Journal of the American Chemical Society
Communication
the core, this means that ∼77% of the pendant olefins (=100 −
10 − 13) were consumed by the intramolecular cross-linking
and/or ladderlike polymerization in the core region upon core
formation. It should be noted that such a detailed analysis of the
core’s chemical structure was made possible by the core
separation and isolation in this work.
The arm cleavage also enabled for the first time the
characterization of the physical properties of the core compared
with its parent star polymer. Generally, globular macromolecules
show solution viscosities smaller than those of their linear
counterparts.22,23 Analysis by SEC-MALLS with a viscosity
detector (Figure 2 and Table 1, entries 1 and 2) showed that S1
From these results, the microgel core is of a swelled and spherical
network structure with a relatively low density. The Rg data for
C1 also revealed that the arm chain density over the core surface
of S1 was 0.21 chain/nm2, indicating a high-density polymer
brush.24
The primary structure and the physical properties of the arm-
cleaved microgels could be tuned by changing the synthetic
conditions of their parent star polymers (Table 1 and Figure S4).
For example, increasing the initial concentration of PEG−
acetal−Cl from 20 to 40 mM while applying the same feed ratio
of EGDMA (r = 10) led to star polymer S2 and arm-free microgel
C2 (∼1030 EGDMA units) with an increased molecular weight
[Mw (MALLS)] and a larger size (Rg) (Table 1, entries 3 and 4)
relative to S1 and C1, respectively. The enhanced core formation
is most likely due to a more efficient intermolecular arm linking.
In spite of the larger Mw, both the viscosity [η] and shape index a
of C2 were smaller than those of C1 (Figure 2), indicating a
higher network density in the former.
As illustrated in Chart 1, an amphiphilic and thermosensitive
microgel, C3 (Mw = 136 000 with 175 short pendent PEG
Chart 1. Amphiphilic and Thermosensitive PEG Microgels
Figure 2. Molecular weight dependence of the intrinsic viscosity for S1
(Mw = 340 000), C1 (Mw = 96 000), C2 (Mw = 309 000), PEO (Mw =
187 000), and PMMA (Mw = 146 000) in DMF.
chains), was obtained from star polymer S3, for which PEG
methyl ether methacrylate (PEGMA) (Mn = 475) was
additionally employed with EGDMA on arm linking. A similar
microgel, C4 (Mw = 240 000 with 370 short PEG spacers) was
also obtained from star polymer S4 with PEG dimethacrylate
(PEGDMA) (Mn = 550) as a linking agent in place of EGDMA
(Table 1 and Figures S4 and S5). While S4 and S1 carried nearly
the same numbers of arms, Mw and Rg of the former were larger as
expected from a longer spacer in the linking agent. As a result, C4
had a larger Mw than C1.
C3 and C4 were amphiphilic and soluble not only in organic
solvents but also in alcohols and water. With either pendant or
spacer PEG units, these microgels were thermosensitive with
lower critical solution temperature-type phase separation in
water25,26 at 60 °C (C3) and 40 °C (C4) (Figure S8). In spite of
the smaller hydrophobic contents [hydrophobic methacrylate/
hydrophilic PEG = 3/1 (C3), 2/1 (C4)], C4 had a cloud point
lower than that for C3. This would be due to the restricted
mobility and conformation of the PEG spacers bridging the
methacrylate units, promoting dehydration of PEG units.
In conclusion, we successfully isolated and directly analyzed
microgel cores in star polymers with an acid-cleavable macro-
initiator (PEG−acetal−Cl). To our knowledge, the structures
and some physical properties of microgel cores were precisely
characterized for the first time in over 40 years since the initial
synthesis of microgel-core star polymers. Despite the absence of
surrounding arm layers, the isolated cores are soluble in various
solvents, in contrast to a previous premise that microgel cores are
solubilized by their linear arms. The cores are spherical, have a
nanoscale network structure (<20 nm) with a relatively large void
space (∼50%), and can be amphiphilic and thermosensitive with
designed linking agents. As a result of their tunable properties
and precise characterization, arm-cleavable microgel star
and C1 actually had intrinsic viscosities ([η]) smaller than those
of their linear counterparts poly(ethylene oxide) (PEO) (Mw =
187 000) and poly(methyl methacrylate) (PMMA) (Mw = 146
000), respectively, with similar molecular weights ([η] = 21, 8.5,
160, and 33 mL/g for S1, C1, PEO, and PMMA, respectively, in
DMF at room temperature).
The double logarithmic plots of [η] versus molar mass (Mw by
MALLS) were fitted using the Mark−Houwink−Sakurada
equation ([η] = KMa), where the index a (the slope) depends
on the polymer conformation (a < 0.5, spherical; 0.5 < a < 1.0,
random coil of a linear chain) (Figure 2). For S1, the slope a over
the SEC peak molecular weight (Mp ≈ 230 000) was 0.27,
indicating that it is globular and different from the linear
polymers (PEO, a = 0.72; PMMA, a = 0.69). In addition, [η]
slightly decreased with increasing Mw to reach a minimum
around Mp (Figure S6). This suggests that the conformation of
the star gradually changes from branched to globular with
increasing Mw or arm number, and this conformational shift, in
turn, offsets an increase in [η] expected for star polymers with
more arms.
For C1, a was 0.56, close to the upper limit of 0.5 for a spherical
shape but small for a random coil. The fact that the shape index
was clearly larger than that of the star suggests that the core
would be a swelled, nearly spherical gel and more flexible in shape
than the parent star polymer surrounded by linear arms.
The core−shell structure of S1 and the nearly spherical
structure of C1 were further supported by small-angle X-ray
scattering (SAXS) analysis in DMF (Figure S7). The radii of
gyration (Rg) as determined by the Guinier plot were 7.1 nm for
S1 and 4.4 nm for C1. With the assumption of a spherical shape,
the EGDMA density in the core microgel was thus 0.44 g/mL,
meaning that C1 is occupied by solvent about 55% in volume.
C
dx.doi.org/10.1021/ja505646p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX