Synthesis and characterization of cyclic brush-like polymers by N-heterocyclic carbene-mediated zwitterionic polymerization of N-propargyl N-carboxyanhydride and the grafting-to approach

Article abstract of DOI:10.1021/ma201948u

Cyclic brush-like polymers were synthesized by tandem organo-mediated zwitterionic polymerization and a grafting-to approach. The cyclic polymer backbone, consisting of poly(N-propargylglycine) (c-PNPG), was synthesized by an N-heterocyclic carbene (NHC)-mediated zwitterionic ring-opening polymerization of N-propargyl N-carboxyanhydride. The polymerization proceeds in a quasi-living manner, allowing access to c-PNPG of well-defined chain length. The cyclic architecture of the polymers was verified by size exclusion chromatography (SEC) and mass spectroscopy (MS), as well as scanning probe characterization. Poly(ethylene glycol) functionalized with azido end-groups was subsequently grafted onto the c-PNPG by the copper-mediated azide/alkyne cycloaddition reaction (CuAAC). The side chain grafting density was determined by ¹H NMR spectroscopy and SEC analysis. The grafting efficiency is low (<19%) when the cyclic backbone is comprised of a c-PNPG homopolymer. The efficiency can be significantly improved (up to 93%) by utilizing cyclic poly(N-propargylglycine)-ran-poly(N-butylglycine) random copolymers (c-PNPG-r-PNBG). This has been attributed to the ease of access to the propargyl groups in c-PNPG and c-PNPG-ran-PNBG: the strong tendency of the former to aggregate in common organic solvents (including the CuAAC reaction medium) restricts access to the propargyl groups.

Full text of DOI:10.1021/ma201948u

ARTICLE
pubs.acs.org/Macromolecules
Synthesis and Characterization of Cyclic Brush-Like Polymers by
N-Heterocyclic Carbene-Mediated Zwitterionic Polymerization of
N-Propargyl N-Carboxyanhydride and the Grafting-to Approach
Samuel H. Lahasky, Wilson K. Serem, Li Guo, Jayne C. Garno, and Donghui Zhang*
Department of Chemistry and Macromolecular Studies Group, Louisiana State University, Baton Rouge, Louisiana, 70803
S Supporting Information
b
ABSTRACT: Cyclic brush-like polymers were synthesized by tandem
organo-mediated zwitterionic polymerization and a grafting-to approach.
The cyclic polymer backbone, consisting of poly(N-propargylglycine) (c-
PNPG), was synthesized by an N-heterocyclic carbene (NHC)-mediated
zwitterionic ring-opening polymerization of N-propargyl N-carboxyanhy-
dride. The polymerization proceeds in a quasi-living manner, allowing access
to c-PNPG of well-defined chain length. The cyclic architecture of the
polymers was verified by size exclusion chromatography (SEC) and mass
spectroscopy (MS), as well as scanning probe characterization. Poly(ethylene glycol) functionalized with azido end-groups was
subsequently grafted onto the c-PNPG by the copper-mediated azide/alkyne cycloaddition reaction (CuAAC). The side chain
grafting density was determined by 1H NMR spectroscopy and SEC analysis. The grafting efficiency is low (<19%) when the cyclic
backbone is comprised of a c-PNPG homopolymer. The efficiency can be significantly improved (up to 93%) by utilizing cyclic poly(N-
propargylglycine)-ran-poly(N-butylglycine) random copolymers (c-PNPG-r-PNBG). This has been attributed to the ease of access to
the propargyl groups in c-PNPG and c-PNPG-ran-PNBG: the strong tendency of the former to aggregate in common organic solvents
(including the CuAAC reaction medium) restricts access to the propargyl groups.
’ INTRODUCTION
synthesis of cyclic polymer brushes bearing randomly grafted
polystyrene (PS) and polyisoprene (PI) side chains. The cyclic
polymer backbone was prepared by end-to-end coupling of an
αꢀω heterofunctional linear precursor, and the PS/PI side
chains were statistically grafted to the backbone in one step.17
While the majority of the polymer brushes are cyclic, linear,
tadpole or ∞-shape backbones have been evidenced by atomic
force microscopy (AFM).17,18 These architectural contaminants
arise from the limitations of synthetic methods used to prepare
the cyclic polymer backbone. Cyclic brush-like polymers can be
synthesized by a grafting-from approach, where PS brushes
are grown from a cyclic poly(ethylene glycol) backbone via
nitroxide-mediated radical polymerization (NMP), as shown by
Jia et al.19 Coulembier et al. reported the synthesis of jellyfish
macromolecular architectures by a grafting-through approach
where a macromonomer (i.e., poly(methyl methacrylate)-modified
L-lactide) was copolymerized with L-lactide in the presence of
in situ generated N-heterocyclic carbene.20 More recently, Xia
et al. and Zhang et al. have synthesized macrocyclic brushes by
ring expansion metathesis polymerization of functionalized
norbornenes.21 In the former report the side chains were
installed prior to polymerization as part of the macromonomers,
whereas in the latter case the side chains were attached by
Brush-like polymers, which consist of a backbone and a bristle,
represent an interesting polymeric architecture.1 Attachment or
growth of the bristles increases the conformational strain of the
backbone2 and accordingly increases the Kuhn length and
reduces the chain entanglement.3,4 As a result, brush-like polymers
have been investigated as precursors of shape-persistent organic
nanotubes,5 lubricants,6 or self-assembled one-dimensional
photonic crystals.7,8 Brush-like polymers can be synthesized by
several synthetic strategies8 including the grafting-through,9 the
grafting-from,10 and the grafting-to approaches.11 The grafting-to
approach is particularly versatile regarding the chemistry of brush-
like polymers. As the backbone and side chains are independently
prepared prior to coupling, their composition can be precisely
characterized. The drawback of this method is that the grafting
density is often limited by the steric congestion of the polymeric
side chains.
The grafting efficiency is also significantly impacted by the
method used to couple the side chains to the polymer backbone.12
In this regard, click chemistry, particularly the copper-mediated
alkyne/azide cycloaddition reaction (CuAAC),13 has become
increasingly employed in the synthesis of brush-like polymers14
due to its high efficiency and stoichiometric bond formation
under mild conditions.11,15
Most synthetic brush-like polymers feature a linear polymer
backbone. Molecular architectures such as cyclic or star-shapes
have received less attention.16 Schappacher et al. reported the
Received: August 25, 2011
Revised:
October 13, 2011
Published: November 07, 2011
r
2011 American Chemical Society
9063
dx.doi.org/10.1021/ma201948u Macromolecules 2011, 44, 9063–9074
|

Macromolecules
ARTICLE
CuAAC post polymerization. While the cyclic polymer backbone
itself has random coil conformations, the grafting of polymeric
side chains rigidifies the backbone, resulting in shape-persistent
ring-like nanostructures. Many of these polymers exhibit intrigu-
ing solution and self-assembly behaviors and have potential uses
in nanotechnology and biomedical sciences. Before these appli-
cations can be realized, it is important to develop robust and
efficient synthetic routes toward these materials. Special atten-
tion needs to be paid to the main challenge in cyclic brush-like
polymer synthesis: the construction of the cyclic backbone
architecture.22
few suitable monomer structures. One strategy to overcome this
challenge is to develop new monomers that are amenable to REP
and contain side chains that are readily derivatized post-polym-
erization. Polymerization of these monomers will lead to cyclic
polymers with diverse structures.
We have recently reported that an NHC mediates the
zwitterionic ring-opening polymerization of N-substituted
N-carboxyanhydrides to yield cyclic poly(N-substituted glycine)
[aka poly(α-peptoids)].29h,i The reaction proceeds in a quasi-
living manner and enables the controlled synthesis of the cyclic
poly(α-peptoid) homopolymers or block copolymers. In the
zwitterionic polymerizations, the N-substituent has been limited
to inert aliphatic29h or aryl groups29i to ensure living polymer-
ization. This significantly limits the structural diversity of the
poly(α-peptoid)s. With demonstrated biocompatibility and en-
hanced proteolytic stability,30,31 poly(α-peptoid)s are potentially
useful for various biotechnological applications (e.g., drug deliv-
ery, and bioactive coatings). For these applications, it is impor-
tant that differing functionalities can be readily installed on the
side chains so as to confer the desired properties. In this work, we
extend the methodology toward the polymerization of N-pro-
pargyl NCA to yield cyclic poly(N-propargyl glycine) (c-PNPG)
and examine its copolymerization with N-butyl NCA to produce
cyclic poly(N-propargyl glycine)-ran-poly(N-butyl glycine) ran-
dom copolymers (c-PNPG-r-PNBG). We demonstrate that the
propargyl side chain can be readily functionalized with an azido-
terminated poly(ethylene glycol) by CuAAC chemistry, produc-
ing a water-soluble cyclic brush-like polymer.
Cyclic polymers are typically synthesized by (1) the end-to-
end coupling of a linear precursor, (2) ringꢀchain equilibrium,
or (3) ring-expansion polymerization. In the first method, two
chain ends are either homo- or heterofunctionalized to enable
intramolecular coupling under conditions of high dilution. While
this approach is versatile as it allows a variety of linear polymeric
precursors to be synthesized by controlled polymerizations (e.g.,
ATRP, RAFT), additional chain-end derivation is often required
to install the desired functionalities for coupling reactions. This
limitation, in addition to the need for highly dilute conditions,
makes this approach impractical for large scale synthesis. Recent
developments in using highly efficient coupling chemistry (i.e.,
CuAAC or thiolꢀene chemistry) under dynamic dilution con-
ditions has dramatically improved the synthetic ease and
efficiency.23,24 Preassociation of linear chain-ends by electrostatic
interaction or micelle formation have also reduced the need for
high dilution.25 However, cyclic polymers prepared by this
method are typically limited to low to moderate molecular
weight. Cyclization of high molecular weight linear precursors
often yields topological contaminants such as knots or catenates.
The second method, ringꢀchain equilibrium, involves a
competition between the propagation of a linear polymer and
chain cyclization through backbiting of the chain ends into the
growing chain. The ringꢀchain equilibrium cyclization is usually
limited to ring-opening polymerization and thermodynamically
controlled step-growth polymerization.26 While the total percen-
tage of cyclic species can be increased by high dilution, smaller
rings are still thermodynamically favored.27 Kricheldorf and
Schwarz demonstrated that, in kinetically controlled step-growth
polymerizations, cyclic polymers are the stable end-products.28
However, the polymers tend to exhibit bimodal mass distribution
where small rings are inevitably present.
’ EXPERIMENTAL SECTION
Materials. Glyoxylic acid monohydrate (98%), progargyl amine
(98%), butyl amine (98%), di-tert-butyl dicarbonate (97%), triethyl-
amine, copper(I) bromide, PMDETA, and phosphorus trichloride were
purchased from Sigma-Aldrich and used as received. All solvents used in
this study were purchased from Sigma-Aldrich and purified by passing
through alumina columns under argon. The compounds 2,6-diisopropyl-
phenylimidazol-2-ylidene (NHC)32 and azido terminated poly(ethylene
glycol) (PEG-N3) (Mn = 2 kg molꢀ1, PDI = 1.03; Mn = 500 g molꢀ1
,
3
3
PDI = 1.05) were synthesized by reported procedures.33 Butylamine was
stirred over CaH2 overnight and distilled under vacuum prior to use as an
initiator. The compounds of N-propargyl NCA (M1) and N-butyl NCA
(M2) were synthesized by adapting a literature procedure.29h
Instrumentation. The 1H and 13C{1H} NMR spectra were
recorded on a Bruker AV-400 spectrometer, and the chemical shifts
were referenced in parts per million (ppm) relative to proton impurities
or 13C isotopes of CDCl3 respectively. The FTIR spectra were collected
on a Bruker Tensor 27 FTIR spectrometer. The ESI spectra were
recorded on an Agilent 6710 TOF mass spectrometer in the positive ion
mode. The SEC analyses were conducted using an Agilent 1200 system
(Agilent 1200 series degasser, isocratic pump, auto sampler and column
heater) equipped with three Phenomenex 5 μm, 300 ꢁ 7.8 mm columns
[100 Å, 1000 Å and Linear(2)], Wyatt DAWN EOS multiangle light
scattering (MALS) detector (GaAs 30 mW laser at λ = 690 nm), Wyatt
ViscoStar viscometry (VISC) detector and Wyatt Optilab rEX differ-
ential refractive index (DRI) detector with a 690 nm light source.
DMF containing 0.1 M LiBr was used as the eluent at a flow rate of
Recent developments in the third method, ring-expansion
polymerization (REP), have enabled the conversion of small
cyclic substrates into cyclic polymers having moderate to high
molecular weight.29 The cyclic polymers obtained are of high
architectural purity and their synthesis does not require high
dilution conditions. For example, Bielawski et al. have reported
the REP of cyclooctene using a cycloalkylidenyl-ruthenium
catalyst/initatior to generate the cyclic poly(octane), which upon
hydrogenation yields high MW cyclic polyethylene (Mn
∼
several million dalton).29a Li et al. have demonstrated the
synthesis of cyclic polycaprolactone by REP of γ-caprolactone
with a cyclic stannane catalyst/initiator.29b Herbert et al. re-
ported a Lewis base-mediated ring-expansion polymerization of
silicon-bridged [1]ferrocenophanes, resulting in high molecular
weight cyclic metallopolymers.29c Waymouth, Hedrick and co-
workers demonstrated that N-heterocyclic carbenes (NHC) can
mediate the REP of cyclic esters to yield their respective cyclic
polyesters.29dꢀg The limitations of the REP approach are that it
requires a specially designed initiator/catalyst and that there are
0.5 mL minꢀ1. The column temperature was 50 °C and the detector
3
temperature was 25 °C. All data analyses were performed using Wyatt
Astra V 5.3 software. Polymer molecular weight (Mn) and molecular
weight distribution (PDI) were obtained by two methods: (1) Zimm
model fit of MALS-DRI data; (2) conventional SEC analysis with
a calibration curve. The calibration curve was constructed from
9064
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
Scheme 1
and linear poly(N-propargylglycine) samples (Mn = 9.3 kg molꢀ1
,
3
PDI = 1.55) were prepared by mixing the four pauci-disperse polymers
with different molecular weight in equal weight fractions. The poly-
disperse samples were then analyzed by SECꢀMALSꢀVISCꢀDRI for
their intrinsic viscosities ([η]) and the absolute molecular weights.
Atomic Force Microscopy (AFM). Imaging of the cyclic brush-
like polymers was accomplished using tapping mode AFM (Agilent 5500
AFM/SPM system) in ambient air with Picoscan v5.3.3 software with
probes acquired from Vista probes. The driving frequency for the tip
during the imaging of the polymers was 181 kHz. Polymer samples were
dissolved in chloroform to make a final concentration of 0.02 mg mLꢀ1
.
3
A volume of polymer solution (∼15 μL) was drop-deposited and dried
on freshly cleaved mica (0001) in ambient conditions for 24 h before
AFM imaging. Minimal processing of the images was done using
Picoscan software from Agilent.
Synthesis of 2-(Prop-2-yn-1-ylamino)acetic Acid Hydro-
chloric Salt (1). Propargyl amine (5.0 g, 90.8 mmol) and glycoxylic
acid (16.72 g, 225 mmol) were both dissolved in CH2Cl2 (230 mL) and
allowed to react overnight at room temperature. The CH2Cl2 was
removed under reduced pressure and aqueous HCl (137 mL, 137 mmol,
1.0 M) was added. The solution was heated at reflux for 24 h, after which
the water was removed by rotary evaporation. The resulting solid was
redissolved in methanol and precipitated by the addition of copious
volumes of ether. The product was collected by filtration and dried under
vacuum to yield a brown solid (8.55 g, 63% yield). 1H NMR (400 MHz,
D2O), δ: 10.03 (s, ꢀOH,1H),4.04(d,ꢀCH2,2H),3.87(t,ꢀCH2CH, 2H),
2.35 (t, ꢀCH, 1H). 13C {1H} NMR (100 MHz, D2O) δ: 173.8 (ꢀCOꢀ),
82.9 (ꢀCt), 73.3 (tCH), 50.4 (NHCH2CO), 40.1 (tCCH2NHꢀ).
Synthesis of 2-((tert-Butoxycarbonyl)(prop-2-yn-1-yl)-
amino)acetic Acid (2). The compound 1 (6.0 g, 40.1 mmol) was
dissolved in distilled water (135 mL), to which di-tert-butyl dicarbonate
(29.3 g, 134 mmol) and triethylamine (37.4 mL, 268 mmol) were
sequentially added. The reaction mixture was stirred overnight at room
temperature and then washed with hexane to remove any unreacted
di-tert-butyl dicarbonate. The aqueous phase was separated and made
acidic (pH = 3) with 1 N HCl(aq). The product was extracted into ethyl
acetate (3 ꢁ 100 mL) and the organic layer was combined and washed
with a saturated aqueous NaCl solution, dried over anhydrous
MgSO4, and concentrated to afford a beige solid. 1H NMR (400 MHz,
CDCl3), δ: 4.13 (t, ꢀCH2ꢀ, 2H), 4.07 (s, ꢀCH2ꢀ, 2H), 2.33 (t, ꢀCH,
1H), 1.46 (d, ꢀ(CH3)3, 9H). 13C {1H} NMR (100 MHz, CDCl3) δ:
173.4 (ꢀCOOH), 154.8 (ꢀNCOOꢀ), 80.5 (ꢀOC(CH3)3), 78.9
(tCꢀ), 73.0 (tCH), 54.5 (ꢀNCH2COOH), 40.1 (tCCH2Nꢀ),
28.0 (ꢀ(CH3)3).
Figure 1. (A) 1H NMR and (B) 13C {1H} NMR spectra of N-propargyl
NCA (M1) in CDCl3.
23 pauci-disperse polystyrene standards (Mn = 590 g molꢀ1 to 1472
3
kg molꢀ1, Polymer Laboratories, Inc.) using Astra's column calibration
3
template. Relative Mn and PDI were then calculated using Astra’s conven-
tional calibration template. Dynamic light scattering (DLS) analysis was
conducted on a Malvern Zetasizer Nano-ZS instrument while using the
Zetasizer software version 6.12. The c-PNPG polymer solution was prepared
by filtering through a 0.2 μm PTFE filter prior to DLS data collection.
Refractive Index Increment (dn/dc) Measurement. The
refractive index increment (dn/dc) of the synthesized polymers was
measured using Wyatt’s rEX DRI detector and Astra software dn/dc
template. Six polymer/DMF/0.1 M LiBr solutions with different and
precise concentrations of polymer were sequentially injected into the
DRI detector. The measured refractive index values were plotted versus
concentration. The slope from a linear fitting of the data is the dn/dc of
the polymer. The measured dn/dc values of c-PNPG and l-PNPG in LiBr
Synthesis of N-Propargyl N-Carboxyanhydride (M1). The
compound 2 (1.74 g, 6.5 mmol) was dissolved in anhydrous CH2Cl2
(230 mL) under a nitrogen atmosphere. PCl3 (1.15 mL, 13.1 mmol) was
added dropwise to the solution at 0 °C. The reaction mixture was stirred
for 3 h and the solvent was removed under vacuum. The solid residue
was extracted with anhydrous CH2Cl2 (20 mL) and filtered. The filtrate
was evaporated to afford a white solid. Further purification by recrys-
tallization from anhydrous CH2Cl2/hexane and sublimation yielded
white crystals (0.5 g, 55% yield). 1H NMR (400 MHz, CDCl3) δ: 4.27
(0.1 M)/DMF at 25 °C and 690 nm wavelength are 0.1094(14) mL gꢀ1
3
and 0.1012(7) mL gꢀ1 respectively.
3
Intrinsic Viscosity Measurement. Eight polydisperse cyclic or
linear poly(N-propargylglycine) samples with different molecular weight
were independently prepared from NHC or butylamine-mediated poly-
merizations of M1. Polydisperse cyclic (Mn = 14.8 kg molꢀ1, PDI = 1.70)
3
9065
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
Scheme 2
Table 1. NHC-Mediated Polymerization of M1 and Copolymerization of M1 and M2
Mn(theor.)b (kg molꢀ1
)
Mn(SEC)c,d (kg molꢀ1
)
Mn(NMR) (kg molꢀ1
)
PDI
convne
a
entry
polymer
[M1]0:[M2]0:[NHC]0
3
3
3
1
c-PNPG
25:0:1
2.4
3.8
-
2.2
4.1
-
100
100
100
96
2
c-PNPG
50:0:1
4.3c
5.7c
9.1c
1.10c
1.03c
1.13c
1.12c
1.10c
1.12d
1.15d
1.30d
1.20d
1.14d
1.11d
3
c-PNPG
75:0:1
5.5
5.8
4
c-PNPG
100:0:1
122:0:1
200:0:1
150:150:1
250:50:1
100:133:1
76:24:1
50:50:1
20:80:1
9.1
10.9
11.5
-
5
c-PNPG
11.6
18.1
20.5
29.4
24.1
9.9
13.3c
15.6c
44.6d
49.2d
49.9d
56.0d
38.9d
37.1d
100
95
6
c-PNPG
7
c-PNPG159-r-PNBG173
c-PNPG150-r-PNBG30
c-PNPG102-r-PNBG73
c-PNPG103-r-PNBG35
c-PNPG62-r-PNBG49
c-PNPG39-r-PNBG99
43.2
17.7
17.7
13.7
10.9
14.9
99
8
99
9
90
10
11
12
99
10.5
11.1
99
99
a [M1]0=[M2]0= 0.4 M for all polymerizations. b Theoretical molecular weights were calculated from the [M1]0:[M2]0:[NHC]0 ratio and the conversion
c
of monomer to polymer (Note: the 6MR content is subtracted in the calculation). Experimental molecular weight and polydispersity index were
determined by a tandem SECꢀMALSꢀDRI system in LiBr (0.1 M)/DMF solution at 50 °C using a measured dn/dc of 0.1094(14) mL gꢀ1
.
3
d Experimental molecular weight and polydispersity index were determined by a SECꢀDRI system in LiBr (0.1M)/DMF solution at 50 °C using
polystyrene standards. e Monomer conversions were determined by FTIR spectroscopy.
(d, ꢀCH2ꢀ, 2H), 4.24 (s, ꢀCH2ꢀ, 2H), 2.45 (t, tCH, 1H). 13C{1H}
NMR (100 MHz, CDCl3) δ: 165.5 (ꢀCH2C(O)Oꢀ), 151.4 (ꢀOC
(O)Nꢀ), 77.7 (CHt), 76.8 (tCꢀ), 49.2 (tCCH2Nꢀ), 32.6 (C(O)
CH2Nꢀ).
{1H} NMR (100 MHz, CDCl3) δ: 162.1 (C=O), 75.9 (HCt), 73.1
(Ct), 49.1 (CH2C=O), 38.4 (CH2Ct).
Representative synthetic procedure for the cyclic poly(N-pro-
pargylglycine)-ran-poly(N-butylglycine) random copolymers (c-PNPGn-r-
PNBGm). In a glovebox, M1 (64 mg, 0.46 mmol) and N-butyl NCA
(M2) (72 mg, 0.46 mmol) were dissolved in anhydrous THF (4 mL).
A stock solution of NHC in THF (91 uL, 2.83 μmol, 31.1 mM) was added
to the reaction flask. The flask was sealed and stirred at 55 °C for 2 d.
The polymerization was terminated by adding cold hexane (20 mL). The
precipitated polymer was isolated by decantation and dried under vacuum
(52 mg, 54% yield). 1H NMR (400 MHz, CDCl3), δ: 4.64ꢀ3.90
(bm, ꢀCCH2Nꢀ, ꢀNCH2COꢀ, ꢀNCH2CO, 6H), 3.51ꢀ3.20
(bm, ꢀCH2CH2Nꢀ, 2H); 2.62ꢀ2.25 (bm, ꢀCCH, 1H), 1.58ꢀ1.21
(bm, ꢀCH2CH2CH2ꢀ,CH3CH2ꢀ,4H),1.06ꢀ0.76 (bm, CH3CH2ꢀ,3H).
Representative Synthetic Procedure for the Cyclic PEG-
Grafted Poly(N-propargylglycine) (c-PNPG-g-PEG). Inside a glove-
Representative Synthetic Procedure for the Cyclic Poly-
(N-propargylglycine) (c-PNPG). Inside a glovebox, M1 (200 mg,
1.44 mmol) was dissolved in THF (2.5 mL) to which a THF stock
solution of NHC (267 μL, 14.4 μmol, 53.8 mM) was added at room tem-
perature. The reaction was stirred and heated at 55 °C for 18 h. An excess
of hexane (10 mL) was added to the remaining reaction mixture. The
suspension was stirred at 50 °C for 8 h and filtered while still warm to
remove low molecular weight oligomers. The yellow solid that was ob-
1
tained was dried under vacuum (120 mg, 88% yield). H NMR (400
MHz, CDCl3), δ: 4.67ꢀ4.04 (bm, ꢀCOCH2Nꢀ, ꢀCCH2Nꢀ, 4H),
2.66ꢀ2.30 (bm, ꢀCCH, 1H). 13C{1H} NMR (100 MHz, CDCl3)
δ: 163.4 (C=O), 76.6 (HCt), 73.9 (Ct), 48.8 (CH2C=O), 35.0
(CH2Ct).
box, c-PNPG137 (85.5 mg, [propargyl]0 = 0.87 mmol, Mn = 13.4 kg molꢀ1
,
3
Representative Synthetic Procedure for the Linear Poly-
(N-propargylglycine) (l-PNPG). Inside a glovebox, M1 (91 mg,
0.654 mmol) was dissolved in THF (2.5 mL) to which a THF stock
solution of butyl amine (143 μL, 7.69 μmol, 53.8 mM) was added at
room temperature. The solution was degassed by freezeꢀpumpꢀthaw
cycle three times and left to react under reduced pressure in a sealed
flask. The reaction was stirred and heated at 50 °C for 48 h. An excess of
hexane (10 mL) was added to the remaining reaction mixture. The sus-
pension was stirred at 50 °C for 8 h and filtered while still warm to remove
low molecular weight oligomers. The white solid that was obtained was
dried under vacuum (31 mg, 50% yield). 1H NMR (400 MHz, CDCl3),
δ: 4.67ꢀ4.04 (bm, ꢀCOCH2Nꢀ, ꢀCCH2Nꢀ, 4H), 2.66ꢀ2.30 13C
PDI = 1.16) was dissolved in CH2Cl2 (3 mL) along with PEG-N3 (288 mg,
0.14 mmol, Mn =2.0 kg molꢀ1, PDI = 1.03, [N3]0:[propargyl]0 = 1:6).
3
A measured volume of CH2Cl2 stock solution containing CuBr/PMDETA
(1.70 mL, 0.202 mmol, 119 mM, [CuBr]0:[PMDETA]0:[propargyl]0 =
23:23:100) was added to the solution which was then stirred at room
temperature for 3 h. The copper catalyst was removed by passing through an
alumina column, and the grafted copolymer was precipitated by adding
an excess of hexane and dried under vacuum at 25 °C (165 mg, 44%
yield, grafting density: 19%). 1H NMR (400 MHz, CDCl3) δ: 7.62ꢀ8.24
(bs, ꢀNC=CHNꢀ, 1.00 H), 4.75ꢀ4.12 (bm, ꢀNCH2COꢀ, CCH2Nꢀ),
3.97ꢀ3.45 (bm, ꢀCH2CH2Oꢀ), 3.43ꢀ3.26 (bs, ꢀCH2CH2OCH3),
2.26ꢀ2.23 (bs, ꢀCCH, 4.21H).
9066
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
Figure 2. (A) 1H NMR and (B) 13C {1H} NMR spectra of c-PNPG (Mn = 4.3 kg molꢀ1, PDI = 1.10) in CDCl3.
3
Figure 3. (A) Representative full and (B) expanded ESI MS spectra of a low molecular weight c-PNPG (Mn = 2.8 kg molꢀ1, PDI = 1.09) as well as (C)
3
their assigned molecular structures.
Representative synthetic procedure for the cyclic PEG-grafted
poly(N-propargylglycine)-ran-poly(N-butylglycine) random copoly-
mers [(c-PNPG-r-PNBG)-g-PEG]. Inside a glovebox, c-PNPG166-r-
PNBG33 (72.8 mg, [propargyl]0 = 0.62 mmol) and PEG-N3 (465 mg,
stock solution containing CuBr/PMDETA (1.70 mL, 0.202 mmol,
119 mM, [Cu]0:[PMDETA]0:[propargyl]0 = 33:33:100) was added
to the solution which was then stirred at 40 °C for 3 h. The reaction
mixture was then passed through a silica column. The filtrate was
concentrated and cold hexane was added to precipitate the polymer (172
mg, 42% yield, grafting density: 42%). 1H NMR (400 MHz, CDCl3) δ:
0.85 mmol, Mn = 550 g molꢀ1, PDI = 1.05, [N3]0:[propargyl]0 = 1.4:1)
3
were both dissolved in CH2Cl2 (5 mL). A measured volume of CH2Cl2
9067
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
Figure 4. (A) Plot of Mn (9) and PDI (2) versus conversion (i.e., ([M1]0-[M1])/[M1]0) for NHC-mediated polymerization of M1 (THF, 50 °C) and
the linearly fitted curve for the Mn-vs-conversion data (ꢀ). (Note: the monomer conversion was determined by FTIR spectroscopy while Mn and PDI
were measured by SEC-MALS-DRI in LiBr (0.1 M)/DMF solution [dn/dc = 0.1094(14) mL gꢀ1]); (B) plot of ln([M1]0/[M1]) versus time for the
3
NHC-mediated polymerization of M1 and their linearly fitted curves ([NHC]0 = 3.0 (9), 4.4(2), 12.2 mM (b); [M1]0:[NHC]0 = 50:1; THF; 50 °C).
7.62ꢀ8.24 (bs, ꢀNC=CHNꢀ, 1.00 H), 4.87ꢀ4.18 (bm, ꢀNCH2COꢀ,
CCH2Nꢀ,ꢀNCH2COꢀ), 4.04ꢀ3.47 (bm, ꢀCH2CH2Oꢀ,ꢀCH2CH2Nꢀ),
3.43ꢀ3.26 (bs, ꢀCH2CH2OCH3), 2.26ꢀ2.23 (bs, ꢀCCH), 1.58ꢀ1.21
(bm, ꢀCH2CH2CH2ꢀ, CH3CH2ꢀ), 1.06ꢀ0.76 (bm, CH3CH2ꢀ, 2.93 H).
CH2CH2OCH3), 2.26ꢀ2.23 (bs, ꢀCCH), 1.58ꢀ1.21 (bm, ꢀCH2CH2CH2ꢀ,
CH3CH2ꢀ), 1.06ꢀ0.76 (bm, CH3CH2ꢀ, 2.93 H).
of 1,4-di(prop-2-ynyl)piperazine-2,5-dione (6MR) along with
cyclic and linear oligomers in small quantities, presumably
formed by an intramolecular “backbiting” mechanism (Figures
S2 and S3, Supporting Information).29i The 6MR constitutes
less than 15% of the reaction product (Figure S4, Supporting
Information). Heating of the isolated and purified polymers
can cause further depolymerization to yield 6MR. This is in
contrast to our previous study on NHC-mediated polymeriza-
tion of N-alkyl NCA (alkyl: Me, Bu) where no “backbiting”
products were observed.29h The polymer products were preci-
pitated by the addition of excess room temperature hexane.
Further purification was achieved by extraction into warm hexane
(50 °C), which removes cyclic oligomers. The samples were
dried under vacuum prior to further analysis.
Representative synthetic procedure for linear PEG-grafted poly-
(N-propargylglycine)-ran-poly(N-butylglycine) random copoly-
mers [(l-PNPG-ran-PNBG)-g-PEG]. Inside a glovebox, l-PNPG250-r-
PNBG50 (17.0 mg, [propargyl]0= 0.14 mmol) and PEG-N3 (201 mg,
0.37 mmol, Mn = 550 g molꢀ1, PDI = 1.05, [N3]0:[propargyl]0 = 2.6:1)
3
were both dissolved in CH2Cl2 (1 mL). A measured volume of CH2Cl2
stock solution containing CuBr/PMDETA (388 μL, 46 μmol,
119 mM, [Cu]0:[PMDETA]0:[propargyl]0 = 33:33:100) was added to the
solution which was stirred at 40 °C for 3 h. The reaction mixture was passed
through a silica column. The collected filtrate was concentrated and cold hexane
was added to precipitate the polymer (54 mg, 57% yield, grafting density: 63%).
1H NMR (400 MHz, CDCl3) δ: 7.62ꢀ8.24 (bs, ꢀNC=CHNꢀ, 1.00H),
4.87ꢀ4.18 (bm, ꢀNCH2COꢀ, CCH2Nꢀ, ꢀNCH2COꢀ), 4.04ꢀ3.47 (bm,
ꢀCH2CH2Oꢀ, ꢀCH2CH2Nꢀ), 3.43ꢀ3.26 (bs, ꢀCH2CH2OCH3),
2.26ꢀ2.23 (bs, ꢀCCH), 1.58ꢀ1.21 (bm, ꢀCH2CH2CH2ꢀ, CH3CH2ꢀ),
1.06ꢀ0.76 (bm, CH3CH2ꢀ, 0.97 H).
1
The H NMR analysis of a low molecular weight polymer
reveals three broad resonances in the 1H NMR spectrum (parts a,
c, and d of Figure 2A), consistent with the targeted poly(N-
propargyl glycine) backbone structure (c-PNPG) (Scheme 2). In
addition, resonances due to NHC moieties are also evident in the
1H NMR spectrum, in agreement with NHC initiator being affixed
to the polymer chain ends as previously reported (Scheme 2).29h,i
The 13C{1H} NMR spectrum (Figure 2B) is also consistent with
the PNPG backbone structure. The ESI MS analysis of a low
molecular weight polymer reveals a major set of doubly charged
mass ions whose mass equals to the sum of integer number of the
desired repeating unit mass (95.10), one NHC mass (388.29)
and two proton masses (1.01), in agreement with a cyclic PNPG
polymeric species with one NHC moiety attached (part a,
Figure 3AꢀC). Apart from the major species, several minor sets
of mass ions that are consistent with PNPG polymeric species
with different end groups or co-ionized with solvent molecules
are also present. For example, doubly charged mass ions indi-
cated by b and c are due to the c-PNPGs with NHC attached that
are co-ionized by Na+ and a proton (b) or Na+, a proton and a
solvent molecule (c). The mass ions indicated by d are the singly
charged linear PNPG bearing carboxyl and amino chain ends,
presumably formed by the reaction between c-PNPG having
NHC attached with adventitious moisture.34 The end groups for
the PNPG polymeric mass ions indicated by e have yet to be
determined. The MALDIꢀTOF MS analysis of the low molec-
ular weight polymer sample also corroborates the ESI MS results
(Figure S5, Supporting Information). In the MALDIꢀTOF MS
experiment, it is critical to use a soft matrix such as α-cyano-4-
hydroxycinnamic acid (CHCA) so that the original polymers
remain structurally intact upon desorption and ionization.
’ RESULTS AND DISCUSSION
Synthesis and Characterization of Cyclic and Linear Poly-
(N-propargylglycine) (c/l-PNPG). The monomer N-propargyl
NCA (M1) was successfully synthesized on a multigram scale by
the Fuch method, i.e., PCl3-mediated cyclization of the corre-
sponding N-Boc-N-propargylglycine precursor 2 (Scheme 1).
The molecular structure of M1 was unambiguously verified by 1H
and 13C {1H} NMR spectroscopy (Figure 1). The monomer is a
white solid at room temperature and can be readily purified by
sublimation prior to polymerization.
Polymerization of M1 was achieved by heating different
initial monomer to initiator ratios ([M1]0:[NHC]0) in THF
solution at 50 °C for 18 h under a nitrogen atmosphere
(Scheme 2). Aliquots of the reaction mixtures were spectro-
scopically analyzed to monitor conversion of M1 to the polymer.
1
Both H NMR or FTIR spectroscopy, where M1 exhibits two
characteristic νCdO stretching modes at 1859 and 1786 cmꢀ1
(Figure S1, Supporting Information) were used to monitor
the conversion. All reactions reached high or quantitative
monomer conversion (95ꢀ100%) under these conditions
1
(Table 1). In addition to polymer formation, H NMR and
ESI MS analysis of the reaction mixture also reveals the formation
9068
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
Figure 5. Representative 1H NMR spectra of c-PNPG192-r-PNBG41 random copolymer in CDCl3. (* indicates peak due to the residual M1 monomer).
.Scheme 3. R0= 2,6-Diisopropylphenyl)
We have previously shown that the NHC moieties that are
attached to the polymers are photolabile.34 Strong laser power or
a hard matrix such as dithranol causes the NHC to dissociate
from the polymers.
distribution (PDI = 1.03ꢀ1.13) (entry 1ꢀ6, Table 1). The
experimental molecular weight agrees reasonably well with the
theoretical values based on single-site initiation and living polym-
erization. Furthermore, NHC-mediated polymerization of M1 also
exhibits a linear increase of molecular weight over conversion
while the molecular weight distribution remains narrow (PDI =
1.10ꢀ1.23) (Figure 4A), suggesting a constant concentration of
propagating species throughout the reaction course, indicative of a
living polymerization. Kinetic studies reveal that the polymeriza-
tion is first-order dependent on the monomer and NHC concen-
1
The polymers were analyzed by H NMR and SEC-MALS-
DRI techniques for their molecular weight and molecular weight
distribution. The polymer molecular weights (Mn) were deter-
mined by integrating the c-PNPG methine proton (a, Figure 2A)
and the NHC phenyl protons (k, Figure 2A) to give the number-
average degree of polymerization (DPn), assuming that each
polymer chain has one NHC affixed to it. Polymerization of
M1 with increasing [M1]0:[NHC]0 leads to the formation of
tration [i.e., d[M1]/dt=kp[NHC]0[M1], kp= 32 (2) Mꢀ1 hꢀ1
]
3
(Figure 4B and S6). The plots of ln([M1]0/[M1]) versus time
(Figure 4B) all pass through (0,0), consistent with an initiation
that is fast or comparable to the propagation.35
c-PNPG with increasing polymer molecular weight (Mn
=
2.2ꢀ15.6 kg molꢀ1) and relative narrow molecular weight
3
9069
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
The SEC chromatograms of c-PNPGs exhibit a multimodal
distribution (Figure S7, Supporting Information) in common
organic solvents [e.g., THF, CHCl3 and LiBr (0.1M)/DMF]
regardless of the molecular weight range. This is in contrast to
MS results where only a monomodal distribution of mass ions
was observed (Figure 3A and S5). The DLS analysis of c-PNPG
reveals the presence of large particles having nonuniform size.
The size of the particles increases from ∼10 nm to several
micrometer over 3 h in THF, strongly suggesting polymer
aggregation (Figure S8, Supporting Information). Aggregation
in solution appears to be an innate property of the PNPG
polymer rather than being induced by the zwitterionic chain
ends, since the linear poly(N-propargyl glycine)s (l-PNPG)
that are independently prepared by primary amine-initiated
polymerization of M1 (Figure S9 and Table S1, Supporting
Information)29h also exhibit multimodal SEC chromatograms
in spite of their neutral chain ends (Figure S10, Supporting
Information). This is in contrast to cyclic poly(N-butyl glycine)s
(c-PNBG) that do not appear to substantially aggregate, as
supported by monomodal SEC chromatograms.29h In view of
this evidence, we attribute the high molecular weight compo-
nents in the SEC chromatograms of PNPGs (i.e., at low elution
time) to polymer aggregation. The polymer molecular weights
(Mn) determined from the SEC mode at long elution times agree
l-PNPG elutes at a shorter elution time than the c-PNPG
(Figure S10, Supporting Information), consistent with the cyclic
polymers being hydrodynamically more compact than their
linear analogues. Intrinsic viscosity measurement is often con-
ducted to verify the polymer architecture.29a,d,h,i However, it was
proven difficult for the analysis of l-PNPG and c-PNPG due to
their strong tendency to aggregate in common organic solvents.
As a result, the intrinsic viscosity difference observed for l-PNPG
and c-PNPG of identical molecular weight cannot be unambigu-
ously attributed to differences in their molecular architecture or
their aggregation state (Figure S11, Supporting Information).
Synthesis and Characterization of Cyclic Poly(N-propar-
gyl glycine)-ran-poly(N-butyl glycine) Random Copolymers
(c-PNPG-r-PNBG). Cyclic poly(N-propargyl glycine)-ran-poly-
(N-butyl glycine) random copolymers (c-PNPG-r-PNBG) were
prepared by NHC-mediated copolymerization of M1 and N-
butyl N-carboxyanhydride (M2) with different initial [M1]0:
[M2]0:[NHC]0 ratios. The c-PNPG-r-PNBG copolymer com-
position and polymer Mn (Table 1, entry 7ꢀ12) were determined
by 1H NMR integration of the PNPG methylene protons (d) and
the PNBG methine proton (g) relative to the phenyl proton
(l) of the NHC moiety (Figure 5). Polymer Mns and PDIs were
also determined by SEC using a calibration curve constructed
with monodisperse polystyrene standards (Table 1, entry 7ꢀ12).
As the amount of M2 incorporated into the c-PNPG-r-PNBG
copolymer increases, the bimodal character of the SEC chroma-
togram appears to decrease, suggesting reduced aggregation of
the c-PNPG-r-PNBG copolymers in the solution (Figure S12,
Supporting Information).
1
reasonably well with those determined by H NMR analysis
(entry 1ꢀ6, Table 1).
A comparison of SEC chromatograms of l-PNPG and c-PNPG
having identical polymer molecular weight reveals that the
Synthesis and Characterization of Cyclic and Linear
Brush-Like Copolymers. Two azido-terminated poly(ethylene
glycol) polymers (PEG-N3) of different molecular weight (Mn =
2 kg molꢀ1, PDI = 1.03; Mn = 500 g molꢀ1, PDI = 1.05) were
3
3
prepared by a previously reported procedure and their Mns were
determined by MALDI TOF MS analysis.33 Grafting of PEG-N3
to PNPG and PNPGn-ran-PNBGm proceeded at room tempera-
ture in CH2Cl2 in the presence of CuBr/PMDETA (1:1)
(∼20ꢀ30 mol % relative to propargyl content) over a period
of 3 h under nitrogen atmosphere (Scheme 3). The polymer
product was purified by passing it through a neutral alumina
column and precipitated by excess hexane.
Figure 6. (A) Representative SEC chromatograms [LiBr (0.1M)/
DMF, 50 °C] of c-PNPG130 and c-PNPG130-g-(PEG2k)24 obtained
after grafting of PEG by CuAAC chemistry; (B) representative SEC
chromatograms [LiBr (0.1M)/DMF, 50 °C] of c-PNPG48-r-PNBG111
and (c-PNPG48-r-PNBG111)-g-(PEG550)48 obtained after grafting of
PEG by CuAAC chemistry.
Successful grafting of the polymeric side chains by CuAAC
chemistry is evidenced by the appearance of characteristic tri-
1
azolium protons at 8.0 ppm in the H NMR spectrum of the
polymer product (c-PNPG-g-PEG) (k, Figure S13, Support-
ing Information). The SEC analysis of the polymer product
Table 2. Cyclic Brush-Like Polymers with Poly(N-substituted glycine) Backbone and PEG Side Chains
entry backbone compositiona backbone Mnb (kg molꢀ1
)
PEG Mn (g molꢀ1
)
[N3]0: [propargyl]0 Mnb (kg molꢀ1
)
PDIb Ygc (%) Ygd (%)
3
3
3
1
2
3
4
5
6
7
8
c-PNPG141
43.8
43.8
44.6
44.6
49.2
49.2
49.2
49.2
2k
0.5:1
0.8:1
1.2:1
1.5:1
0.75:1
2:1
121
174
194
204
142
175
145
155
1.68
2.28
1.27
1.23
1.82
1.73
1.78
1.71
19
19
37
47
45
62
77
52
-
-
c-PNPG141
2k
c-PNPG221-r-PNBG196
c-PNPG221-r-PNBG196
c-PNPG166-r-PNBG33
c-PNPG166-r-PNBG33
c-PNPG166-r-PNBG33
c-PNPG150-r-PNBG30
2k
42
45
-
2k
550
550
550
550
-
3:1
-
1.6:1
-
a All reactions were carried out in THF (entries 1 and 2) or CH2Cl2 (entries 3ꢀ8) and the polymer composition is determined by 1H NMR analysis. b All
Mns and PDIs were determined by SEC [LiBr(0.1M)/DMF, 50 °C] using polystyrene standards. c Grafting density was determined by 1H NMR analysis.
d Grafting density was determined by the changes of PEG-N3 concentration, i.e., percentage intensity change of their SECꢀDRI response.11
9070
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
Figure 7. Representative AFM topographic (A, E) and amplitude (B, F) images of c-PNPG141-g-(PEG2k)27 and (c-PNPG166-r-PNBG33)-
g-(PEG550)154 (entry 1 and 7, Table 2) respectively on mica (0001) and the cross-section (C, G) and histogram analysis (D, H) of selected ring
polymers within the respective sample (sampling size =50). (Note: the black line in Figure A and E indicate the specific nanostructure whose cross-
section analysis is shown in Figure C and G.)
9071
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
also reveals an increase in the molecular weight relative to that
of c-PNPG, confirming successful grafting of the PEG side
chains to the c-PNPG backbone (Figure 6). However, the mol-
ecular weight distribution is fairly broad (PDI = 1.68ꢀ2.28)
(entry 1ꢀ2, Table 2). The integration ratio of the triazolium
proton (k, Figure S13, Supporting Information) relative to the
propargyl methine proton (y, Figure S13, Supporting In-
formation) has been used to determine the grafting density
(YGRAFTING). The maximum grafting density is ∼20%, which is
low relative to typical polymer brushes prepared by the graft-to
method (entry 1ꢀ2, Table 2).11,36 Varying the ratio of
[N3]0:[propargyl]0 does not appear to have an appreciable effect
in the grafting density. While steric crowding of the affixed
polymeric side chains limits the grafting density, PNPG aggrega-
tion also contributes to the restricted access of propargyl groups
by azido-ended PEG. As a result, we reason that random
copolymers (c-PNPG-r-PNBG) will likely provide improved
grafting efficiency due to the reduced aggregation tendency of
these copolymers.
The side chain grafting density is indeed substantially in-
creased to 37ꢀ77% (entry 3ꢀ8, Table 2), corresponding to high
CuAAC coupling efficiency (70ꢀ93%), when the random co-
polymers (c-PNPG-r-PNBG) are used. The molecular weight
distributions are also narrower (PDI = 1.23ꢀ1.82) than those
obtained when c-PNPGs are used in the PEG grafting experi-
ments. The grafting density was determined by integrating
the resonance for the single triazolium proton (k, Figure S14,
Supporting Information) relative to the three protons on the
methyl group of the PNBG repeating unit (i, Figure S14,
Supporting Information), which is then multiplied by the molar
percentage of PNBG repeating unit in the random copolymers.
Grafting densities can also be enhanced by increasing the ratio of
[N3]0:[propargyl]0 (entries 5ꢀ7, Table 2). We have shown that
increasing PNBG content results in reduced aggregation of the
random copolymers c-PNPG-r-PNBG, as manifested in increas-
ingly monomodal SEC chromatograms (Figure S12, Supporting
Information). The molecular weight distribution (PDI) of the
cyclic brush-like polymers [i.e., (c-PNPG-r-PNBG)-g-PEG] also
appears to decrease with increasing PNBG backbone content,
suggesting that perhaps the reduced aggregation facilitates the
statistical grafting of the side chains, resulting in lowered PDIs
(entry 3ꢀ8, Table 2).
(Figure 7, parts C and G). Amplitude images constructed by
mapping the cantilever oscillation as it is raster scanned across
the surface also reveal ring-shaped nanostructures (Figure 7, parts
B and F). Differences in the nanostructure size were observed for
the cyclic brush-like polymers with variable composition. The
cross-section and histogram analysis of the nanostructures reveals
an average diameter of 283 and 362 nm for c-PNPG141
-
g-(PEG2k)27 and (c-PNPG166-r-PNBG33)-g-(PEG550)154 re-
spectively (Figure 7, parts D and H). The lateral dimensions
are exaggerated when compared to the theoretical diameters
based on the polymer composition (i.e., 49 and 32 nm). This is
attributed to the tip effect in AFM imaging that displays a
convolution of the geometry of the sample and tip, resulting in
overestimation of the lateral features.37
While the ‘donut-shape’ is not as evident for c-PNPG141-g-
(PEG2k)27 (Figure 7, parts A and B), it is clearly visible for
(c-PNPG166-r-PNBG33)-g-(PEG550)154 (Figures 7, parts E and F).
This is likely to arise from the difference in side chain length
relative to the diameter of the cyclic backbone. If the cyclic
backbone and the side chains are assumed to adopt a fully
extended zigzag conformation, the cyclic backbone diameter of
c-PNPG141-g-(PEG2k)27 is estimated to be 17 nm and PEG
chain length is 16 nm, resulting in a theoretical diameter of 49 nm
for the nanostructure. In comparison, the theoretical diameter of
the (c-PNPG166-r-PNBG33)-g-(PEG550)154 nanostructure is
estimated to be 32 nm, which is the sum of the cyclic backbone
diameter (24 nm) and the twice the PEG side chain length
(4 nm). As a result, one would expect to observe a “donut hole” in
the AFM images of the latter sample, but not necessarily the
former. Additionally, the cyclic brush-like polymers can also self-
assemble into larger aggregates (Figure S16, Supporting In-
formation), but small ring nanostructures are still visible along
with the aggregates. By contrast, AFM imaging of the linear
brush-like polymer (l-PNPG-r-PNBG)-g-PEG reveals only large
and ill-defined aggregates.
’ CONCLUSIONS
The monomer N-propargyl N-carboxyanhydrides were suc-
cessfully polymerized using NHC initiators to yield cyclic poly-
(N-propargyl glycine) (c-PPNG) in a controlled manner. The
propargyl groups enable further side chain derivation by CuAAC
chemistry. This was demonstrated in the synthesis of brush-like
polymers having poly(N-substituted glycine) backbone and PEG
side chain. The grafting efficiency of polymer side chain is low
when c-PPNG homopolymer is used, presumably due to polymer
aggregation that hinders access to the propargyl groups. High-to-
quantitative grafting efficiencies can be obtained by utilizing
cyclic random block copolymers (i.e., PPNG-r-PBNG) where
the reactive propargyl groups are spaced by inert butyl side
chains. The increase in grafting density is attributed to either a
decrease in cyclic backbone aggregation in the CuAAC reaction
medium or a decrease in steric crowding among the side chains
on the cyclic backbones. The AFM analysis of the cyclic brush-like
polymers reveals the formation of donut-shape nanostructures
whose dimension correlates well with the molecular composition
of the polymers. The successful development of NHC-mediated
polymerization of NCA bearing side chains amendable to CuAAC
chemistry will enable future development of structurally and
functionally diverse polymer materials with novel architectures.
For example, appending thermo- and pH-responsive side chains
on to the cyclic backbones will produce stimuli-responsive
To further validate the grafting density obtained by 1H NMR
analysis, we quantified the percentage decrease of PEG-N3
content prior to and post CuAAC by the SEC-DRI method
(Figure S15, Supporting Information).11 As the initial [N3]0:
[propargy]0 ratio is known, the grafting density can be deduced.
The grafting densities obtained by 1H NMR or SEC analysis are
in good agreement (entry 3ꢀ4, Table 2). Linear brush-like
copolymers [(l-PNPG-r-PNBG)-g-PEG] can also be synthesized
in a similar manner as the cyclic analogs with high grafting
efficiency.
AFM Analysis of Cyclic Brush-Like Copolymers. Atomic
force microscopic (AFM) analysis of the cyclic brush-like poly-
mers [i.e., c-PNPG141-g-(PEG2k)27 and (c-PNPG166-r-PNBG33)-
g-(PEG550)154] (entry 1 and 7, Table 2) is presented in Figure 7.
Bright areas in the topography images (Figure 7, parts A and E)
display disc- or ring-shaped nanostructures, which exhibit a
narrow size distribution. The darker areas in these topographical
figures are considered to be the mica background. The height
profiles for all samples were relatively low (<2.5 nm), consistent
with a single layer of nanostructures lying flat on the mica surface
9072
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
nano-objects useful for drug delivery applications or smart
membranes. Cyclic brush-like polymers bearing mutually incom-
patible side chains are potentially useful as polymeric blend
compatibilizers.
(13) Kolb, H. C.; Sharpless, K. B.; Finn, M. G. Angew. Chem., Int. Ed.
2001, 40, 2004–2021.
(14) (a) Sumerlin, B. S.; Tsarevsky, N. V.; Louche, G.; Lee, R. Y.;
Matyjaszewski, K. Macromolecules 2005, 38, 7540–7545. (b) Helms, B.;
Mynar, J. L.; Hawker, C. J.; Frꢀechet, J. M. J. J. Am. Chem. Soc. 2004,
126, 15020–15021. (c) Parrish, B.; Breitenkamp, R. B.; Emrick, T. J. Am.
Chem. Soc. 2005, 127, 7404–7410.
’ ASSOCIATED CONTENT
(15) (a) Gondi, S. R.; Vogt, A. P.; Sumerlin, B. S. Macromolecules 2007,
40, 474–481. (b) Golas, P. L.; Matyjaszewski, K. Chem. Soc. Rev. 2010,
39, 1338–1354. (c) Meldal, M. Macromol. Rapid Commun. 2008,
29, 1016–1051. (d) Tsarevsky, N. V. J. Polym. Sci., Part A: Polym. Chem.
2010, 48, 966–974. (e) Wu, J.; Gao, C. Macromolecules 2010, 43, 7139–7146.
(f) Zhang, Y.; He, H.; Gao, C. Macromolecules 2008, 41, 9581–9594. (g) Li,
C.; Ge, Z.; Fang, J.; Liu, S. Macromolecules 2009, 42, 2916–2924. (h) Fu,
G. D.; Xu, L. Q.; Yao, F.; Zhang, K.; Wang, X. F.; Zhu, M. F.; Nie, S. Z. ACS
Appl. Mater. Interfaces 2009, 1, 239–243. (i) Yuan, Y.-Y.; Du, Q.; Wang, Y.-C.;
Wang, J. Macromolecules 2010, 43, 1739–1746.
S
Supporting Information. Time-dependent FTIR spectra
b
1
of a NHC-mediated polymerization of M1, H NMR, ESI
MS, MALDIꢀTOF MS spectra of the polymer product from
NHC-mediated polymerizations of M1 prior to removal of cyclic
oligomers, the molar percentage of M1, c-PNPG, and 6MR as a
function of polymerization time, the plot of the observed rate
constant (kobs) versus the initial NHC concentration, SEC chro-
matograms of c-PNPG, l-PNPG, and c-PNPG-r-PNBG, DLS
1
traces of c-PNPG in THF, H NMR spectrum of l-PNPG, the
(16) Kreutzer, G.; Ternat, C.; Nguyen, T. Q.; Plummer, C. J. G.;
polymerization results of BuNH2-initiated polymerizations of M1
to yield l-PNPG, MarkꢀHouwinkꢀSakurada plot for c-PNPG
and l-PNPG, 1H NMR spectra of c-PNPG-g-PEG and c-PNPG-r-
PNBG-g-PEG, SEC chromatograms that display the conversion
of c-PNPG-r-PNBG into c-PNPG-r-PNBG-g-PEG, and AFM
images of c-PNPG-r-PNBG-g-PEG (entry 8, Table 2) along with
the section and histogram analysis. This material is available free
of charge via the Internet at http://pubs.acs.org.
ꢂ
Manson, J.-A. E.; Castelletto, V.; Hamley, I. W.; Sun, F.; Sheiko, S. S.;
Herrmann, A.; Ouali, L.; Sommer, H.; Fieber, W.; Velazco, M. I.; Klok,
H.-A. Macromolecules 2006, 39, 4507–4516.
(17) Schappacher, M.; Deffieux, A. Science 2008, 319, 1512–1515.
(18) (a) Schappacher, M.; Deffieux, A. J. Am. Chem. Soc. 2008,
130, 14684–14689. (b) Schappacher, M.; Deffieux, A. Angew. Chem., Int.
Ed. 2009, 48, 5930–5933.
(19) Jia, Z.; Fu, Q.; Huang, J. Macromolecules 2006, 39, 5190–5193.
(20) Coulembier, O.; Moins, S. b.; De Winter, J.; Gerbaux, P.;
Leclꢀere, P.; Lazzaroni, R.; Dubois, P. Macromolecules 2010, 43, 575–579.
(21) (a) Xia, Y.; Boydston, A. J.; Grubbs, R. H. Angew. Chem., Int. Ed.
2011, 50, 5882–5885. (b) Zhang, K.; Lackey, M. A.; Wu, Y.; Tew, G. N.
J. Am. Chem. Soc. 2011, 133, 6906–6909.
(22) (a) Semlyen, J. A. Cyclic Polymers; Kluwer Academic Publishers:
Norwell, MA, 2000. (b) Laurent, B. A.; Grayson, S. M. Chem. Soc. Rev.
2009, 38, 2202–2213.(c) Kobayashi, S.; Endo, K. In New Frontiers in
Polymer Synthesis; Springer: Berlin and Heidelberg, Germany, 2008, 217,
121ꢀ183. (d) Kricheldorf, H. R. J. Polym. Sci., Part A: Polym. Chem.
2010, 48, 251–281.
(23) (a) Laurent, B. A.; Grayson, S. M. J. Am. Chem. Soc. 2006,
128, 4238–4239. (b) Whittaker, M. R.; Goh, Y.-K.; Gemici, H.; Legge,
T. M.; Perrier, S.; Monteiro, M. J. Macromolecules 2006, 39, 9028–9034.
(c) Misaka, H.; Kakuchi, R.; Zhang, C.; Sakai, R.; Satoh, T.; Kakuchi, T.
Macromolecules 2009, 42, 5091–5096.
(24) Stanford, M. J.; Pflughaupt, R. L.; Dove, A. P. Macromolecules
2010, 43, 6538–6541.
(25) (a) Schappacher, M.; Deffieux, A. J. Am. Chem. Soc. 2011,
133, 1630–1633. (b) Schappacher, M.; Deffieux, A. Macromolecules
2011, 44, 4503–4510. (c) Hu, J.; Zheng, R.; Wang, J.; Hong, L.; Liu,
G. Macromolecules 2009, 42, 4638–4645.
(26) Szymanski, R.; Kubisa, P.; Penczek, S. Macromolecules 1983,
16, 1000–1008.
(27) Jacobson, H.; Beckmann, C. O.; Stockmayer, W. H. J. Chem.
Phys. 1950, 18, 1607–1612.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: dhzhang@lsu.edu.
’ ACKNOWLEDGMENT
The authors thank Dr. Rafael Cueto for assisting with the SEC
studies. S.H.L. thanks Dr. Haoyu Tang for helpful discussion. This
work was supported by Louisiana State University, the National
Science Foundation (CHE-0955820 and DMR-0906873), and
Louisiana State Board of Regents [LEQSF(2008-11)-RD-A-11].
W.K.S. acknowledges financial support for study-leave from
Masinde Muliro University in Kenya.
’ REFERENCES
(1) Muir, H. Biochem. Soc. Trans. 1983, 11, 613–622.
(2) Wintermantel, M.; Gerle, M.; Fischer, K.; Schmidt, M.; Wataoka,
I.; Urakawa, H.; Kajiwara, K.; Tsukahara, Y. Macromolecules 1996,
29, 978–983.
(3) Zhang, M.; M€uller, A. H. E. J. Polym. Sci., Part A: Polym. Chem.
2005, 43, 3461–3481.
(4) Sheiko, S. S.; Sun, F. C.; Randall, A.; Shirvanyants, D.; Rubinstein,
M.; Lee, H.-i.; Matyjaszewski, K. Nature 2006, 440, 191–194.
(5) Huang, K.; Rzayev, J. J. Am. Chem. Soc. 2009, 131, 6880–6885.
(6) M€uller, M. T.; Yan, X.; Lee, S.; Perry, S. S.; Spencer, N. D.
Macromolecules 2005, 38, 5706–5713.
(28) Kricheldorf, H. R.; Schwarz, G. Macromol. Rapid Commun.
2003, 24, 359–381.
(29) (a) Bielawski, C. W.; Benitez, D.; Grubbs, R. H. Science 2002,
297, 2041–2044. (b) Li, H.; Debuigne, A.; Jꢀerome, R.; Lecomte, P.
Angew. Chem., Int. Ed. 2006, 45, 2264–2267. (c) Herbert, D. E.; Gilroy,
J. B.; Chan, W. Y.; Chabanne, L.; Staubitz, A.; Lough, A. J.; Manners, I.
J. Am. Chem. Soc. 2009, 131, 14958–14968. (d) Culkin, D. A.; Jeong, W.;
Csihony, S.; Gomez, E. D.; Balsara, N. P.; Hedrick, J. L.; Waymouth,
R. M. Angew. Chem., Int. Ed. 2007, 46, 2627–2630. (e) Jeong, W.;
Hedrick, J. L.; Waymouth, R. M. J. Am. Chem. Soc. 2007, 129,
8414–8415. (f) Jeong, W.; Shin, E. J.; Culkin, D. A.; Hedrick, J. L.;
Waymouth, R. M. J. Am. Chem. Soc. 2009, 131, 4884–4891. (g) Shin,
E. J.; Brown, H. A.; Gonzalez, S.; Jeong, W.; Hedrick, J. L.; Waymouth,
R. M. Angew. Chem., Int. Ed. 2011, 50, 6388–6391. (h) Guo, L.; Zhang,
D. J. Am. Chem. Soc. 2009, 131, 18072–18074. (i) Guo, L.; Li, J.; Brown,
Z.; Ghale, K.; Zhang, D. Biopolym.: Peptide Sci. 2011, 96, 596–603.
(7) Kumaraswamy, G.; Dibaj, A. M.; Caruso, F. Langmuir 2002,
18, 4150–4154.
(8) Sheiko, S. S.; Sumerlin, B. S.; Matyjaszewski, K. Prog. Polym. Sci.
2008, 33, 759–785.
(9) Yamada, K.; Miyazaki, M.; Ohno, K.; Fukuda, T.; Minoda, M.
Macromolecules 1998, 32, 290–293.
(10) Beers, K. L.; Gaynor, S. G.; Matyjaszewski, K.; Sheiko, S. S.;
Moller, M. Macromolecules 1998, 31, 9413–9415.
(11) Gao, H.; Matyjaszewski, K. J. Am. Chem. Soc. 2007, 129,
6633–6639.
(12) Binder, W. H.; Sachsenhofer, R. Macromol. Rapid Commun.
2008, 29, 952–981.
9073
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074

Macromolecules
ARTICLE
(30) Patch, J. A.; Barron, A. E. Curr. Opin. Chem. Biol. 2002, 6, 872–877.
(31) Barron, A. E.; Zuckermann, R. N. Curr. Opin. Chem. Biol. 1999,
3, 681–687.
(32) Arduengo, A. J.; Krafczyk, R.; Schmutzler, R.; Craig, H. A.;
Goerlich, J. R.; Marshall, W. J.; Unverzagt, M. Tetrahedron 1999,
55, 14523–14534.
(33) Ostaci, R.-V.;Damiron, D.;Grohens, Y.;Lꢀeger, L.; Drockenmuller,
E. Langmuir 2009, 26, 1304–1310.
(34) Li, X.;Guo, L.; Casiano-Maldonado, M.;Zhang, D.; Wesdemiotis,
C. Macromolecules 2011, 44, 4555–4564.
(35) Beste, L. F.; Hall, H. K., Jr. J. Phys. Chem. 1964, 68, 269–274.
(36) (a) Ostaci, R.-V.; Damiron, D.; Capponi, S.; Vignaud, G.; Leger,
L.; Grohens, Y.; Drockenmuller, E. Langmuir 2008, 24, 2732–2739.
(b) Fallais, I.; Devaux, J.; Jꢀer^ome, R. J. Polym. Sci., Part A: Polym. Chem.
2000, 38, 1618–1629.
(37) Villarrubia, J. S. J. Res. Natl. Inst. Stand. Technol. 1997, 102,
425–454.
9074
dx.doi.org/10.1021/ma201948u |Macromolecules 2011, 44, 9063–9074