Synthesis of ultralarge molecular weight bottlebrush polymers using Grubbs' catalysts

Article abstract of DOI:10.1021/ma049647k

This paper describes two methods to synthesize bottlebrush polymers with molecular weights from 1 million to over 60 million g mol^-1 using Grubbs' first and second generation catalysts. In the first method, macromonomers of poly(L-lactide) were synthesized using tin(II) 2-ethylhexanoate and terminated on one end with a norbornyl group. Grubbs' first generation catalyst polymerized macromonomers with one poly(L-lactide) chain per norbornene, and Grubbs' second generation catalyst polymerized macromonomers with two poly(L-lactide) chains per norbornene. The predicted and measured molecular weights closely matched each other, and the polydispersities of the bottlebrush polymers were between 1.05 and 1.39. These examples are the first where Grubbs' second generation catalyst can be considered living for ROMP. In the second method, the backbone was polymerized first, and polylactide arms with molecular weights from 15 000 to 50 000 g mol^-1 were polymerized from the backbone. Polymers that were shaped as spheres or rigid rods were synthesized. The polymers were analyzed by GPC, MALLS, QELS, and ¹H NMR.

Full text of DOI:10.1021/ma049647k

Macromolecules 2004, 37, 4365-4374
4365
Synthesis of Ultralarge Molecular Weight Bottlebrush Polymers Using
Grubbs’ Catalysts
Sh eu li J h a , Sa m r a t Du tta , a n d Ned B. Bow d en *
Department of Chemistry, University of Iowa, Iowa City, Iowa 52242
Received February 21, 2004; Revised Manuscript Received April 5, 2004
ABSTRACT: This paper describes two methods to synthesize bottlebrush polymers with molecular weights
from 1 million to over 60 million g mol-1 using Grubbs’ first and second generation catalysts. In the first
method, macromonomers of poly(L-lactide) were synthesized using tin(II) 2-ethylhexanoate and terminated
on one end with a norbornyl group. Grubbs’ first generation catalyst polymerized macromonomers with
one poly(L-lactide) chain per norbornene, and Grubbs’ second generation catalyst polymerized macro-
monomers with two poly(L-lactide) chains per norbornene. The predicted and measured molecular weights
closely matched each other, and the polydispersities of the bottlebrush polymers were between 1.05 and
1.39. These examples are the first where Grubbs’ second generation catalyst can be considered living for
ROMP. In the second method, the backbone was polymerized first, and polylactide arms with molecular
weights from 15 000 to 50 000 g mol-1 were polymerized from the backbone. Polymers that were shaped
1
as spheres or rigid rods were synthesized. The polymers were analyzed by GPC, MALLS, QELS, and H
NMR.
In tr od u ction
synthesized by the copolymerization of different macro-
monomers.8 In the second method arms are polymerized
from the backbone to yield bottlebrush polymers. This
method is often challenging as small amounts of cou-
pling between arms can lead to cross-linked, insoluble
polymers. In addition, steric crowding between the arms
will limit the concentration of arms along the backbone.
Thus, the syntheses of the arms are typically run at high
dilutions. In the third method, backbones and arms are
polymerized separately and coupled in a final step. This
method has the advantage that both the arms and
backbones can be well characterized prior to their
coupling. A disadvantage is that steric crowding may
limit the number of arms that are ultimately coupled
to a backbone.
These polymers offer challenges to their full charac-
terization. As they may assume shapes from random
coils to rigid rods, these polymers are not correctly
characterized using standard size exclusion chromatog-
raphy connected to a refractive index detector. These
polymers must be characterized using light scattering
to determine their molecular weights.5,14 In addition,
others have reported that these polymers have anoma-
lous elution times and separations in SEC.10
The synthesis and applications of well-defined, soluble
organic nanomaterials remain important areas of chem-
istry as nanoscience is recognized as a field that will
have a large and wide impact in the next number of
years.1 Organic nanomaterials that assume the shape
of spheres are well-known, yet only a few examples of
the synthesis of more complex, soluble architectures
have been reported.2-4 Polymers have been self-as-
sembled in solution and the solid state, cross-linked in
an inorganic matrix, and synthesized in multiple steps
to yield organic nanomaterials with unique architec-
tures.2-4 The synthesis of nonspherical, organic archi-
tectures remains an important goal that will have wide
applications in nanotechnology. In this paper we will
report our efforts to synthesize ultrahigh molecular
weight bottlebrush polymers that are shaped as spheres
and nanorods.
Bottlebrush polymers are a new architecture of
polymers with regular and densely spaced arrange-
ments of arms along a backbone (Figure 1).4-12 Both the
arms and backbones are polymers; thus, these polymers
have ultrahigh molecular weights and may assume
nonspherical shapes due to steric crowding between
arms. For arms with medium to high molecular weights
steric crowding between adjacent arms causes the
backbone to assume a mostly linear conformation.7,10,13
One consequence of this steric crowding is that bottle-
brush polymers may be shaped as rigid rods with
occasional bends and kinks.
The growth of living and controlled polymerizations
over the past dozen years has yielded a series of
methods that can be tailored to synthesize bottlebrush
polymers.3,9,15-19 One polymerization catalyst that is
curiously absent from use in the synthesis of bottlebrush
polymers despite its high activity and excellent toler-
ance of functional groups is Grubbs’ catalyst. Only a few
examples of Grubbs’ first generation catalyst have been
reported in the synthesis of bottlebrush polymers; no
examples of Grubbs’ second generation catalyst have
been reported.11 We wished to explore the use of Grubbs’
first and second generation catalysts to synthesize
bottlebrush polymers with molecular weights in excess
Several excellent syntheses of bottlebrush polymers
have been reported.4-11 These polymers are typically
synthesized using one of three methods. In the first
method macromonomers are polymerized to yield bottle-
brush polymers. This method has found widespread use
as the physical properties of the arms can be character-
ized prior to the synthesis of the bottlebrush polymers,
and a wide variety of polymer architectures are readily
of 10 million g mol-1
.
Grubbs’ first and second generation catalysts are well-
studied.15-18,20,21 The first generation catalyst typically
is used to synthesize polymers with small to medium
* Corresponding author: e-mail ned-bowden@uiowa.edu; Tel
(319) 335-1198; Fax (319) 335-1270.
10.1021/ma049647k CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/21/2004

4366 J ha et al.
Macromolecules, Vol. 37, No. 12, 2004
F igu r e 1. An example of a bottlebrush polymer that we synthesized and a schematic of the polymer beneath it. In (a) the polylactide
(wavy line) is terminated on one end with a norbornyl group (square). The macromonomer is polymerized to yield a bottlebrush
polymer as shown in (c). A macromonomer is a polymer with a polymerizable group on one end. (b) Lactide is polymerized from
a backbone (connected squares) to yield a bottlebrush polymer. (c) The backbone is a polynorbornene and the arms are polylactides.
performed at room temperature. A Waters 515 HPLC pump
was used for SEC. Four Waters columns (styragel HR2, HR4,
HR4, and HR6) were used in series. The DAWN EOS 18 angle
laser light scattering detector from Wyatt Corp., the Wyatt-
QELS for quasi-elastic light scattering, and the Wyatt Optilab
DSP for refractive index were used to measure the absolute
molecular weights of the polymers. We used a value of 6.59 ×
10-5 mol mL g-2 for the second viral coefficient for the bottle-
brush polymers in Tables 3 and 4. This value was the average
of those measured for entries 1-3 in Table 4. For molecular
weights of less than 10 million g mol-1, the molecular weights
and polydispersities varied less than 20% when the second
viral coefficient was changed from 0 to 6.59 × 10-5 mol mL
g-2. For higher molecular weight polymers the changes in the
molecular weights and polydispersities were more substantial.
degrees of polymerization along the backbone (from tens
to several hundred monomer units). We wished to study
whether this catalyst could be used to synthesize well-
defined polymers with degrees of polymerization in
excess of 500 that could function as backbones for the
synthesis of bottlebrush polymers. Grubbs’ second gen-
eration catalyst is not described as a living catalyst as
polymerizations with it are not well controlled.16,17,20
Yet, this catalyst is one of the most active ring-opening
metathesis polymerization (ROMP) catalysts that are
commercially available.20 In this paper we will report
the polymerization of small molecule monomers and
high molecular weight macromonomers with Grubbs’
catalysts.
Syn th esis of Hom op olyla ctid e Usin g 1-Hexa d eca n ol
a s th e In itia tor . The polymerizations in Table 1 were run
from 2 to 24 h. We will describe the procedure for one
polymerization as a representative example. 1-Hexadecanol
(0.121 g, 0.5 mmol) was added to a glass vial and evacuated.
L-Lactide (2.38 g, 16.5 mmol) was added to the vial. At an
initiator-to-catalyst ratio of 50/1, tin(II) 2-ethylhexanoate (3.28
µL, 1.0 µmol) was added to the glass vial. The vial was
evacuated and backfilled with N2 five times. The vial was
flame-sealed under vacuum and heated to 110 °C for 2 h. The
vial was cooled in ice and opened, and the contents were
dissolved in CH2Cl2. The polymer was precipitated into
methanol and filtered. It was dried under vacuum for 48 h.
In this paper we will describe two approaches to
synthesize bottlebrush polymers using ROMP and ring-
opening polymerizations (Figure 1). We synthesized
bottlebrush polymers by polymerizing macromonomers
of L-lactide terminated with norbornene and by polymer-
izing L-lactide onto backbones of polynorbornene. Poly-
lactide is important as it is biologically compatible and
easily degraded. Both methods yielded bottlebrush
polymers, but each method had unique advantages and
disadvantages. In addition, we will outline evidence
based on light scattering that our highest molecular
weight bottlebrush polymers are shaped as rigid rods.
Polylactide had a low value for dn/dc and did not scatter
light well. We could not determine the absolute molecular
weight of homopolylactide; thus, we investigated the molecular
weights of the polylactides using polystyrene standards (pur-
chased from Waters) and by end-group analysis from the 1H
NMR spectrum. To determine the molecular weight by NMR,
the peaks at 1.3, 0.9, 5.1, and 1.6 ppm were integrated against
each other. The triplet at 0.9 ppm corresponded to the methyl
on the initiator. The broad peak at 1.3 ppm corresponded to
26 protons on the initiator. The quartet at 5.1 ppm cor-
responded to the CO-CH(CH3)-O peak. The doublet at 1.6
ppm corresponded to the CO-CH(CH3)-O peak. We found the
molecular weights of the polylactides by measuring the ratio
of the integration of the peaks at 1.3 and 5.1 ppm. For
Exp er im en ta l Section
Ma ter ia ls. 1,12-Dodecanediol, tert-butyldimethylsilyl chlo-
ride, 1-hexadecanol, imidazole, oxalyl chloride, tetraethylene
glycol, 4-(dimethylamino)pyridine, p-toluenesulfonic acid mono-
hydrate, 1,3-dicyclohexylcarbodiimide, 1,1,1-tris(hydroxy-
methyl)ethane, 1-hexadecanol, 5-norborne-2-carboxylic acid (as
an endo/exo mixture), tetrabutylammonium fluoride (1 M in
THF), and DOWEX 50X2-400 were purchased from Aldrich
or Acros Organics at their highest purity and used as received.
The GPC solvent was HPLC grade chloroform and purchased
from Acros Organics and used as received. All other solvents
were reagent grade and purchased from Acros Organics.
(3S)-cis-3,6-Dimethyl-1,4-dioxane-2,5-dione (98%) was pur-
chased from Aldrich. It was purified by recrystallization from
ethyl acetate three times and stored under N2 in a glovebox.
Tin(II) 2-ethylhexanoate was purchased from Aldrich and used
as received. The pure exo isomer of 5-norborne-2-carboxylic
acid was synthesized according to a literature procedure.22
2-Methoxyethyl ether was dried over activated molecular
sieves (type 3A) and stored in a Kontes flask. Three freeze-
pump-thaw cycles were performed, and it was taken into the
glovebox, poured over activated aluminum oxide, and stored.
Isopropylidine-2,2-bis(methoxy)propionic acid was synthesized
according to a literature procedure.23 Geduran silica gel 60 was
purchased from Fisher and used for all purifications.
polylactide with Mn ) 5200, 9500, 14 000, and 21 000 g mol-1
,
the ratio of the peak at 5.1 to 1.3 ppm was 2.6, 4.9, 7.3, and
11.1, respectively. The peaks at 0.9 and 1.6 ppm were used to
find a second value for Mn that was checked against the first
value.
Syn th esis of Ma cr om on om er s of P olyla ctid e. The syn-
thesis of macromonomers followed the same general procedure
as in the previous section. The molar ratio of alcohol to tin
was 50/1, and the reactions were run from 2 to 24 h. We will
describe the synthesis of exo-A-4300. A (0.644 g, 0.002 mol)
was added to a glass vial. Next, L-lactide (9.36 g, 0.065 mol)
was added. The vial was evacuated and backfilled with N2 five
times. Tin(II) 2-ethylhexanoate (13.0 µL, 0.04 mmol) was
added. The vial was evacuated and backfilled with N2 five
times. The glass vial was flame-sealed. The vial was placed
in an oil bath at 110 °C for 2 h. The vial was cooled in ice and
opened, and the contents were dissolved in 50 mL of CH2Cl2.
In str u m en ta tion . 1H and 13C NMR spectra were recorded
on a Bruker DPX 300 using CDCl3. The solvent signal was
used as an internal standard. Size exclusion chromatography
(SEC) using CHCl3 as the mobile phase (1.0 mL min-1) was

Macromolecules, Vol. 37, No. 12, 2004
Bottlebrush Polymers from Grubbs’ Catalysts 4367
The polymer was precipitated into 400 mL of methanol. The
polymer was redissolved in CH2Cl2 and precipitated into
methanol three times. The polymer was isolated and dried
under vacuum at room temperature for 48 h. 1H NMR
(CDCl3): δ 1.58 (d, 3H, J ) 7.1 Hz), 5.17 (q, 1H, J ) 7.1 Hz).
ROMP of Ma cr om on om er s w ith Gr u bbs’ F ir st Gen -
er a tion Ca ta lyst. The exo-A macromonomers in Table 2 were
polymerized at the highest possible concentrations from 4 to
24 h. The macromonomers were polymerized in a glovebox
under an atmosphere of N2. We will describe the synthesis of
entry 2 in Table 3. The macromonomer (2.50 g) was dissolved
in 6.68 mL of CH2Cl2. Grubbs’ first generation catalyst was
added to a vial and dissolved with CH2Cl2 to a concentration
of 1.00 mg mL-1. From this catalyst solution, 0.82 mL was
added to the monomer. The reaction was allowed to run for
15 h. The reaction was removed from the glovebox, and ethyl
vinyl ether (1 mL) was added. The reaction was stirred for 1
h, and the polymer was precipitated into 300 mL of methanol.
The polymer was isolated by filtration, redissolved in CH2Cl2,
and precipitated into methanol. The polymer was dried under
vacuum for 48 h.
was cooled to room temperature. tert-Butyldimethylsilyl chlo-
ride (5.587 g, 37.1 mmol) was added, and the reaction mixture
was stirred for 18 h at room temperature. Water (150 mL) was
added to the reaction mixture, and the product was extracted
with four portions of CH2Cl2. After the solvent had been
removed under reduced pressure, the product was dissolved
in 50 mL of CH2Cl2 and extracted with four portions of water
(100 mL). The crude product was purified by liquid chroma-
tography on silica gel, eluting with 3:97 ethyl acetate/hexane
gradually increasing to 15:85 ethyl acetate/hexane to give (tert-
butyldimethylsilyloxy)dodecanol as a colorless viscous oil:
4.568 g, yield ) 39%. 1H NMR (CDCl3): δ -0.06 (s, 6H,
Si(CH3)2), 0.78 (s, 9H, C(CH3)3), 1.11-1.16 (m, 16H), 1.31-
1.43 (m, 4H), 3.13 (s, 1H, OH), 3.47 (m, 4H, CH2O).13C NMR
(CDCl3): δ -5.56, 13.89, 18.07, 20.65, 25.47, 25.58, 25.62,
25.71, 29.23, 29.30, 29.41, 29.44, 32.51, 32.62, 60.12, 62.24,
63.03, 170.85. HRMS: Calcd for C18H40O2Si + H+: 317.2876.
Found: 317.2892.
Bicyclo[2.2.1]h ep t -5-en e-2-ca r b oxylic Acid 12-(ter t-
Bu tyld im eth ylsilyloxy)d od ecyl Ester . 5-Norborne-2-car-
boxylic acid (1.635 g, 11.9 mmol) was dissolved in 15 mL of
CH2Cl2 at 0 °C under a nitrogen atmosphere. The mixture was
stirred at 0 °C for 30 min. Oxalyl chloride (4.07 mL, 47.4 mmol)
was added. The temperature was raised to 25 °C, and the
reaction was allowed to react for 6 h. The CH2Cl2 was
evaporated, and a slightly yellow viscous oil was obtained. The
oil was dissolved in 12 mL of THF at 0 °C under N2. The
mixture was stirred at 0 °C for 30 min, and then (tert-
butyldimethylsilyloxy)dodecanol (4.50 g, 14.2 mmol) and py-
ridine (3.83 g, 47.4 mmol) were added. The temperature was
raised to 25 °C, and the reaction was allowed to run overnight.
The reaction mixture was extracted with CH2Cl2. The crude
product was purified by liquid chromatography on silica gel,
eluting with 2:98 ethyl acetate/hexane to give bicyclo[2.2.1]-
hept-5-ene-2-carboxylic acid 12-(tert-butyldimethylsilyloxy)-
dodecyl ester as a colorless viscous oil: 4.543 g (88%). 1H NMR
(CDCl3): δ -0.03 (s, 6H, Si(CH3)2), 0.83 (s, 9H, C(CH3)3), 1.21-
1.31 (m, 18H) 1.39-1.48 (m, 3H), 1.52-1.58 (m, 2H), 1.86 (m,
1H), 2.14 (m, 1H), 2.83 (m, 1H), 2.96 (m, 1H), 3.52 (t, 2H, J )
6.6 Hz, CH2O), 4.00 (d of t, 2H, J ) 0.6 Hz and J ) 6.8 Hz,
CH2-O-CO-), 6.04 (m, 2H, CHdCH). 13C NMR (CDCl3): δ
-5.42, 18.16, 25.68, 25.83, 28.58, 29.13, 29.31, 29.39, 29.42,
29.44, 29.51, 30.15, 32.74, 41.48, 43.00, 46.17, 46.46, 63.04,
64.29, 135.58, 137.79, 175.85. HRMS: Calcd for C26H48O3Si +
Na+: 459.3270. Found: 459.3261.
Bicyclo[2.2.1]h ept-5-en e-2-car boxylic Acid 12-Hydr oxy-
d od ecyl Ester (A). Bicyclo[2.2.1]hept-5-ene-2-carboxylic acid
12-(tert-butyldimethylsilyloxy)dodecyl ester (4.543 g, 10.4
mmol) was dissolved in 11 mL of THF at room temperature
under a nitrogen atmosphere. Tetrabutylammonium fluoride
(31.2 mL, 31.2 mmol) was added, and the reaction mixture
was stirred for 22 h at room temperature. Water was added
to the reaction mixture, and the product was extracted with
CH2Cl2. The crude product was purified by liquid chromatog-
raphy on silica gel, eluting with 25:75 ethyl acetate/hexane to
give A as colorless viscous oil: 3.219 g (96%). 1H NMR
(CDCl3): δ 1.22-1.34 (m, 18H), 1.45-1.59 (m, 5H), 1.86 (t of
d, 1H, J ) 3.9 Hz and J ) 11.8 Hz), 2.15 (d of d, 1H, J ) 4.5
Hz and J ) 10.2 Hz), 2.24 (s, 1H, OH), 2.85 (s, 1H), 2.97 (br,
s, 1H), 3.54 (t, 2H, J ) 6.6 Hz, CH2O), 4.01 (t, 2H, J ) 6.8 Hz,
CH2-O-CO), 6.06 (m, 2H, CHdCH). 13C NMR (CDCl3): δ
25.63, 25.77, 28.51, 29.06, 29.29, 29.32, 29.36, 29.38, 29.43,
30.13, 32.59, 41.46, 43.05, 46.19, 46.44, 62.62, 64.44, 135.59,
137.83, 176.20. HRMS: Calcd for C20H35O3 + Na+: 323.2586.
Found: 323.2584.
(B). 5-Norborne-2-carboxylic acid (4.00 g, 29.0 mmol) was
dissolved in 35 mL of CH2Cl2 at 0 °C under a nitrogen
atmosphere. The mixture was stirred at 0 °C for 30 min before
oxalyl chloride (9.96 mL, 116 mmol) was added. The temper-
ature was raised to 25 °C, and the reaction was allowed to
react for 6 h. The CH2Cl2 was evaporated, and a slightly yellow
viscous oil was obtained. The oil was dissolved in 25 mL of
THF under a nitrogen atmosphere. 1,1,1-Tris(hydroxymethyl)-
ethane (6.96 g, 58 mmol) was dissolved in 90 mL of warm THF
and added to the reaction mixture. Pyridine (4.69 mL, 57.9
ROMP of Ma cr om on om er s w ith Gr u bbs’ Secon d Gen -
er a tion Ca ta lyst. The exo-B macromonomers in Table 2 were
polymerized from 0.5 to 2.0 h using the same procedure as
described for the ROMP of macromonomers with Grubbs’ first
generation catalyst.
P olym er iza tion of A. A (1.10 g, 3.4 mmol) was taken into
the glovebox. CH2Cl2 (1.06 mL) was added to the monomer.
Grubbs’ first generation catalyst (27.4 mg) was added to 3.8
mL of CH2Cl2. From this catalyst solution, 0.94 mL was added
to the monomer, and the reaction was allowed to proceed for
4 h. The reaction was removed from the glovebox, and ethyl
vinyl ether (1 mL) was added. The reaction was stirred for 1
h before being precipitated into 20 mL of hexanes. The polymer
was dissolved in THF and precipitated into methanol; this
procedure was repeated three times. The polymer was isolated
1
and dried under vacuum for 48 h. H NMR (CDCl3): δ 1.05-
1.20 (br, m, 17H), 1.41-1.65 (br, m, 6H), 1.82-2.10 (br, m,
2H), 2.41-3.18 (br, m, 3H), 3.57 (br, m, 2H), 4.02 (br, m, 2H),
5.18-5.42 (br, m, 2H). GPC: Mn ) 143 000 g mol-1, PDI )
1.02.
P olym er iza tion of D. D (0.800 g, 1.86 mmol) was added
to a vial in the glovebox. CH2Cl2 (1.5 mL) was added to D.
Grubbs’ first generation catalyst (21.0 mg) was added to 3.4
mL of CH2Cl2. From this catalyst solution, 0.50 mL was added
to the monomer. The reaction was stirred for 1.5 h and
removed from the glovebox. Ethyl vinyl ether (1 mL) was
added, and the reaction was stirred for 1 h. The reaction was
filtered through a plug of silica gel, and the polymer was
precipitated into diethyl ether. The polymer was redissolved
in minimal CH2Cl2 and precipitated into diethyl ether. This
procedure was repeated two times. The polymer was isolated
1
and dried under vacuum for 48 h. H NMR (CDCl3): δ 1.01-
1.20 (br, m, 4H), 1.62 (br, s, 1H), 1.82-2.05 (br, m, 2H), 2.41-
3.18 (br, m, 3H), 3.20 (br, s, 2H), 3.57-3.90 (br, m, 16H), 4.10-
4.38 (br, m, 4H), 5.10-5.41 (br, m, 2H). GPC: Mn ) 226 000
g mol-1, PDI ) 1.12.
Syn th esis of Ar m s of P oly(L-la ctid e) fr om th e Ba ck -
bon e. The same procedure was followed for all entries in Table
4. We will describe the synthesis of entry 5 in Table 4 as an
example. Poly-D (41.4 mg, 96.1 µmol) and L-lactide (2.84 g,
19.7 mmol) were added to a Schlenk flask in the glovebox.
2-Methoxyethyl ether (4.26 mL) was added to the flask. The
flask was sealed, removed from the glovebox, and attached to
a N2 line. It was heated at 110 °C for 20 min to dissolve all of
the polymer. Tin(II) 2-ethylhexanoate (6.2 µL, 19.2 µmol) was
added under a positive pressure of N2. The polymerization was
allowed to run for 2 h at 110 °C and then cooled in ice. The
polymer was dissolved in CHCl3 and precipitated into metha-
nol. This process was repeated three times. The polymer was
dried under vacuum for 48 h. Yield ) 2.65 g (92%). The 1H
NMR was identical to that of homopolylactide.
(ter t-Bu tyldim eth ylsilyloxy)dodecan ol. 1,12-Dodecanediol
(15.00 g, 74.1 mmol) and imidazole (6.31 g, 92.7 mmol) were
dissolved in 75 mL of warm DMF, and the reaction mixture

4368 J ha et al.
Macromolecules, Vol. 37, No. 12, 2004
mmol) was added, and the reaction was allowed to reach
completion overnight. The product was extracted from water
(50 mL) with four portions of CH2Cl2 (50 mL). The organic
phase was evaporated, and the crude product was purified by
liquid chromatography on silica gel, eluting with 40:60 ethyl
acetate/hexane gradually increasing to 50:50 ethyl acetate/
hexane to give B as white crystals: 3.76 g (56%). 1H NMR
(CDCl3): δ 0.84 (s, 3H, CH3), 1.11 (br, s, 1H), 1.35-1.43 (m,
2H), 1.47-1.50 (m, 1H), 1.91 (m, 1H), 2.26 (m, 1H), 3.04 (m,
3H), 3.53 (d of d, 4H, J ) 11.4 Hz and J ) 18.3 Hz, CH2O),
4.18 (d of d, 2H, J ) 11.4 Hz and J ) 17.4 Hz, CH2-O-CO),
6.12 (d of d, 2H, J ) 3 Hz and J ) 5.4 Hz, CHdCH). 13C NMR
(CDCl3): δ 16.80, 30.41, 40.83, 41.59, 43.24, 46.37, 46.67, 66.39,
67.43, 67.48, 135.55, 138.08, 177.41. HRMS: Calcd for C13H21O4
+ Na+: 241.1440. Found: 241.1431.
Sch em e 1. Syn th esis of Hom op olya ctid e fr om a
Sm a ll Molecu le In itia tor
Ta ble 1. Molecu la r Weigh ts of P olyla ctid es Syn th esized
in Sch em e 1
a
predicted
Mn from 1H
Mn from
Mn(GPC)/
Mn(NMR) PDI
Mn (g mol-1
)
NMR (g mol-1
)
GPC (g mol-1
)
5 000
10 000
15 000
20 000
40 000
50 000
75 000
5 200
9 500
14 000
21 000
b
8 600
17 000
25 000
36 000
70 000
80 400
114 000
1.7
1.8
1.8
1.7
b
1.04
1.06
1.06
1.11
1.14
1.16
1.09
Tetr a eth ylen e Glycol Bicyclo[2.2.1]h ep t-5-en e-2-ca r -
boxyla te. 5-Norborne-2-carboxylic acid (4.00 g, 29 mmol) was
dissolved in 35 mL of CH2Cl2 at 0 °C under N2. The reaction
was stirred at 0 °C for 30 min before oxalyl chloride (9.96 mL,
116 mmol) was added. The temperature was raised to 25 °C,
and the reaction was allowed to react for 6 h. The CH2Cl2 was
evaporated, and a slightly yellow viscous oil was obtained. The
oil was dissolved in 15 mL of CH2Cl2 at 0 °C under a nitrogen
atmosphere. Tetraethylene glycol (20.0 mL, 115 mmol) was
dissolved in 25 mL of CH2Cl2 and added to the reaction
mixture. Pyridine (4.69 mL, 57.9 mmol) was added, and the
reaction was allowed to reach completion overnight at room
temperature. Water (100 mL) was added to the reaction
mixture, and the product was extracted with CH2Cl2. The
organic phase was evaporated, and the crude product was
purified by liquid chromatography on silica gel, eluting with
100% ethyl acetate to give tetraethylene glycol bicyclo[2.2.1]-
hept-5-ene-2-carboxylate as a colorless viscous oil: 7.198 g
b
b
b
b
a
b
Calibrated against polystyrene standards. The molecular
weights of the three highest molecular weight polylactides could
not be determined by 1H NMR.
Resu lts a n d Discu ssion
Bottlebr u sh P olym er s fr om th e ROMP of P oly-
(L-la ctid e) Ma cr om on om er s. In this section we will
describe the synthesis and polymerization of mac-
romonomers of polylactide terminated on one end with
a norbornyl group, as shown in Figure 1a. Before we
synthesized the macromonomers, we studied the po-
lymerization of L-lactide to investigate the molecular
weights and polydispersities of the polymers that we
could synthesize (Scheme 1 and Table 1). The polymer-
izations of L-lactide were initiated from 1-hexadecanol,
catalyzed by a Sn(II) catalyst, and run for 2-24 h. Each
polymerization went to >95% conversion as judged by
the amount of polymer isolated at the end of the
reaction. These conditions were chosen as others have
shown that the polymers have narrow polydispersities
and predictable molecular weights.24
1
(80%). H NMR (CDCl3): δ 1.33-1.40 (m, 2H), 1.53 (m, 1H),
1.91-1.96 (m, 1H), 2.24-2.28 (m, 1H), 2.37 (br, s, 1H, OH),
2.92 (s, 1H), 3.05 (m, 1H), 3.59-3.74 (m, 14H, CH2O), 4.25
(m, 2H, CH2-O-CO), 6.12 (m, 2H, CHdCH). 13C NMR
(CDCl3): δ 16.88, 30.11, 41.40, 42.82, 46.06, 46.43, 49.50, 63.12,
63.23, 66.72, 68.55, 68.96, 70.15, 70.25, 70.33, 135.47, 137.84,
175.30, 176.00. HRMS: Calcd for C11H21O6 (parent peak minus
cyclopentadiene): 249.1338. Found: 249.1328.
(C). Tetraethylene glycol bicyclo[2.2.1]hept-5-ene-2-carbox-
ylate (5.00 g, 15.9 mmol), isopropylidine-2,2-bis(methoxy)-
propionic acid (3.87 g, 22.2 mmol), 4-(dimethylamino)pyridine
(555 mg, 4.54 mmol), and p-toluenesulfonic acid monohydrate
(605 mg, 3.18 mmol) were mixed in 75 mL of CH2Cl2 under a
nitrogen atmosphere. DCC (5.91 g, 28.6 mmol) was added. The
reaction was stirred at room temperature for 15 h. The DCU
was filtered with a glass frit and washed with a small volume
of CH2Cl2. The crude product was purified by liquid chroma-
tography on silica gel, eluting with 60:40 ethyl acetate/hexane
to give C as colorless viscous oil: 4.96 g (66%). 1H NMR
(CDCl3): δ 1.15 (s, 3H, CH3), 1.21 (m, 1H), 1.29-1.43 (m, 8H),
1.90-1.94 (m, 1H), 2.27 (m, 1H), 2.92 (s, 1H), 3.04 (s, 1H),
3.60-3.72 (m, 14H, CH2O), 4.21-4.32 (m, 6H), 6.12 (m, 2H,
CHdCH).13C NMR (CDCl3): δ 18.65, 23.04, 24.19, 30.32, 41.61,
41.74, 41.99, 43.01, 46.26, 46.65, 63.43, 63.83, 65.90, 68.99,
69.23, 70.52, 70.57, 70.60, 98.01, 135.68, 138.05, 174.10, 176.18.
HRMS: Calcd for C24H38O9 + Na+: 493.2414. Found: 493.2404.
(D). C (4.06 g, 10.5 mmol) was dissolved in 80 mL of
methanol. One teaspoon of DOWEX 50X2-400 was added, and
the reaction mixture was stirred for 7 h at room temperature.
DOWEX 50X2-400 was filtered and carefully washed with
methanol. The filtrate was evaporated, and the crude product
was purified by liquid chromatography on silica gel, eluting
with 100% ethyl acetate to give D as a colorless viscous oil:
3.85 g (85%). 1H NMR (CDCl3): δ 1.17 (s, 3H, CH3), 1.25-
1.31 (m, 2H), 1.44 (m, 1H), 1.83 (m, 1H), 2.17 (m, 1H), 2.83 (s,
1H), 2.96 (m, 1H, OH), 3.19 (s, 2H), 3.58-3.73 (m, 16H, CH2O),
4.16 (m, 2H, CH2-O-CO), 4.24 (m, 2H, CH2-O-CO), 6.03 (m,
2H, CHdCH). 13C NMR (CDCl3): δ 16.88, 30.11, 41.40, 42.82,
46.06, 46.43, 49.50, 63.12, 63.23, 66.72, 68.55, 68.96, 70.15,
70.25, 70.33, 135.47, 137.84, 175.30, 176.00. HRMS: Calcd for
These polymers scattered light poorly; thus, we could
not determine their absolute molecular weights using
multiangle laser light scattering. Instead, we used two
techniques to measure their molecular weights. In one
method we used gel permeation chromatography (GPC)
with a refractive index detector to measure the molec-
ular weights against polystyrene standards. In the
second method we calculated the molecular weights
1
from H NMR using end-group analysis as described in
the Experimental Section.
From the results in Table 1 we see that the predicted
molecular weights closely matched those measured by
1H NMR. These polymerizations were well controlled
as shown by the narrow polydispersities for all molec-
ular weights measured and the close agreement between
the predicted molecular weights and the molecular
1
weights measured by H NMR. The ratio of the molec-
ular weights measured against polystyrene standards
1
to those measured by H NMR were close to 1.7 and
can be understood by differences in hydrodynamic
volume for polystyrene relative to polylactide.25 These
experiments demonstrated that we synthesized high
molecular weight polylactide with low polydispersities.
Synthesis of the Macromonomers. We choose nor-
bornene-based monomers for the synthesis of the back-
bone for three reasons (Figure 2). First, these monomers
have large ring strains and are readily polymerized by
C
16H29O9 (parent peak minus cyclopentadiene): 365.1812.
Found: 365.1803.

Macromolecules, Vol. 37, No. 12, 2004
Bottlebrush Polymers from Grubbs’ Catalysts 4369
F igu r e 3. (a) GPC of exo-A-4300. (b) GPC of a bottlebrush
polymer synthesized from exo-A-4300 with a macromonomer-
to-catalyst ratio of 580/1 (entry 2 in Table 3). (c) GPC of a
bottlebrush polymer synthesized from exo-A-4300 with a
macromonomer-to-catalyst ratio of 1160/1 (entry 3 in Table
3). The conversion of macromonomer to the bottlebrush
polymer was greater than 95%.
F igu r e 2. Synthesis of monomers A, B, and D that we used
in this study. The full synthetic procedures are described in
the Experimental Section.
Sch em e 2. Gr u bbs’ F ir st, E, a n d Secon d , F ,
Gen er a tion Ca ta lysts
Ta ble 2. Ma cr om on om er s Th a t Wer e Syn th esized Usin g
La ctid e a n d Tin (II) 2-Eth ylh exa n oa te
c
Mn
PDI duration of
polymeri-
GPC zationb (h)
from GPC from
initiatora
macromonomer
(g mol-1
)
exo- A
exo-A
exo-B
exo-B
exo-A-4300
exo-A-19000
exo-B-9800
exo-B-52000
4 300
19 000
9 800
52 000
11 000
1.06
1.08
1.09
1.09
1.19
4
15
10
24
12
catalyst to greater than 95% conversion (Figure 3 and
Table 3). Small high and low molecular weight shoul-
ders were seen for the peaks, as shown in Figure 3b,c.
Despite these shoulders, the polydispersities of the
polymers were low. The polymerizations in Table 3 were
run from 4 to 24 h, and extended reaction times did not
significantly diminish the small amount of unreacted
macromonomer that was present. Our polymers had
high degrees of polymerization along the backbone, and
the measured molecular weights matched the predicted
molecular weights well. We note that this catalyst has
been described as living for polymerizations of low
molecular weight monomers, but we do not make this
claim for polymerizations of macromonomers. More
work would be needed to show that this catalyst is living
for the ROMP of macromonomers.
We polymerized exo-A-4300 with Grubbs’ second
generation catalyst (entry 4 in Table 3). This polymer-
ization resulted in greater than 95% conversion of
macromonomer to bottlebrush polymer, but the mea-
sured molecular weight was significantly higher than
the predicted molecular weight. This result was consis-
tent with the high rate of propagation relative to the
rate of initiation for Grubbs’ second generation cata-
lyst.16-18
endo/exo-B endo/exo-B-11000
a
The prefix “exo” refers to norbornenes that are solely the
exo isomer. The prefix “endo/exo” refers to norbornenes that
b
are approximately 80% endo and 20% exo. The yields of the
polymerizations based on isolated polymer were greater than
90%. c The molecular weights were measured against polylactide
standards. Macromonomers initiated from exo-B had two arms
with combined molecular weights of the values reported.
Grubbs’ catalysts at room temperature. Second, these
monomers are synthesized in a few steps. Third, the
pure exo monomer can be isolated. The exo monomer is
more active than the endo monomer toward polymeri-
zation; thus, the choice of monomers can tune the rate
of polymerization.
Norbornenes with one or two alcohols were used as
initiators for the synthesis of the macromonomers. In
our synthesis each alcohol becomes an arm; thus, we
could vary the density of arms along the backbone
through our choice of macromonomer. In Table 2 we
show the results of the synthesis of our macromonomers.
To characterize these polymers, we used the poly-
lactide standards that were synthesized in Table 1 and
1
the molecular weights found using H NMR spectros-
copy to calibrate the GPC. Although the molecular
weights found using H NMR spectroscopy are values
1
We wished to test the limits of this method by
polymerizing macromonomers with high molecular
weights. One possible limitation to the polymerization
of high molecular weight macromonomers was the high
dilution necessary to completely dissolve the macro-
monomer. For example, in the ROMP of exo-A-4300 the
for the number-average molecular weight, Mn, rather
than the molecular weight at the top of the peak, Mp,
the values for Mn and Mp will be close to one another
due to the narrow polydispersities of these polymers.
We sacrificed a small amount of error in the value for
Mp to use polylactide standards rather than polystyrene
standards. Each macromonomer was synthesized with
a low polydispersity, including those synthesized from
an initiator with two alcohols.
Polymerization of Macromonomers Using Grubbs’
Catalysts. We investigated the polymerization of macro-
monomers with one polylactide per norbornene using
Grubbs’ first generation catalyst (E in Scheme 2). Exo-
A-4300 polymerized well with Grubbs’ first generation
concentration of E was approximately 0.11 mg mL-1
.
This concentration may be compared to the ROMP of
monomers A and D as described in the Experimental
Section where the concentration of E was 3.4 and 2.4
mg mL-1. To polymerize macromonomers with molec-
ular weights greater than 4300 g mol-1, the concentra-
tion of E was less than 0.11 mg mL-1. Thus, we expected
the polymerizations of high molecular weight macro-
monomers to be slow. Experiments to polymerize exo-

4370 J ha et al.
Macromolecules, Vol. 37, No. 12, 2004
Sch em e 3. Ma cr om on om er s w ith On e P olyla ctid e Ch a in p er Nor bor n en e P olym er ized Well w ith E, bu t
P olym er iza tion s w ith F Wer e Un con tr olled ; Ma cr om on om er s w ith Tw o P olyla ctid e Ch a in s p er Nor bor n en e
P olym er ized Ver y Slow ly w ith E, bu t P olym er iza tion s w ith F P r oceed ed Ra p id ly a n d Resu lted in P olym er s w ith
Na r r ow P olyd isp er sities a n d P r ed icted Molecu la r Weigh ts
Ta ble 3. Syn th esis of Bottlebr u sh P olym er s fr om Ma cr om on om er s
b
macromonomer/
catalysta
predicted Mn
(g mol-1
measured Mn
(g mol-1
entry
macromonomer
catalyst
)
)
PDI
timec (h)
1
2
3
4
5
6
7
8
9
exo-A-4300
exo-A-4300
exo-A-4300
exo-A-4300
exo-A-19000
exo-B-9800
exo-B-9800
exo-B-9800
exo-B-9800
E
E
E
F
E
E
E
F
F
F
F
F
290/1
580/1
1160/1
1160/1
660/1
51/1
1020/1
255/1
510/1
1020/1
230/1
460/1
1 250 000
2 500 000
5 000 000
5 000 000
12 500 000
500 000
9 800 000
2 500 000
5 000 000
10 000 000
2 500 000
5 000 000
1 800 000
2 700 000
5 200 000
13 000 000
6 400 000
430 000
1 000 000
1 900 000
4 700 000
12 000 000
940 000
1.05
1.07
1.19
1.30
1.25
1.03
1.07
1.39
1.22
1.17
1.13
1.08
4
15
23
1.8
22
4
25
1
1
10
11
12
exo-B-9800
endo/exo-B-11000
endo/exo-B-11000
1
1.5
1.5
520 000
a
b
The ratio of macromonomer to Grubbs’ catalyst. Measured by GPC and MALLS. c The duration of the polymerizations. The yields
for recovery of macromonomer and bottlebrush polymer were greater than 90% for each reaction.
A-19000 with E resulted in low conversions at long
reaction times. For example, in entry 5 exo-A-19000
polymerized to 51% conversion after 22 h. Attempts to
polymerize this macromonomer with F resulted in
polymers with several peaks in the GPC.
We next investigated the polymerizations of macro-
monomers with two polylactide arms per norbornene.
We attempted to polymerize exo-B-9800 with E , but
the conversions were low. For example, in entry 7 in
Table 3 the polymerization went to 10% conversion after
25 h. To investigate whether the concentration of the
norbornene ends of the macromonomer and catalyst
were responsible for the slow polymerization, we low-
ered the ratio of macromonomer to catalyst to 51/1 as
shown in entry 6. The concentrations of the norbornene
chain ends and the catalyst were higher than that in
F igu r e 4. (a) GPC of exo-D-9800. (b) GPC of a bottlebrush
polymer synthesized from exo-D-9800 (entry 9 in Table 3).
entry 1, yet this polymerization went to only 86%
conversion after 4 h. Similar polymerizations with exo-
A-4300 went to greater than 95% conversion in 4 h.
Thus, E was found to be a slow and ineffective catalyst
for the polymerization of norbornene with two poly-
polylactide arms. This hindrance may prevent the
catalyst from reacting with these olefins.
We investigated the polymerization of exo-B-9800
with F at various ratios of macromonomer to catalyst
(entries 8-10 in Table 3). These polymerizations went
to greater than 95% conversion within 2 h (Figure 4
and Scheme 3). The agreement between the predicted
molecular weight and the measured molecular weight
was excellent, and the polydispersities decreased as the
macromonomer-to-catalyst ratio was raised.
lactide arms.
It is well-known that F is a more active catalyst than
E.16,18,20 Despite its superior activity, F is rarely used
as a ROMP catalyst as the polymerizations are not
living. Two proposed reasons for the lack of living
character of this catalyst are that it has a high value
for the ratio of rate of propagation, kp, to rate of
initiation, ki, and the catalyst chain transfers to olefins
along the backbone of the polymers.16,18,20 We hypoth-
esized that our macromonomers would address these
two problems due to their unique characteristics. First,
steric crowding develops between macromonomers as
they are added to the growing bottlebrush polymer. This
steric crowding only occurs after multiple macro-
monomers have been added to the polymer. Thus, we
expect that the steric crowding will affect kp more than
ki and lower the ratio of kp/ki. Second, access to the
olefins along the backbone is sterically hindered by the
To study whether we could polymerize high molecular
weight macromonomers with two polylactide arms per
norbornyl group, we investigated the polymerization of
exo-B-52000 with F . These polymerizations were not
well controlled. At a macromonomer-to-catalyst ratio of
250/1 the polymerization went to completion, but there
were several broad peaks in the GPC. Thus, the polym-
erization of the macromonomers with two polylactide
arms per norbornene is limited to those with molecular
weights lower than 52 000 g mol-1
.

Macromolecules, Vol. 37, No. 12, 2004
Bottlebrush Polymers from Grubbs’ Catalysts 4371
F igu r e 5. We polymerized norbornyl monomers to yield well-defined backbone polymers. Polylactides were synthesized from
the alcohols on the backbone.
Ta ble 4. P olym er iza tion of La ctid e fr om Ba ck bon e P olym er s a s Sh ow n in F igu r e 5
predicted Mn
predicted Mn
measured for bottlebrush polymer
for each armb
for bottlebrush polymer
d
entry
backbonea
(g mol-1
)
(g mol-1
)
Mwc (g mol-1
)
PDId
Rzc (nm)
Rh (nm)
Rz/Rh
1
2
3
4
5
6
poly-A
poly-A
poly-A
poly-D
poly-D
poly-D
15 000
30 000
50 000
15 000
30 000
50 000
4 700 000
9 400 000
15 700 000
15 800 000
31 600 000
52 300 000
4 990 000
8 920 000
13 600 000
18 300 000
36 100 000
64 000 000
1.05
1.02
1.02
1.03
1.21
e
62
64
80
110
131
243
48
54
55
53
65
e
1.3
1.2
1.5
2.1
2.0
a
Poly-A had a Mn ) 143 000 g mol-1; PDI ) 1.02; degree of polymerization ) 444. Poly-D had a Mn ) 226 000 g mol-1; PDI ) 1.12;
b
degree of polymerization ) 525. We assumed a 100% conversion of lactide to calculate the Mn for each arm. For poly-D each alcohol was
treated as a separate arm. c Measured from a Zimm plot. Measured with a GPC and a multiangle laser light spectrometer. e This polymer
d
was too large to pass through the GPC columns.
We investigated the polymerization of endo/exo-B-
11000 with F (entries 11 and 12 in Table 3). The
polymerizations were slow and went to 40% and 10%
conversions for macromonomer-to-catalyst ratios of
230/1 and 460/1 after 1.5 h. For comparison, the
polymerizations of exo-B-9800 with E were run for 1 h
and resulted in conversions of greater than 95%.
Bottlebr u sh P olym er s by P olym er izin g La ctid e
fr om Ba ck bon e P olym er s. We developed a second
method to synthesize bottlebrush polymers (Figure 5).
In this method we synthesized the backbone first and
grafted polylactide arms onto it in the next step. We
wished to study whether this method would yield higher
molecular weight bottlebrush polymers than were syn-
thesized using the macromonomer approach.
We first investigated the polymerizations of A, B, and
D (Figure 2). Polymers synthesized from A were soluble
in methylene chloride and chloroform. The polymeriza-
tions of A proceeded smoothly at all concentrations we
attempted. Notably, we synthesized a polymer with a
molecular weight of 1 240 000 g mol-1 and a polydis-
persity of 1.13. This polymer has a degree of polymer-
ization of 4000 and may be used to synthesize bottle-
brush polymers with long backbones. We investigated
monomers structurally related to A but synthesized
using 1,6-hexanediol and 1,10-decanediol. Polymers
synthesized from these monomers were insoluble in
methylene chloride and chloroform and not pursued.
Polymers synthesized from B were insoluble in meth-
ylene chloride, so we developed a short synthesis of D.
This monomer yielded polymers with narrow polydis-
persities (PDI < 1.2) and degrees of polymerization up
to 1500 along the backbone. Monomers synthesized in
an analogous manner to D but with triethylene glycol
resulted in polymers that were insoluble in CH2Cl2 and
were not pursued.
1-24 h.24 When we attempted to grow polylactide from
the backbone polymers under these conditions, the
bottlebrush polymers were insoluble. The backbone
polymer never dissolved in the molten lactide, and
extensive cross-linking occurred.
To dissolve the backbone polymer, we added 2-meth-
oxyethyl ether as a solvent to the polymerization. In
addition, we changed the ratio of initiator to catalyst
from 50/1 for neat polymerizations to 10/1 for polymer-
izations with 2-methoxyethyl ether. Polymerizations
with an initiator-to-catalyst ratio of 50/1 using 2-meth-
oxyethyl ether were slow, and conversions were low.
Polymerizations with solvent and a 10/1 initiator-to-
catalyst ratio yielded excellent results, as shown in
Table 4. Under these conditions the polymer dissolved
in the lactide and solvent prior to the addition of the
tin catalyst. We synthesized bottlebrush polymers with
ultrahigh molecular weights and narrow polydispersi-
ties. The polydispersities were narrow even for the
highest molecular weight polymer that passed through
the GPC columns. In addition, one should note the ratio
of Rz/Rh, where Rz (m) is the root-mean-square radius
and Rh (m) is the hydrodynamic radius. For values of
less than 1.5, this ratio indicates that the polymer is a
random coil or a sphere.25 For values of 2.0, this ratio
indicates that the polymer is shaped as a rigid rod.25
Thus, our initial results indicate that we synthesized
bottlebrush polymers shaped as rigid rods and spheres.
In Figure 6, we show the Zimm plot of entry 6 in Table
4. The molecular weights of the bottlebrush polymers
were ultralarge so the assumptions used to measure
their molecular weights using GPC/MALLS were not
valid. In particular, the second viral coefficient, A2, is
set equal to zero in GPC/MALLS systems so the absolute
molecular weights of polymers can be measured.14 This
assumption is valid for polymers with molecular weights
below approximately 1 million g mol-1, but our polymers
had molecular weights that were much higher. We
measured the Zimm plots for the polymers in Table 4
Polymerization of L-Lactide. We investigated the
polymerization of L-lactide from the backbone. Typically,
L-lactide polymerizations are run neat at 110 °C for

4372 J ha et al.
Macromolecules, Vol. 37, No. 12, 2004
F igu r e 7. Relationship between the intensity and scattering
angle for the GPC/MALLS trace of entry 5 in Table 4. The
strong dependence of angle on the scattering intensity indi-
cates that the polymer has an ultrahigh molecular weight.
F igu r e 6. Zimm plot of entry 6 in Table 4. We show this plot
as this polymer has the highest molecular weight in Table 4.
The Zimm plots were increasingly difficult to plot as the
molecular weight increased. The 5245 in the x axis is a stretch
factor that is used to put the data in a clear format. The data
were fitted to a quadratic equation.
medium molecular weights (up to 9800 g mol-1). At
higher molecular weights the polymerizations of the
macromonomers were not well controlled and had low
conversions, or multiple peaks were seen in the GPC.
During the course of this work we discovered the first
example where Grubbs’ second generation catalyst may
be considered “living” as applied toward the synthesis
of polymers using ROMP. As mentioned previously, this
catalyst is not considered living as its rate of initiation
is slow relative to its rate of propagation, and it reacts
with olefins along the polymer backbone. In our work
we used Grubbs’ second generation catalyst to synthe-
size polymers with narrow polydispersities and molec-
ular weights that were predicted on the basis of the ratio
of macromonomer to catalyst. These results strongly
suggest, but do not prove, that polymerizations of
macromonomers with Grubbs’ second generation cata-
lyst were living.
We hypothesize that at least three factors contributed
to the success of these polymerizations with Grubbs’
second generation catalyst. First, the polymerizations
were run at much lower concentrations than typical
ROMP experiments. The concentrations may have an
as yet unexplored effect on these polymerizations.
Second, steric interactions between an incoming macro-
monomer and those already polymerized should lower
the rate of polymerization but have little effect on the
rate of initiation. Third, the rate of chain transfer to
polymer may be slowed due to the steric crowding of
the polylactide arms along the backbone.
One piece of evidence that supports our hypothesis
is the comparison of the ROMP of exo-A-4300 and exo-
B-9800. These macromonomers had similar values for
the molecular weights of the polylactide arms but
differed in the number of arms per norbornene (exo-B-
9800 had two arms of average Mn of 4900 g mol-1 and
exo-A-4300 had one arm with Mn of 4300 g mol-1). The
steric crowding of polymerizing exo-A-4300 should be
less than the steric crowding of polymerizing exo-B-
9800. The polymerization of exo-A-4300 with Grubbs’
second generation catalyst was uncontrolled, and the
measured molecular weight was much higher than the
predicted molecular weight. This result is consistent
with a low value for ki/kp. The polymerization of exo-
B-9800 with Grubbs’ second generation catalyst was
well controlled, and the measured molecular weights
matched the predicted molecular weights. These results
are consistent with the hypothesis that steric crowding
may influence the reactivity of the catalyst but does not
prove it. More experiments are needed to determine the
reasons behind the success of these macromonomers
with Grubbs’ second generation catalyst.
as these plots report the values for Mw, Rg, and A2
independent of any assumptions about A2. We then used
the values for A2 from the Zimm plots to measure
polydispersities of the polymers from the GPC spectra.
The GPC spectra of entries 4-6 in Table 4 were
unique. The peaks were broad and monomodal and had
confusing retention times. The retention times of the
peaks ranged from 40 to 65 min. For comparison, the
retention times of small molecules (such as monomer)
had retention times of 42-44 min in our GPC, and
linear polymers had retention times from 22 to 35 min.
Despite these characteristics, the peaks exhibited a
strong dependence for light scattering intensity as a
function of angle of detection (see Figure 7), and values
for the molecular weights and Rg were constant over
the entire peak. The polymers had narrow polydisper-
sities despite their broad appearance. We do not offer
an explanation of this result but merely point it out.
We only observed this result for the last three entries
in Table 4 where light scattering data indicate the
polymers were shaped as rigid rods.
One method to determine the shape of the polymer
is to measure the Kuhn length from GPC and MALLS
data. The Kuhn length is the distance between kinks
in the polymer; thus, polymers with large Kuhn lengths
are mostly linear and shaped as rigid rods. To determine
the Kuhn length, a plot of Rg vs Mw must be modeled
using different values for the Kuhn length. As our
bottlebrush polymers had narrow polydispersities, the
plot of Rg vs Mw did not cover a range large enough to
find a value for the Kuhn length.
In Figure 7 we show light scattering data from the
peak corresponding to entry 5 in Table 4. This figure
demonstrates two important points. First, the GPC peak
is monomodal and lacks high or low molecular weight
humps. High molecular weight humps would indicate
that the bottlebrush polymers are cross-linking. Low
molecular weight humps would indicate that homopoly-
lactide is being synthesized. Thus, the bottlebrush
polymers are free of homopolylactide and not cross-
linked. Second, the light scattering data show a strong
angle dependence that indicates that the polymer has
a high molecular weight.
Com p a r ison of th e Tw o Meth od s To Syn th esize
Bottlebr u sh P olym er s. In the first method we polym-
erized polylactide macromonomers to yield bottlebrush
polymers with degrees of polymerization up to 1160
along the backbone and narrow polydispersities. This
method worked well for macromonomers with low to

Macromolecules, Vol. 37, No. 12, 2004
Bottlebrush Polymers from Grubbs’ Catalysts 4373
Huang, H.; Remsen, E. E.; Kowalewski, T.; Wooley, K. L. J .
Am. Chem. Soc. 1999, 121, 3805-3806. (g) Reches, M.; Gazit,
E. Science 2003, 300, 625-627. (h) Discher, D. E.; Eisenberg,
A. Science 2002, 297, 967-973. (i) Wooley, K. L.; Moore, J .
S.; Wu, C.; Yang, Y. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
11147-11148. (j) Ma, Q.; Remsen, E. E.; Clark, C. G., J r.;
Kowalewski, T.; Wooley, K. L. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 5058-5063. (k) Liu, T.; Xie, Y.; Chu, B. Langmuir
2000, 16, 9015-9022. (l) Lui, G.; Yan, X.; Qiu, X.; Li, Z.
Macromolecules 2002, 35, 774-7747. (m) Schnur, J . M.
Science 1993, 262, 1669-1676. (n) Fuhrhop, J .-H.; Spiroski,
D.; Boettcher, C. J . Am. Chem. Soc. 1993, 115, 1600-1601.
(o) Mertig, M.; Kirsch, R.; Pompe, W. Appl. Phys. A 1998,
66, 723-727. (p) Khazanovich, N.; Granja, J . R.; McRee, D.
E.; Milligan, R. A.; Ghadiri, M. R. J . Am. Chem. Soc. 1994,
116, 6011-6012.
In our second method to synthesize bottlebrush
polymers we polymerized lactide from a backbone. This
method was more versatile than the first method. We
synthesized backbones with degrees of polymerization
up to 4000. For our experiments we used polymers with
degrees of polymerization of approximately 500 so the
final molecular weights of the bottlebrush polymers
would be within a range that we could easily character-
ize. We demonstrated that lactide could be polymerized
from the backbone with little or no cross-linking be-
tween the bottlebrush polymers. In addition, we were
able to synthesize arms with molecular weights up to
approximately 50 000 g mol-1. The arms of the bottle-
brush polymers synthesized using this method were of
much higher molecular weight than those synthesized
using the macromonomer method.
(3) J ain, S.; Bates, F. S. Science 2003, 300, 460-464.
(4) Percec, V.; Ahn, C.-H.; Ungar, G.; Yaeardley, D. J . P.; Moller,
M.; Sheiko, S. S. Nature (London) 1998, 391, 161-164.
(5) (a) Borner, H. G.; Beers, K.; Matyjaszewski, K.; Sheiko, S.
S.; Moller, M. Macromolecules 2001, 34, 4375-4383. (b)
Wintermantel, M.; Gerle, M.; Fischer, K.; Schmidt, M.;
Wataoka, I.; Urakawa, H.; Kajiwara, K.; Tsukahara, Y.
Macromolecules 1996, 29, 978-983.
One disadvantage of this method is that we lacked
an opportunity to directly measure the molecular weights
or polydispersities of the arms. Thus, the arms were not
fully characterized, and we must be careful when
interpreting these results. We expect that the polydis-
persities of the arms were narrow on the basis of two
related experiments. First, the polydispersities of homo-
polylactide were narrow as shown in Tables 1 and 2.
Second, we polymerized L-lactide using 1-hexadecanol
in 2-methoxyethyl ether at the same concentrations as
used to synthesize the bottlebrush polymers. These
homopolylactides had narrow polydispersities (PDI’s
from 1.11 to 1.28) for molecular weights from 5000 to
(6) (a) Deimede, V.; Kallitsis, J . K. Chem.sEur. J . 2002, 8,
467-473. (b) Grande, D.; Six, J .-L.; Breunig, S.; Heroguez,
V.; Fontanille, M.; Gnanou, Y. Polym. Adv. Technol. 1998,
9, 601-612. (c) Heroguez, V.; Amedro, E.; Grande, D.;
Fontanille, M.; Gnanou, Y. Macromolecules 2000, 33, 7241-
7248. (d) Nomura, K.; Takahashi, S.; Imanishi, Y. Macro-
molecules 2001, 34, 4712-4723. (e) Tsukahara, Y.; Kohjiya,
S.; Tsutsumi, K.; Okamoto, Y. Macromolecules 1994, 27,
1662-1664. (f) Tsukahara, Y.; Namba, S.; Iwasa, J .; Nakano,
Y.; Kaeriyama, K.; Takahashi, M. Macromolecules 2001, 34,
2624-2629. (g) Wintermantel, M.; Fischer, K.; Gerle, M.;
Ries, R.; Schmidt, M.; Kajiwara, K.; Urakawa, H.; Wataoka,
I. Angew. Chem., Int. Ed. Engl. 1995, 34, 1472-1474. (h)
Yamada, K.; Miyazaki, M.; Ohno, K.; Fukuda, T.; Minoda,
M. Macromolecules 1999, 32, 290-293. (i) Dziezok, P.; Sheiko,
S. S.; Fischer, K.; Schmidt, M.; Moller, M. Angew. Chem., Int.
Ed. Engl. 1997, 36, 2812-2815. (j) Djalali, R.; Hugenberg,
N.; Fischer, K.; Schmidt, M. Macromol. Rapid Commun.
1999, 20, 444-449. (k) Djalali, R.; Li, S.-Y.; Schmidt, M.
Macromolecules 2002, 35, 4282-4288. (l) Cheng, G.; Boker,
A.; Zhang, M.; Krausch, G.; Muller, A. H. E. Macromolecules
2001, 34, 6883-6888. (m) Hartikainen, J .; Lahtinen, M.;
Torkkeli, M.; Serimaa, R.; Valkonen, J .; Rissanen, K.; Ikkala,
O. Macromolecules 2001, 34, 7789-7795. (n) Terao, K.; Takeo,
Y.; Tazaki, M.; Nakamura, Y.; Norisuye, T. Polym. J . 1999,
31, 193-198. (o) Vlassopoulos, D.; Fytas, G.; Loppinet, B.;
Isel, F.; Lutz, P.; Benoit, H. Macromolecules 2000, 33,
5960-5969. (p) Shinoda, H.; Matyjaszewski, K. Macromol.
Rapid Commun. 2001, 22, 1176-1181. (q) Shinoda, H.;
Miller, P. J .; Matyjaszewski, K. Macromolecules 2001, 34,
3186-3194. (r) Hong, S. C.; Pakula, T.; Matyjaszewski, K.
Macromol. Chem. Phys. 2001, 202, 3392-3402. (s) Boerner,
H. G.; Duran, D.; Matyjaszewski, K.; da Silva, M.; Sheiko,
S. S. Macromolecules 2002, 35, 3387-3394. (t) Qin, S.;
Matyjaszewski, K.; Xu, H.; Sheiko, S. S. Macromolecules
2003, 36, 605-612. (u) Pakula, T.; Minkin, P.; Beers, K. L.;
Matyjaszewski, K. Polym. Mater. Sci. Eng. 2001, 84,
1006-1007. (v) Neugebauer, D.; Zhang, Y.; Pakula, T.;
Sheiko, S. S.; Matyjaszewski, K. Macromolecules 2003, 36,
6746-6755. (w) Lecommandoux, S.; Checot, F.; Borsali, R.;
Schappacher, M.; Deffieux, A.; Brulet, A.; Cotton, J . P.
Macromolecules 2002, 35, 8878-8881. (x) Hadijchristidis, N.;
Pitsikalis, M.; Iatrou, H.; Pispas, S. Macromol. Rapid
Commun. 2003, 24, 979-1013. (y) Denesyuk, N. A. Phys. Rev.
E 2003, 68, 031803/031801-031803/031811. (z) Snyder, J .
F.; Hutchison, J . C.; Ratner, M. A.; Shriver, D. F. Chem.
Mater. 2003, 15, 4223-4230. (aa) Meuhlebach, A.; Rime, F.
J . Polym. Sci., Part A: Polym. Chem. 2003, 41, 3425-3439.
(bb) Valkama, S.; Lehtonen, O.; Lappalainen, K.; Kosonen,
H.; Castro, P.; Repo, T.; Torkkeli, M.; Serimaa, R.; ten Brinke,
G.; Leskelae, M.; Ikkala, O. Macromol. Rapid Commun. 2003,
24, 556-560. (cc) Malkoch, M.; Carlmark, A.; Woldegiorgis,
A.; Hult, A.; Malmstro¨m, E. E. Macromolecules 2004, 37,
322-329. (dd) Feast, W. J .; Gibson, V. C.; J ohnson, A. F.;
Khosravi, E.; Mohsin, M. A. Polymer 1994, 35, 3542-3548.
(ee) Feast, W. J .; Gibson, V. C.; J ohnson, A. F.; Khosravi, E.;
Mohsin, M. A. J . Mol. Catal. A: Chem. 1997, 115, 37-42.
(ff) Rizmi, A. C. M.; Khosravi, E.; Feast, W. J .; Mohsin, M.
A.; J ohnson, A. F. Polymer 1998, 39.
50 000 g mol-1
.
Con clu sion s
We report two methods to synthesize bottlebrush
polymers using ROMP of norbornene-based monomers
and ROP of L-lactide. The polymers had several char-
acteristics. First, they had medium to high degrees of
polymerization for the arms and backbones. Second,
they were synthesized from L-lactide and are biologically
compatible. Third, the polymers were shaped as spheres
or cylinders as determined by light scattering.
What is the impact of this work? We synthesized
ultrahigh molecular weight bottlebrush polymers with
narrow polydispersities and biocompatible arms. These
polymers are among the highest molecular weight
bottlebrush polymers synthesized. We also described the
first ROMP experiments where Grubbs’ second genera-
tion catalyst may be described as living. The polymers
offer an excellent opportunity to synthesize controlled
organic nanomaterials that are shaped as spheres or
rigid rods and may be integrated with inorganic nano-
materials.
Ack n ow led gm en t. We gratefully acknowledge the
University of Iowa, Math and Physical Sciences Funding
Program, Carver Scientific Research Initiative Grants
Program, and American Chemical Society Petroleum
Research Fund for funding.
Refer en ces a n d Notes
(1) (a) Alivisatos, P. A. Sci. Am. 2001, 285, 67-73. (b) Seeman,
N. C. Nature (London) 2003, 421, 427-431. (c) Lieber, C. M.
Sci. Am. 2001, 285, 58-64. (d) Dai, H. Acc. Chem. Res. 2002,
35, 1035-1044.
(2) (a) Choucair, A. A.; Kycia, A. H.; Eisenberg, A. Langmuir
2003, 19, 1001-1008. (b) Luo, L.; Eisenberg, A. Langmuir
2001, 17, 6804-6811. (c) Yu, K.; Zhang, L.; Eisenberg, A.
Langmuir 1996, 12, 5980-5984. (d) Zhang, L.; Eisenberg, A.
Science 1995, 268, 1728-1731. (e) Zhang, Q.; Remsen, E. E.;
Wooley, K. L. J . Am. Chem. Soc. 2000, 122, 3642-3651. (f)

4374 J ha et al.
Macromolecules, Vol. 37, No. 12, 2004
(7) (a) Hokajo, T.; Terao, K.; Nakamura, Y.; Norisuye, T. Polym.
J . 2001, 33, 481-485. (b) Nakamura, Y.; Norisuye, T. Polym.
J . 2001, 33, 874-878. (c) Shiokawa, K.; Itoh, K.; Nemoto, N.
J . Chem. Phys. 1999, 111, 8165-8173. (d) Subbotin, A.;
Saariaho, M.; Ikkala, O.; ten Brinke, G. Macromolecules 2000,
33, 3447-3452. (e) Tsoukatos, T.; Pispas, S.; Hadjichristidis,
N. Macromolecules 2000, 33, 9504-9511. (f) Sheiko, S. S.;
Prokhorova, S. A.; Beers, K. L.; Matyjaszewski, K.; Potemkin,
I. I.; Khokhlov, A. R.; Moller, M. Macromolecules 2001, 34,
8354-8360.
(19) (a) McConville, D. H.; Wolf, J . R.; Schrock, R. R. J . Am. Chem.
Soc. 1993, 115, 4413-4414. (b) Schrock, R. R. Tetrahedron
1999, 55, 8141-8153. (c) Kennedy, J . P. J . Polym. Sci., Part
A: Polym. Chem. 1999, 37, 2285-2293. (d) Hadjichristidis,
N.; Pitsikalis, M.; Pispas, S.; Iatrou, H. Chem. Rev. 2001, 101,
3747-3792. (e) Smid, J . J . Polym. Sci., Part A: Polym. Chem.
2002, 40, 2101-2107. (f) Hashimoto, K. Prog. Polym. Sci.
2000, 25, 1411-1462.
(20) Bielawski, C. W.; Grubbs, R. H. Angew. Chem., Int. Ed. 2000,
39, 2903-2906.
(8) (a) Matyjaszewski, K. Polym. Mater. Sci. Eng. 2001, 84, 363-
364. (b) Mijovic, J .; Sun, M.; Pejanovic, S.; Mays, J . W.
Macromolecules 2003, 36, 7640-7651.
(21) (a) Lynn, D. M.; Kanaoka, S.; Grubbs, R. H. J . Am. Chem.
Soc. 1996, 118, 784-790. (b) Weck, M.; Schwab, P.; Grubbs,
R. H. Macromolecules 1996, 29, 1789-1793. (c) Wilhelm, T.
E.; Belderrain, T. R.; Brown, S. N.; Grubbs, R. H. Organo-
metallics 1997, 16, 3867-3869.
(22) Gordon, E. J .; Gestwicki, J . E.; Strong, L. E.; Kiessling, L. L.
Chem. Biol. 2000, 7, 9-16.
(23) Ihre, H.; Hult, A.; Fre´tchet, J . M. J .; Gitsov, I. Macromolecules
1998, 31, 4061-4068.
(24) (a) Ryner, M.; Kajsa, S.; Albertsson, A.-C.; von Schenck, H.;
Svensson, M. Macromolecules 2001, 34, 3877-3881. (b)
Kricheldorf, H. R.; Kreiser-Saunders, I.; Boettcher, C. Poly-
mer 1995, 36, 1253-1259. (c) Moller, M.; Nederberg, F.; Lim,
L. S.; Kange, R.; Hawker, C. J .; Hedrick, J . L.; Gu, Y.; Shah,
R.; Abbott, N. L. J . Polym. Sci., Part A: Polym. Chem. 2001,
39, 3529-3538. (d) Atthoff, B.; Trollsas, M.; Claesson, H.;
Hedrick, J . L. Macromol. Chem. Phys. 1999, 200, 1333-1339.
(e) Int Veld, P. J . A.; Velner, E. M.; Van de Witte, P.;
Hamhuis, J .; Dijkstra, P. J .; Feijen, J . J . Polym. Sci., Part A:
Polym. Chem. 1997, 35, 219-226.
(25) (a) Huglin, M. B., Ed. Light Scattering from Polymer
Solutions; Academic Press: New York, 1972. (b) Rodriguez,
F. Principles of Polymer Systems; Taylor and Francis:
Philadelphia, PA, 1996. (c) Berne, B. J .; Pecora, R. Dynamic
Light Scattering: With Applications to Chemistry, Biology,
and Physics; Dover Publications: Mineola, NY, 2000.
(9) Matyjaszewski, K.; Xia, J . Chem. Rev. 2001, 101, 2921-2990.
(10) Gerle, M.; Fischer, K.; Roos, S.; Muller, A. H. E.; Schmidt,
M. Macromolecules 1999, 32, 2629-2637.
(11) (a) Lahitte, J .-F.; Pelascini, F.; Peruch, F.; Meneghetti, S. P.;
Lutz, P. J . C. R. Chim. 2002, 5, 225-234. (b) Allcock, H. R.;
Laredo, W. R.; de Denus, C. R.; Taylor, J . P. Macromolecules
1999, 32, 7719-7725. (c) Allcock, H. R.; de Denus, C. R.;
Prange, R.; Laredo, W. R. Macromolecules 2001, 34, 2757-
2765.
(12) Pitsikalis, M.; Pispas, S.; Mays, J . W.; Hadjichristidis, N. Adv.
Polym. Sci. 1998, 135, 1-137.
(13) (a) Saariaho, M.; Subbotin, A.; Szleifer, I.; Ikkala, O.;
ten Brinke, G. Macromolecules 1999, 32, 4439-4443. (b)
Fredrickson, G. H. Macromolecules 1993, 26, 2825-2831.
(14) Wyatt, P. J . Anal. Chim. Acta 1993, 272, 1-40.
(15) Schwab, P.; Grubbs, R. H.; Ziller, J . W. J . Am. Chem. Soc.
1996, 118, 100-110.
(16) (a) Sanford, M. S.; Love, J . A.; Grubbs, R. H. J . Am. Chem.
Soc. 2001, 123, 6543-6554. (b) Love, J . A.; Sanford, M. S.;
Day, M. W.; Grubbs, R. H. J . Am. Chem. Soc. 2003, 125,
10103-10309.
(17) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-
29.
(18) Bielawski, C. W.; Grubbs, R. H. Macromolecules 2001, 34,
8838-8840.
MA049647K