Helix-coil transition in cylindrical brush polymers with poly-l-lysine side chains

Article abstract of DOI:10.1021/ma300377v

Cylindrical brush polymers with poly-l-lysine side chains were prepared by grafting lysine NCA from a macroinitiator via living ring-opening polymerization. The main chain degree of polymerization of the methacrylate main chain was P_w = 870, the side chains consisted of 25 and 55 lysine repeat units, respectively. Upon deprotection, the cylindrical brush polymers in 0.005 M NaBr exhibited an almost rodlike conformation with a Kuhn statistical segment length of several hundred nanometers. Cryo-TEM as well as AFM in aqueous solution clearly demonstrated pronounced undulations along the main chain at low ionic strength which could not be detected at higher salt concentrations. With increasing concentration of NaClO₄ the PLL side chains underwent a coil-to-helix transition as revealed by CD measurements. The effect of the side chain coil-to-helix transition on the main chain stiffness could not be followed by light scattering due to intramolecular attraction ("folding") of the cylindrical brushes at high salt concentration, which is somewhat more pronounced for the helical as compared to the coiled PLL side chain conformation. Comparison with linear PLL revealed the coil-to-helix transition to be hardly affected by the high grafting density of the PLL side chains in the cylindrical brush structures.

Full text of DOI:10.1021/ma300377v

Article
pubs.acs.org/Macromolecules
Helix−Coil Transition in Cylindrical Brush Polymers with Poly-L-lysine
Side Chains
Mike Sahl, Sandra Muth, Robert Branscheid, Karl Fischer, and Manfred Schmidt*
Institute of Physical Chemistry, University of Mainz, Welder Weg 11, 55099 Mainz
*
S Supporting Information
ABSTRACT: Cylindrical brush polymers with poly-L-lysine side chains were
prepared by grafting lysine NCA from a macroinitiator via living ring-opening
polymerization. The main chain degree of polymerization of the methacrylate main
chain was P = 870, the side chains consisted of 25 and 55 lysine repeat units,
w
respectively. Upon deprotection, the cylindrical brush polymers in 0.005 M NaBr
exhibited an almost rodlike conformation with a Kuhn statistical segment length of
several hundred nanometers. Cryo-TEM as well as AFM in aqueous solution clearly
demonstrated pronounced undulations along the main chain at low ionic strength
which could not be detected at higher salt concentrations. With increasing
concentration of NaClO the PLL side chains underwent a coil-to-helix transition
4
as revealed by CD measurements. The effect of the side chain coil-to-helix
transition on the main chain stiffness could not be followed by light scattering due
to intramolecular attraction (“folding”) of the cylindrical brushes at high salt concentration, which is somewhat more pronounced
for the helical as compared to the coiled PLL side chain conformation. Comparison with linear PLL revealed the coil-to-helix
transition to be hardly affected by the high grafting density of the PLL side chains in the cylindrical brush structures.
INTRODUCTION
After its discovery in 1994, cylindrical brush polymers
side chain length in contrast to all other experimental data and
■
38
1
in contrast to very recent computer simulations. The origin of
this discrepancy might well originate from finite concentration
effects which were neglected in ref 37 as already scrutinized in
(
sometimes called “bottlebrushes”) have received considerable
2
−27
attention,
because any flexible main chain can be forced to
7
the literature.
adopt an expanded cylindrical conformation due to the
repulsive interaction of the densely grafted side chains. Thus,
a variety of cylindrical brush polymers have been synthesized by
16,39−44
Theories on regular comblike polymers
do not
provide a unique picture. Mean field theory predicts l ∼
k
2
16,39
sc
15/8
sc
M
which is close to l ∼ M
derived by blob
arguments. Somewhat smaller exponents were reported by
k
“
grafting through”, “grafting from”, and “grafting onto”
43
techniques with various combinations of main and side chain
monomers. Also branched or dendritic side chains were
introduced as “dendronized polymers”, leading to cylindrical
Scheutjens−Fleer self-consistent field calculations for good (l
k
1.82
∼
M
) and poor solvents for long side chains (l ∼
sc
.59 44 16
k
1
2
8−35
M
). Experimental results by Nakamura et al. on
sc
objects of similar conformations.
Unfortunately, the
cylindrical brushes with flexible side chains in good and theta
precise main and side chain conformation is extremely difficult
to analyze, because the cylinder length might be shorter than
the (all-trans) contour length of the main chain depending on
1
6,39
solvent are well described by the relation
2
lk = l + kMsc
k,0
6
−8
subtle details of solvent quality,
on side chain-side chain
with lk,0 the intrinsic Kuhn length of the main chain without
side chain repulsion, whereas deviating results were reported by
interactions in cylindrical brushes with chemically different side
3
6
chains and possibly on main chain-side chain interactions. A
combination of light and neutron scattering has been employed
7
Zhang et al. who obtained an exponent of 1.79 in good and
1
.42 in poor solvents. All experiments yielded unusually large
to experimentally determine the Kuhn length l of the cylinder
by application of the Kratky−Porod wormlike chain model and
the cross-sectional radius of gyration R and the cross-sectional
density profiles ρ(R) of the side chains has been determined as
a function of different side chain lengths.
experimental results qualitatively agree that the Kuhn length
increases with increasing side chain length and that Rgc
increases with side chain length quantitative agreement has
not been achieved so far due to different experimental
k
values for lk,0 in the regime of 5.3 nm < l < 12 nm which is
k,0
7
,16
not well understood.
Whereas first computer simulations seemed to be compatible
gc
45
7
,37
with scaling laws (with deviating exponents as compared to
analytical theory, though) newer investigations reveal a
considerable gradient of chain stiffness along the contour of
Although
Received: February 23, 2012
Revised: May 5, 2012
Published: June 5, 2012
7
,12−16
37
techniques and theories employed.
have reported that the chain stiffness does not depend on the
Rathgeber et al.
©
2012 American Chemical Society
5167
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
the “bottlebrush”, resulting in a higher flexibility at both
In order to obtain the free macroinitiator (6) the Boc-protection
4
6−48
groups of the precursor were cleaved in a mixture of dichloromethane
ends.
Concomitantly, the simulations also show a
61
1
and trifluoroacetic acid according to R. A. Houghten et al. H NMR
analysis of the macroinitiator and its precursor clearly showed that the
Boc-protection groups were removed quantitatively (see Supporting
Information, Figure S2b).
pronounced nonexponential decay of the bond angle
45−49
correlation function
which additionally questions the
applicability of the wormlike chain model to experimental
data. Only for very long main and side chains presently outside
of the experimentally accessible regime the results of computer
“
Grafting from” Polymerization of the NCA. The NCA of ε-
62
benzyloxycarbonyl-L-lysine (7) was polymerized via ROP from the
macroinitiator 6. According to Schlaad,
48
63,64
simulations and analytical theory are believed to converge.
the ionic form of the
Very little data exist on cylindrical brush polymers with
rodlike side chains, in particular, on how a coil to rod transition
of the side chains affects the main chain conformation.
Analytical theories do not yield a significantly different scaling
behavior of the main chain Kuhn length for flexible and rodlike
side chains: For rodlike side chains mean field calculations
macroinitiator was deliberately used to suppress the activated
monomer mechanism and to prevent the formation of linear
contaminations. In a typical experiment 35 mg of the macroinitiator
6
2
and 2 g of the Z-protected L-lysine NCA were placed under argon in
65
a 100 mL Schlenk flask and dissolved in 6 mL of purified DMF. The
reaction flask was sealed with a septum, purged with argon for 30 min
and placed in an oil bath preheated to 60 °C. The ring-opening
polymerization was carried out for 3 days. After precipitation in water,
the pure Z-protected poly L-lysine brushes 8 were isolated.
5
0
2
yield l ∼ M /ln M which is not very different from the
k
sc
sc
various results for flexible side chains. Computer simulations
5
1,52
reveal
the main chain to become significantly stiffer upon a
It should be noted that the terminal thiol group of the
6
6
coil to rod transition of the side chains, but within the limited
range of investigated side chain lengths smaller exponents result
macroinitiator could also initiate a PLL chain at the end of the
macroinitiator. Given the very many side chains in a cylindrical brush
polymer one more PLL chain would not make a significant difference
so that this question was not pursued in detail.
1
.15
for both, flexible and stiff side chains (l = 26 + kM
for
k
sc
1
.6
flexible and l = 20 + kM
for stiff side chains) and unusually
large values for lk,0. Kawaguchi et al have investigated a
cylindrical brush with rigid helical poly(n-hexyl isocyanate) side
k
sc
53
54
Two different monomer-initiator-ratios (25:1; 55:1) were chosen to
produce brushes with different side chain length. Both brushes, Z-CB-
PLL25 and Z-CB-PLL55 were soluble in HFIP and could be analyzed
by static and dynamic light scattering (see Table 1 and Supporting
Information for Zimm plots).
0
sc
.87
chains and obtain lk ∼ M and Rgc ∼ M
in clear
sc
contradiction to both, computer simulation and scaling results.
Our group has recently published the synthesis and
characterization of cylindrical brushes with poly-L-lysine side
Table 1. Light Scattering Results of the Cylindrical Brushes
with Z-Protected PLL Side Chains Measured in HFIP with
5
5
chains which were shown to induce a helical conformation of
−3
the main chain, if the side chains are forced into a β-sheet
10
M LiBr
5
6,57
conformation.
Since poly-L-lysine is known to also exhibit a
Rg/
nm
Rh/
nm
ρ-
M /
w
Mwtheo
−1
/
(dn/dc)/
3 −1
58
helical conformation under appropriate solvent conditions,
−1
sample
Z-CB-
PLL55
Z-CB-
PLL25
ratio
g·mol
g·mol
cm g
cylindrical brushes with poly-L-lysine side chains of different
lengths were prepared in the present work, in order to monitor
the effect of a coil to helix transition of the poly-L-lysine side
chains on the main chain conformation.
79.7
47.4
1.68
1.36 × 107
1.28 × 107
0.228
0.228
6.06 × 106
5.95 × 106
68.9
39.1
1.76
Deprotection of Poly-L-lysine Brushes. In order to obtain the
deprotected poly-L-lysine brushes (9), the protected precursors were
treated with a mixture of trifluoroacetic acid and hydrogen bromide
EXPERIMENTAL SECTION
■
Synthesis of N-tert-Butoxycarbonyl-1,6-hexanediamine (2).
N-tert-Butoxycarbonyl-1,3-hexanediamine (N6NB; 2) was synthesized
6
7
(33 wt %) in acetic acid according to literature. Then, 300 mg of the
59
according to Stahl et al. Subsequently, monomer 3 (MA-N6NB; 3)
was obtained by reaction between N6NB and methacryloyl chloride.
In a typical experiment compound N6NB (7.58 g, 0.046 mol) was
suspended in methylene chloride (200 mL) and cooled in an ice bath
before triethylamine (0.12 mol) was added. To the stirred solution, 5.2
g (0.05 mol) methacryloyl chloride, dissolved in 50 mL methylene
chloride, was added dropwise and the solution was kept stirring
overnight at 0 °C. After filtration, the solution was washed with
aqueous sodium bicarbonate solution (0.1 N, 250 mL), water (3 × 250
mL), dried over anhydrous sodium sulfate, and evaporated in vacuum.
The product 3 (MA-N6NB) was purified two times by column
chromatography (ethyl acetate/toluene (50:50). Yield: 80%. ESI(M +
precursor was dissolved in 3 mL of trifluoroacetic acid followed by the
addition of 40 mL of hydrogen bromide in acetic acid. The mixture
was stirred for 1.5 h and precipitated in 500 mL of diethyl ether. After
washing with 3 × 200 mL of ethanol, the white solid was solved in 30
mL of distilled H O and dialyzed for 5 days. Afterward, the polymer 9
2
was isolated by lyophilization. UV−vis analysis of the precursor and
the protected polymer showed that the Z-protection groups were
removed quantitatively. The bands at 254 nm, originating from the
benzene ring of the Z-group, disappeared in the spectra of the
deprotected polymer brushes (see Supporting Information, Figure S5).
It should be noted that the hydrobromide of the lysine units is isolated
by the procedure described above.
Static and Dynamic light Scattering. Static light scattering
(SLS) measurements were performed with an ALV-SP86 goniometer,
an ALV-3000 correlator, a Uniphase HeNe Laser (25 mW output
power at λ = 632.8 nm wavelength) and ALV/High QE APD
avalanche diode fiber optic detection system. For dynamic light
scattering (DLS) an ALV-SP125 goniometer, an ALV-5000 correlator,
and a Spectra Physics 2060 Argon ion laser (500 mW output power at
λ = 514.5 nm wavelength) were utilized. The scattered intensity was
divided by a beam splitter (approximately 50:50), each portion of
which was detected by a photomultiplier. The two signals were cross-
correlated in order to eliminate nonrandom electronic noise.
+
1
13
Na ): 307 g/mol, M = 284 g/mol, FD: M = 284 g/mol. H and
C
NMR spectra are shown in the Supporting Information (Figures S1a
and S1b) as well as the HPLC elution curve (Figure S1c).
Synthesis of the Macroinitiator 5. 3 was polymerized by RAFT
to protected macroinitiator (5) using 4-cyano-4-thiobenzylvaleric
60
acid (4) as the RAFT-agent and AIBN as the initiator. In a typical
−6
polymerization 0.46 mg (2.8 × 10 mol) AIBN and 3,93 mg (1.4 ×
−
5
10
mol) 4-cyano-4-thio-benzyl-valeric-acid dissolved in 2.5 mL
anisole were placed with 20 g (0.704 mol) MA-N6NB in a Schlenk
flask. Oxygen was removed by 4 freeze pump cycles. Subsequently,
polymerization was performed at 90 °C for 3 h. Then the solid
reaction mixture was diluted with 50 mL THF, precipitated in 700 mL
diethyl ether and dried in high vacuum for 24 h. Yield: 6.7 g (30%).
Solutions were typically measured from 30 to 150° in steps of 5°
(SLS) or in steps of 20° (DLS). The static scattering intensities were
analyzed according to Zimm in order to yield the weight-average molar
1
For H NMR spectrum see Supporting Information, Figure S2a.
5
168
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
Figure 1. Reaction scheme for the synthesis of cylindrical poly-L-lysine brushes.
6
8
mass, M , the square root of z-average mean square radius of gyration,
elsewhere. The experimental uncertainties are estimated to ±2% for
dn/dc.
w
2
R ≡ ⟨R ⟩ , and the second viral coefficient A . The experimental
g
g
z
2
uncertainties are estimated to ±5% for Mw and R . The correlation
Gel Permeation Chromatography (GPC). GPC was performed
with a Hitachi L7100 pump and a L-2490 refractive index detector, a
Waters 486 UV detector operating at 280 nm wavelength and a Waters
AF degasser. The flow rate of the solvent DMF was set to 1 mL/min at
a temperature of 333 K. Three columns filled with polystyrene beads
g
functions showed a monomodal decay and were fitted by a sum of two
exponentials, from which the first cumulant Γ was calculated. The z-
2
average diffusion coefficient D was obtained by extrapolation of Γ/q
z
for to q = 0 leading to the inverse z-average hydrodynamic radius R ≡
h
−1
4
5
6
⟨
1/R ⟩
by formal application of Stokes law. The experimental
(nominal pore sizes 10 , 10 , and 10 Å pore size, 5 μm size, MZ-
h
z
uncertainties are estimated to ±2% for Rh.
Analysentechnik) were utilized as separation medium.
Stock solutions of each sample were prepared at c = 0.1 g/L and
filtered through 0.45 μm pore size Millipore GHP filters into 20 mm
diameter quartz cuvettes (Hellma). Further dilutions were made by
subsequent addition of filtered solvent into the LS cuvette and the
respective concentrations were obtained by weighing.
Circular Dichroism (CD). CD measurements were performed with
a JASCO J-815 instrument equipped with a solid-state detector.
Measurements were done in the step-scan mode. One mm thick quarz
cuvettes were used and filled with solutions of c = 0.1 g/L polymer
concentration.
The refractive index increments at λ = 632.8 nm wavelength were
measured by a home-built Michelson interferometer as described
Atomic Force Microscopy (AFM). Measurements were per-
formed with a Veeco MultiMode Scanning Probe Microscope and a
5
169
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
Nanoscope IIIa Controller. All micrographs were taken in the Tapping
mode. For measurements on the dried samples Olympus OMCL-
AC160-W2 cantilevers with a resonance frequency of 300 kHz (spring
constant 42 N/m) were utilized, for measurements in solution Veeco
SNL-10 cantilevers with a resonance frequency of 40−75 kHz, (spring
constant 0.58 N/m).
Cryo-TEM. The sample was vitrified in liquid propane using the
climate chamber of a Vitrobot (FEI, Eindhoven, Netherlands) at room
temperature and 80% humidity. The cryo-TEM measurements were
performed at 120 kV by a Tecnai 12 (FEI, Eindhoven, Netherlands)
equipped with a BIO-TWIN-lens and a 4k CCD (Tietz Video and
Image Processing Systems GmbH, Gauting, Germany).
Table 2. Light Scattering Results of Cylindrical Brushes with
Deprotected PLL Side Chains (9) Measured in 0.005 M
NaBr Solution
Rg/
nm
Rh/
nm
ρ-
ratio
M /
Mwtheo
/
(dn/dc)/
3 −1
w
−1
−1
sample
g·mol
g·mol
cm g
9.95 × 106
1.08 × 107
CB-
84.1
56.7
1.48
0.164
0.164
PLL55
CB-
PLL25
75.9
42.0
1.80
8.14 × 106
4.81 × 106
RESULTS AND DISCUSSION
The synthesis of the poly-L-lysine brushes was performed by
ring-opening polymerization of the Z-protected L-lysine NCA
The light scattering results of the cylindrical brushes with
protected poly-L-lysine side chains 8 with nominal side chain
degrees of polymerization 25 and 55, Z-CB-PLL25 and Z-CB-
PLL55, are summarized in Table 1.
■
55
from a macroinitiator as described by Zhang et al. (Figure 1).
theo
The theoretical molar mass Mw in Table 1 is calculated by
Mw theo = Mw + ([M]/[I])P MIM
MI
Z‐lys
w
0
(1)
with MwMI and PwMI the molar mass and degree of
polymerization of the macroinitiator, respectively, [M]/[I]
Z‑lys
the monomer to initiator ratio and M0 the molar mass of the
Z-protected lysine repeat unit. The agreement of the theoretical
and the experimentally determined molar masses demonstrates
a near to complete conversion of the protected lysine
monomers and that no significant amount of linear PLL has
formed. The latter could be confirmed by capillary electro-
69
phoresis in aqueous solution on the deprotected samples. It
should be noted, that these results do not prove that each of the
macroinitiator sites has started a PLL side chain. Since the side
chains cannot be cleaved off the main chain for an independent
molar mass determination without degradation of the PLL side
chains the length of the side chains given above is derived
assuming 100% grafting density. If the true grafting density
were less the length of the side chains would increase
accordingly. Thus, the precise length of the side chains cannot
be determined.
Figure 2. Zimm plot of the Boc-protected macroinitiator 5: M = 2.59
w
5
−7
3
−2
2
1/2
×
10 g/mol, A = 1.6 × 10 mol·dm ·g , ⟨R ⟩
dn/dc) = 0.04 cm /g.
= 15.7 nm with
2
g
z
3
(
AFM analysis on a mica surface shows the typical wormlike
chain conformation expected for cylindrical brush polymers
(
Figure 3).
Upon deprotection both cylindrical brush polymers were
soluble in aqueous solution. Static and dynamic light scattering
results in 0.005 M NaBr solution are summarized in Table 2
(
see Figure 7 below and Supporting Information for Zimm
plots and angular dependent DLS measurements, Figures S3
and S4).
Upon deprotection, the molar mass of the protected
Z‑CB
cylindrical brushes, Mw , is expected to decrease due to the
loss of the Z-protection groups according to
MwCB,cal = MwZ−CBMoHBr−Lys/MoZ−lys
(2)
with MoZ‑lys = 262 g/mol and MoHBr‑Lys = 208 g/mol the molar
masses of the Z-protected lysine and deprotected lysine
hydrobromide repeat units. Whereas the experimentally
determined molar mass of CB-PLL55 agrees well with the
calculated one, the molar mass of CB-PLL25 is too large by
Figure 3. AFM-picture of Z-CB-PLL55 (8) spin-cast from HFIP
solution on graphite (HOPG).
7
0%, which is well outside the experimental uncertainty. We
have no explanation for this discrepancy except for a subtle
aggregation tendency. The short PLL side chains might not be
long enough in order to completely shield the hydrophobic
main chain. As shown below the calculated molar mass is
indeed found if CB-PLL25 is dissolved in a helicogenic solvent
for the PLL side chains.
The absolute molar mass of the protected macroinitiator 5
was determined by light scattering to MW = 259000 g/mol
(
Figure 2) which corresponds to 870 monomer units (weight-
average). The polydispersity was estimated to M /M = 1.5 by
w
n
GPC (polystyrene calibration).
5
170
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
Figure 4. AFM micrographs in aqueous solution (left, height; middle, amplitude) and cryo-TEM picture (right) of sample CB-PLL55 in 0.005 M
NaBr solution.
Table 3. Contour Lengths Determined by AFM in Different
Solvents
LnAFM/
nm
LwAFM/
nm
Lw
AFM
/
AFM
sample
solvent
Ln
counts
CB-PLL25
CBPLL-25
0.005 M NaBr
0.5 M NaClO4
149
149
186
181
160
200
209
226
244
214
1.34
1.40
1.22
1.35
1.34
276
359
321
309
254
Z-CB-PLL55 spin-cast HFIP
CB-PLL55
CB-PLL55
0.005 M NaBr
0.5 M NaClO4
The dimensions R and R of the deprotected PLL brushes
g
h
increase somewhat, which is to be expected by the introduction
Figure 6. Content of α-helix versus NaClO concentration for CB-
4
70−72
PLL25 (black), Cb-PLL55 (red), and linear PLL with P = 50 (green)
of ionic charges. Application of the wormlike chain model
w
and linear PLL with P = 1500 (blue).
w
and accounting for polydispersity as well as for the contribution
of the side chains at both chain ends to the contour length of
73
the cylinders (see below), the Kuhn statistical segment length
was calculated to increase from l = 300 nm for the protected
k
side chain sample Z-CBB-PLL55 in HFIP (for comparison: l =
k
1
70 nm for CBB-Z-PLL25 in HFIP) to l = 400 nm of the
k
deprotected sample in aqueous 0.005 M NaBr solution. These
values are among the largest reported for cylindrical brush
polymers and reveal a particularly strong repulsion between the
polypeptide side chains. It should be noted, however, that the
determination of the Kuhn length becomes quite inaccurate if
the Kuhn length becomes comparable to or even exceeds the
contour length because R is getting too close to the rod limit
g
rod
(
Rg = 103 nm, l = ∞). Moreover, the application of the
k
Kratky−Porod wormlike chain model to cylindrical brush
polymers might not be justified, because recent computer
simulations have demonstrated the main chain persistence to
significantly decrease from the middle of the cylindrical brushes
46−49
toward the ends.
In order to qualitatively confirm the large persistence of the
cylindrical brushes, AFM in aqueous 0.005 M NaBr solution as
well as cryo-TEM were performed on sample CB-PLL55. The
Cryo-TEM picture shown in Figure 4 (right) clearly shows
almost rodlike structures whereas the AFM-pictures (left) show
significantly curved cylinders. Obviously, the interaction with
the negatively charged mica surface disturbs the conformation
of the cylindrical brushes during the adsorption process. Both,
AFM and Cryo-TEM pictures reveal a peculiar shape of the
cross-section reminiscent of “pinned clusters” postulated by
Figure 5. CD spectra of (a) CBN-PLL55 and (b) CB-PLL25 in
aqueous solution for different concentrations of NaClO . The solid
4
7
8−80
lines represent the fits to the data by the program Dichroweb.
(●,
H O; red
2
●
, 0.1 N NaClO ; green ●, 0.25 N NaClO ; blue ●, 0.5 N
4
4
NaClO ; purple ●, 1.0 N NaClO ). Note that θ is given in
4
4
7
4,75
millidegrees and not in terms of molar ellipticity.
scaling arguments.
It should be noted, however, that scaling
5
171
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
6
−8
3
−2
2
1/2
Figure 7. (a) Zimm plot for sample CB-PLL55 in 0.005 M NaBr: M = 9.95 × 10 g/mol, A = 4.87 × 10 mol·dm ·g , ⟨R ⟩ = 84.1 nm with
w
2
g
z
3
6
−9
3
−2
2
1/2
(
6
dn/dc) = 0.164 cm /g. (b) Zimm plot for sample CB-PLL55 in 0.5 M NaCl: M = 8.68 × 10 g/mol, A = 7.73 × 10 mol·dm ·g , ⟨R ⟩
=
w
2
g
z
3
6
−8
0.4 nm with (dn/dc) = 0.164 cm /g. (c) Zimm plot for sample CB-PLL55 in 0.5 M NaClO : M = 8.13 × 10 g/mol, A = 1.29 × 10
4 w 2
3 2 1/2 3 2
−2
mol·dm ·g , ⟨R ⟩
= 49 nm with (dn/dc) = 0.164 cm /g. (d) Apparent diffusion coefficient Dapp versus q for CB-PLL55 in 0.005 M NaBr
g
z
(
triangles), 0.5 M NaCl (circles), and 0.5 M NaClO (squares).
4
Table 4. Light Scattering Results of Samples CB-PLL25 and CB-PLL55 in Aqueous Salt Solutions
3
‑2
‑1
sample
solvent
Rg/nm
Rh/nm
A /mol·dm ·g
M /g·mol
w
ρ-ratio
structure
2
8
8
8
9
5.01 × 106
CB-PLL 25
CB-PLL 25
CB-PLL 55
CB-PLL 55
0.5 N NaClO4
0.5 N NaBr
56.4
64.8
49.0
60.4
37.1
42
−6.53 × 10−
1.52
1.54
1.28
1.35
helix
coil
−
6
5.85 × 10
6.90 × 10
0.5 N NaClO4
0.5 N NaCl
38.4
44.8
1.29 × 10−
7.73 × 10−
8.13 × 106
helix
coil
6
8.86 × 10
arguments were derived for flexible main chains with a Gaussian
segment distribution between the branch points. Thus, the
coincidence with the “pinned cluster” scenario might be
accidental.
In addition the contour length of the deprotected cylindrical
PLL brushes was determined by AFM in aqueous solution at a
mica surface. The number and weight-average contour lengths
The contour lengths agree within experimental uncertainty
(approximately ±10% comprising the statistical error due to the
limited number of molecules counted and the reliability of the
software for length determination) and compare well with the
values deducted from the light scattering and GPC results of
the macroinitiator. The polydispersity M /Mn = 1.35 is
w
somewhat lower than determined by GPC via polystyrene
calibration.
Since it is well-known, that linear PLL adopts a coillike
L and L for CB-PLL 25 and CB-PLL 55 could be measured
n
w
and are given in Table 3 along with the calculated contour
lengths L based on the number-average degree of polymer-
w
conformation in aqueous NaBr solutions and an α-helical
MI
ization Pw = 870, the contour length of the polymethacrylate
76,77
structure in aqueous NaClO solution (>0.25 M),
the
4
main chain repeat unit, l = 0.25 nm, and an estimate of the side
chain contribution at both ends to the total length of the
cylinder
conformation of the PLL side chains in the cylindrical brush
polymers was investigated by CD spectroscopy in the
respective solvents. The results are shown in Figure 5 for
CB-PLL55 (a) and for CB-PLL25 (b).
MI
chain end
Lw = Pw l + 2L
(3a)
The typical minima at 208 and 222 nm clearly reveal a major
fraction of the PLL side chains to adopt a helical conformation.
Quantitative analysis of the spectra was performed by the
and
78−80
Lchain end = R2 1/2 = (LPLLl
PLL 1/2
program Dichroweb (routine CDSSTR)
α-helix and β-sheet content. The α-helix contents as a function
of NaClO concentration are shown in Figure 6 for CB-PLL25
and CB-PLL55 cylindrical brushes as well as for a short (P
50) and long (P = 1500) chain linear PLL for comparison. As
in terms of coil,
)
(3b)
k
with LPLL and lkPLL the contour and the Kuhn statistical
4
segment lengths, respectively, of the PLL side chains. Utilizing
=
w
PLL
lk = 4 nm (due to some stretching of the densely grafted side
w
chains) the contour lengths are calculated to be L = 223 nm
expected the α-helix content is higher for the longer PLL chain
w
for CB-PLL25 and L = 226 nm for CB-PLL55.
since the stability of the helix decreases for shorter chains.
w
5
172
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
Figure 8. (a) AFM micrographs of CB-PLL25 in 0.5 M NaClO solution, where the side chains adopt a helical conformation. (b) AFM micrographs
4
of CB-PLL25 in 0.5 M NaBr solution, where the side chains are coiled. Left: height. Right: amplitude.
Unfortunately, a shorter linear PLL with P = 25 was not
ionic strength as expected. At constant ionic strength the
dimensions are consistently smaller for the helical conformation
of the side chains as compared to coiled side chains. However, a
quantitative analysis in terms of the wormlike chain model does
not seem to be meaningful for the following reasons: The radii
of gyration of the short side chain polymer CB-PLL25 are
consistently larger as compared to the larger side chain polymer
CB-PLL55. This is not at all expected since the extension of the
cylindrical brush polymers should increase with increasing side
chain length. At the same time the hydrodynamic radii of CB-
PLL25 and CB-PLL55 do not change significantly. This
inconsistency may be partly caused by the aggregation tendency
of CB-PLL25 in aqueous solution. Irrespective of molar mass,
the ratio R /R of 1.35 and 1.28 of CB-PLL55 in 0.5 M NaBr
w
available in order to elucidate the chain length dependence of
the coil-to-helix transition of linear PLL for very short chains.
On the basis of the comparison of CB-PLL55 with the linear
PLL50 the topological constraint of the cylindrical brush seems
to favor a somewhat higher helix content, but does not
significantly affect the PLL secondary structure within
experimental uncertainty in terms of PLL side chain length. It
should be noted that the PLL side chain length could not be
independently determined but was calculated with the
assumption of 100% grafting density. If the actual grafting
density is smaller, the side chains become larger and could get
closer to the long chain limit of the helix content. This small
effect is in contrast to cylindrical brush polymers with
polyproline side chains which were observed to undergo a
transition of proline II to proline I helices in contrast to linear
g
h
and 0.5 M NaClO , respectively, indicates a partial collapse of
4
the brushes. This is actually confirmed by the AFM
measurements shown in Figure 8. It is seen that the cylindrical
brushes predominantly fold back on themselves apparently by
some attractive interaction between the side chains. This
behavior is expected for cylindrical brushes in poor solvents as
81
polyproline chains. This observation was explained by the
steric overcrowding of the polyproline side chains in the
cylindrical brush which favors the “slimmer” proline II
conformation.
8
2−84
predicted theoretically
as well as observed experimentally
Since the overall conformation of the cylindrical brushes may
be expected to depend on both, the secondary structure of the
side chains and on the ionic strength, static and dynamic light
scattering experiments were performed on both samples in 10−3
M and 0.5 M NaCl/0.5 M NaBr solution as well as in 0.5 M
71,85
on dried cylindrical brushes with PMMA side chains.
Here
it is demonstrated that the folding phenomenon also occurs in
solution and does not constitute a drying artifact.
CONCLUSION
NaClO solution. Representative measurements are shown in
■
4
Figure 7 for CB-PLL55. A summary of all measurements is
given in Table 4 (also see Supporting Information, Figures S6
Cylindrical brushes with polypeptide side chains have advanced
to interesting research subjects, because the main chain exhibits
a large directional persistence and the conformation may be
tuned from coillike to helical or to β-sheet as demonstrated
−
S9).
The light scattering results qualitatively show that the
dimensions of the cylindrical brushes shrink with increasing
57
before. Whereas the twist of the side chain β-sheets forced the
5
173
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
main chain into a helix with a pitch of 20 nm, the influence of
the coil-to-helix side chain conformation on the main chain
conformation is more subtle and less spectacular. The
theoretically interesting influence of the coil-to-rod transition
of the side chains onto the main chain stiffness could not be
detected due to the decreasing solvent quality of the helical side
chains which caused the cylindrical brushes to adopt a more
compact “folded” conformation due to side chain-side chain
attraction.
At low ionic strength the main chains were observed to show
regular undulations in the cross-section, which may be
interpreted in terms of “pinned cluster” or “string of spheres”
scenarios. However, to the best of our knowledge, a profound
theory on the conformation of cylindrical brushes with ionic
side chains as a function of added salt is still missing.
(15) Terao, K.; Hayashi, S.; Nakamura, Y.; Norisuye, T. Polym. Bull.
2
(
2
000, 44, 309.
16) Hokajo, T.; Terao, K.; Nakamura, Y.; Norisuye, T. Polym. J.
001, 33, 481.
(17) Jha, S.; Dutta, S.; Bowden, N. B. Macromolecules 2004, 37, 4365.
(18) Runge, M. B.; Dutta, S.; Bowden, N. B. Macromolecules 2006,
3
9, 498.
19) Sun, F.; Sheiko, S. S.; Mo
Phys. Chem. 2004, 108, 9682.
20) Beers, K. L.; Gaynor, S. G.; Matyjaszewski, K.; Sheiko, S. S.;
Moller, M. Macromolecules 1998, 31, 9413.
(21) Borner, H. G.; Beers, K. L.; S., G.; Matyjaszewski, K.; Sheiko, S.
S.; Moller, M. Macromolecules 2001, 34, 4375.
22) Sheiko, S. S.; Borisov, O. V.; Prokhorova, S. A.; Mo
Polym. J. E 2004, 13, 125.
23) Cheng, G.; Boker, A.; Zhang, M.; Krausch, G.; Mu
Macromolecules 2001, 34, 6883.
24) Zhang, M.; Breiner, T.; Mori, H.; Mu
4, 1449.
25) Deffieux, A.; Schappacher, M. Macromolecules 1999, 32, 1797.
(
̈
ller, M.; Beers, K.; Matyjaszewski, K. J.
(
̈
̈
̈
(
̈
ller, M. Eur.
(
̈
̈
ller, A. H. E.
(
̈
ller, A. H. E. Polymer 2003,
ASSOCIATED CONTENT
Supporting Information
■
4
(
*
S
2
NMR and UV spectra, Zimm plots and q -dependence of the
(26) Deffieux, A.; Schappacher, M. Macromolecules 2000, 33, 7371.
(27) Lecommandoux, S.; Checot, F.; Borsali, R.; Schappacher, M.;
2
reduced first cumulant, Γ/q .This material is available free of
charge via the Internet at http://pubs.acs.org.
Deffieux, A.; Brulet, A.; Cotton, J. P. Macromolecules 2002, 35, 8878.
28) Zhang, A.; Shu, Z. B.; Schluter, A. D. Macromol. Chem. Phys.
2003, 204, 328.
29) Schluter, A. D.; Rabe, J. P. Angew. Chem., Int. Ed. Engl. 2000, 39,
64.
30) Zhang, A.; Zhang, B.; Wac
A. D. Chem.Eur. J. 2003, 9, 6083.
(31) Kasemi, E.; Zhuang, W.; Rabe, J. P.; Fischer, K.; Schmidt, M.;
Colussi, M.; Keul, H.; Yi, D.; Colfen, H.; Schluter, A. D. J. Am. Chem.
Soc. 2006, 128, 5091.
(
̈
AUTHOR INFORMATION
Corresponding Author
E-mail: mschmidt@uni-mainz.de.
■
(
̈
8
*
(
̈
̈
htersbach, E.; Schmidt, M.; Schluter,
Notes
The authors declare no competing financial interest.
̈
̈
ACKNOWLEDGMENTS
We are grateful to Dipl. Chem. D. Weller for conducting some
of the CD measurements and to E. Wac
measurements of the refractive index increments. This work
was financially supported by the Deutsche Forschungsgemein-
schaft (SFB 625) and the Graduate School “Materials Science
in Mainz” (stipend for M.S.).
■
(
̈
32) Barner, J.; Mallwitz, F.; Shu, L.; Schluter, A. D.; Rabe, J. P.
Angew. Chem., Int. Ed. Engl. 2003, 42, 1932.
̈
htersbach for
(
33) Percec, V.; Ahn, C.-H.; Ungar, G.; Yeardley, D. J. P.; Mo
M.; Sheiko, S. S. Nature 1998, 39, 161.
34) Percec, V.; Ahn, C.-H.; Cho, W.-D.; Jamieson, A. M.; Kim, J.;
Leman, T.; Schmidt, M.; Gerle, M.; Moller, M.; Prokhorova, S. A.;
̈
ller.,
(
̈
Sheiko, S. S.; Ghemg, S. Z. D.; Zhang, A.; Ungar, G.; Yeardley, D. J. P.
J. Am. Chem. Soc. 1998, 120, 8619.
(35) Zhang, B.; Wepf, R.; Fischer, K.; Schmidt, M.; Besse, S.;
Lindner, P.; King, B. T.; Sigel, R.; Schurtenberger, R.; Talmon, Y.;
REFERENCES
■
(
1) Wintermantel, M.; Schmidt, M.; Tsukahara, Y.; Kajiwara, K.;
Kohjiya, S. Makromol. Chem., Rapid Commun. 1994, 15, 279.
Ding, Y.; Kro
Ed. 2011, 50, 737.
36) Stephan, T.; Muth, S.; Schmidt, M. Macromolecules 2002, 35,
857.
37) Rathgeber, S.; Pakula, T.; Wilk, A.; Matyjaszewski, K.; Beers, K.
L. J. Chem. Phys. 2005, 122, 12490.
38) Hsu, H.-P.; Paul, W.; Rathgeber, S.; Binder, K. Macromolecules
̈
̈
ger, M.; Halperin, A.; Schluter, A. D. Angew. Chem. Int
(
2) Wintermantel, M.; Gerle, M.; Fischer, K.; Schmidt, M.; Wataoka,
I.; Urakawa, H.; Kajiwara, K.; Tsukahara, Y. Macromolecules 1996, 29,
78.
3) Sheiko, S.; Gerle, M.; Fischer, K.; Schmidt, M.; Mo
Langmuir 1997, 13, 5368.
4) Dziezok, P.; Sheiko, S.; Fischer, K.; Schmidt, M.; Mo
Angew. Chem., Int. Ed. 1997, 36, 2812.
(
9
(
9
(
̈
ller, M.
ller, M.
(
(
̈
2010, 43, 1592.
(
(
5) Djalali, R.; Li, S. Y.; Schmidt, M. Macromolecules 2002, 35, 4282.
6) Gerle, M.; Fischer, K.; Muller, A. H. E.; Schmidt, M.; Sheiko, S.
ller, M. Macromolecules 1999, 32, 2629.
hn, F.; Pedersen, J. S.; Fischer, K.; Schmidt, M.
Macromolecules 2006, 39, 8440.
8) Fischer, K.; Schmidt, M. Macromol. Rapid Commun. 2001, 22,
87.
9) Zhang, B.; Zhang, S.; Okrasa, L.; Pakula, T.; Stephan, T.;
Schmidt, M. Polymer 2004, 45, 4009.
10) Neiser, M. W.; Okuda, J.; Schmidt, M. Macromolecules 2003, 36,
437.
11) Li, C.; Gunari, N.; Fischer, K.; Janshoff, A. Angew. Chem. 2004,
16, 1121.
12) Terao, K.; Takeo, Y.; Tazaki, M.; Nakamura, Y. Polym. J. 1999,
1, 193.
13) Terao, K.; Nakamura, Y.; Norisuye, T. Macromolecules 1999, 32,
11.
14) Terao, K.; Hokajo, T.; Nakamura, Y.; Norisuye, T. Macro-
molecules 1999, 32, 3690.
(39) Subbotin, A.; Saariaho, M.; Ikkala, O.; ten Brinke, G.
Macromolecules 2000, 33, 3447.
(40) Birshtein, T. M.; Borisov, O. V.; Zhulina, E. B.; Khokhlov, A. R.;
Yurasova, T. A. Polym. Sci. USSR 1987, 29, 1293.
(41) Guo, Y.; van Beek, J. D.; Zhang, B.; Colussi, M.; Walde, P.;
̈
S.; Prokhorova, S.; Mo
7) Zhang, B.; Gro
̈
(
̈
(
7
(
Zhang, A.; Kro
2009, 131, 11841.
(42) Kroger, M.; Peleg, O.; Halperin, A. Macromolecules 2010, 43,
6213.
̈
̈
ger, M.; Halperin, A.; Schluter, A. D. J. Am. Chem. Soc.
̈
(
5
(
1
(
3
(
7
(43) Frederickson, G. H. Macromolecules 1993, 26, 2825.
(44) Rouault, Y.; Borisov, O. V. Macromolecules 1996, 29, 2605.
(45) Saariaho, M.; Ikkala, O.; Szleifer, I.; Erukhimovich, I.; ten
Brinke, G. J. Chem. Phys. 1997, 107, 3267.
(46) Elli, S.; Ganazzoli, F.; Timoshenko, E. G.; Kuznetsov, Y. A.;
Connolly, R. J. Chem. Phys. 2004, 120, 6257.
(47) Connolly, R.; Bellesia, G.; Timoshenko, E. G.; Kuznetsov, Y. A.;
Elli, S.; Ganazzoli, F. Macromolecules 2005, 38, 5288.
(48) Hsu, H.-P.; Paul, W.; Binder, K. Eur. Phys. Lett. 2010, 92, 28003.
(
(49) Hsu, H.-P.; Paul, W.; Binder, K. Macromolecules 2010, 43, 3094.
5
174
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175

Macromolecules
Article
(
50) Subbotin, A.; Saariaho, M.; Stepanyan, R.; Ikkala, O.; ten Brinke,
G. Macromolecules 2000, 33, 6168.
51) Saariaho, M.; Szleifer, I.; Ikkala, O.; ten Brinke, G. Macromol.
Theory Simul. 1998, 7, 211.
52) Saariaho, M.; Szleifer, I.; Ikkala, O.; ten Brinke, G. Macro-
molecules 1999, 32, 4439.
53) Relations obtained from a fit of the data shown in Figure 3 of ref
2.
54) Kikuchi, M.; Lien, L. T. N.; Narumi, A.; Jinbo, Y.; Izumi, Y.;
Nagai, K.; Kawaguchi, S. Macromolecules 2008, 41, 6564.
55) Zhang, B.; Fischer, K.; Schmidt, M. Macromol. Chem. Phys.
005, 205, 157.
56) Gunari, N.; Cong, Y.; Zhang, B.; Janshoff, A.; Schmidt, M.
Macromol. Rapid Commun. 2008, 29, 821.
57) Gunari, N.; Cong, Y.; Zhang, B.; Janshoff, A.; Schmidt, M.
Langmuir 2009, 25, 6392.
(
(
(
5
(
(
2
(
(
(
(
58) Cassim, J. Y.; Yang, J. T. Biopolymers 1970, 9, 1475.
59) Stahl, G. L.; Walter, R.; Smith, C. W. J. Org. Chem. 1978, 43,
2
285.
60) Aamer, K. A.; Tew, G. N. J. Polym. Sci., Part A: Polym. Chem.
007, 45, 5618.
61) Houghten, R. A.; Beckman, A.; Oestres, J. M. Int. J. Pept. Protein
Res. 1986, 27, 653.
(
2
(
(
(
(
(
́
62) Daly, W. H.; Poche, D. Tetrahedron Lett. 1988, 29, 5859.
63) Dimitriov, I.; Schlaad, H. Chem. Commun. 2003, 2994.
64) Meyer, M.; Schlaad, H. Macromolecules 2006, 39, 3967.
65) Thomas, A. B.; Rochow, E. G. J. Am. Chem. Soc. 1957, 57, 1843
(
BaO Method)..
66) Zhang, X.; Oddon, M.; Giani, O.; Monge, S.; Robin, J.-J.
Macromolecules 2010, 43, 2654.
(
(
67) Ben-Ishai, D.; Berger, A. J. Org. Chem. 1952, 17, 1564.
(
68) Becker, A.; Kohler, W.; Muller, B. Ber. Bunsen-Ges. 1995, 99,
̈
̈
600.
(
69) Sahl, M. Ph.D. Thesis, Mainz 2011.
(
(
(
(
(
70) Kratky, O.; Porod, G. Recl. Trav. Chim. Pays-Bas 1949, 68, 1106.
71) Benoit, H.; Doty, P. J. Phys. Chem. 1953, 57, 958.
72) Yamakawa, H.; Fujii, M. Macromolecules 1973, 6, 407.
73) Schmidt, M. Macromolecules 1984, 17, 553.
̈
74) Sheiko, S. S.; Borisov, O. V.; Prokhorova, S. A.; Moller, M. Eur.
Phys. J. E 2004, 13, 125.
(
(
75) Halperin, A.; Zhulina, E. B. Europhys. Lett. 1991, 15, 417.
76) Peggion, E.; Cosani, A.; Terbojevic, M.; Borin, G. Biopolymers
1
972, 11, 633.
(
(
(
77) Ebert, Ch.; Ebert, G. Colloid Polym. Sci. 1977, 255, 1041.
78) Whitmore, L.; Wallace, B. A. Biopolymers 2008, 89, 392.
79) Whitmore, L.; Wallace, B. A. Nucleic Acid Research 2004, 32,
668.
(
(
(
80) See also: http://dichroweb.cryst.bbk.ac.uk/html/home.shtml
81) Zhang, A.; Guo, Y. Chem.Eur. J. 2008, 14, 8939.
82) Subbotin, A.; de Jong, J.; ten Brinke, G. Eur. Phys. J. E 2006, 20,
9
9.
83) De Jong, J.; Subbotin, A.; ten Brinke, G. Macromolecules 2005,
8, 6718.
84) Flikkema, E.; ten Brinke, G. Macromol. Theory Simul. 2002, 11,
77.
85) Khalatur, P. G.; Khokhlov, A. R.; Prokhorova, S. A.; Sheiko, S.
S.; Moller, M.; Reineker, P.; Shirvanyanz, D. G.; Starovoitova, N. Eur.
Phys. J. E 2000, 1, 99.
(
3
(
7
(
̈
5
175
dx.doi.org/10.1021/ma300377v | Macromolecules 2012, 45, 5167−5175