. Polym. Sci.,Part A: Polym. Chem., 2012, 50, 918–926.

33 Q. W. Pan, L. C. Gao, X. F. Chen, X. H. Fan and Q. F. Zhou,Macromolecules, 2007, 40, 4887–4894.

34 D. Zhang, Y. X. Liu, X. H. Wan and Q. F. Zhou,Macromolecules, 1999, 32, 5183–5185.

35 C. Ye, H. L. Zhang, Y. Huang, E. Q. Chen, Y. L. Lu, D. Y. Shen,X. H. Wan, Z. H. Shen, S. Z. D. Cheng and Q. F. Zhou,Macromolecules, 2004, 37, 7188–7196.

4956 | Polym. Chem., 2014, 5, 4948–4956

36 Q.-H. Zhou, J.-K. Zheng, Z. Shen, X.-H. Fan, X.-F. Chen andQ.-F. Zhou, Macromolecules, 2010, 43, 5637–5646.

37 B. Zhang, F. Groehn, J. S. Pedersen, K. Fischer andM. Schmidt, Macromolecules, 2006, 39, 8440–8450.

38 M. Kikuchi, L. T. N. Lien, A. Narumi, Y. Jinbo, Y. Izumi,K. Nagai and S. Kawaguchi, Macromolecules, 2008, 41,6564–6572.

39 K. Amitani, K. Terao, Y. Nakamura and T. Norisuye, Polym. J.,2005, 37, 324–331.

40 G. Cheng, Y. B. Melnichenko, G. D. Wignall, F. Hua, K. Hongand J. W. Mays, Macromolecules, 2008, 41, 9831–9836.

41 S. L. Pesek, X. Li, B. Hammouda, K. Hong and R. Verduzco,Macromolecules, 2013, 46, 6998–7005.

42 M. A. J. Gillissen, T. Terashima, E. W. Meijer,A. R. A. Palmans and I. K. Voets, Macromolecules, 2013, 46,4120–4125.

43 S. S. Sheiko, F. C. Sun, A. Randall, D. Shirvanyants,M. Rubinstein, H. Lee and K. Matyjaszewski, Nature, 2006,440, 191–194.

44 A. Guinier and A. Fournet, Small angle scattering of X-rays,Wiley & Sons, New York, 1955.

45 O. Glatter and O. Kratky, Small Angle X-Ray Scattering,Academic Press, London, 1982.

46 P. V. Konarev, V. V. Volkov, A. V. Sokolova, M. H. J. Koch andD. I. Svergun, J. Appl. Crystallogr., 2003, 36, 1277–1282.

47 D. I. Svergun, J. Appl. Crystallogr., 1992, 25, 495–503.48 S. Rathgeber, T. Pakula, A. Wilk, K. Matyjaszewski, H.-i. Lee

and K. L. Beers, Polymer, 2006, 47, 7318–7327.49 P. Englebienne, P. A. J. Hilbers, E. W.Meijer, T. F. A. De Greef

and A. J. Markvoort, So Matter, 2012, 8, 7610–7616.50 E. Temyanko, P. S. Russo and H. Ricks, Macromolecules,

2001, 34, 582–586.51 S. Y. Fung, C. Keyes, J. Duhamel and P. Chen, Biophys. J.,

2003, 85, 537–548.

This journal is © The Royal Society of Chemistry 2014