Polymeric 19F MRI Contrast Agents Prepared by Ring-Opening Metathesis Polymerization/Dihydroxylation

Article abstract of DOI:10.1021/acs.macromol.0c01585

The capability of ring-opening metathesis polymerization (ROMP) to efficiently incorporate bulky monomers and conserve olefin bonds during polymerization was exploited to design water-soluble fluoropolymers, which were evaluated as potential quantitative 19F magnetic resonance imaging (MRI) contrast agents. The fluoromonomeric units comprised 3, 6, 9, or 18 magnetically equivalent fluorine atoms. Aqueous solubility was achieved through dihydroxylation of the partially unsaturated polymeric backbone and by tetraethylene glycol (TEG)-based linker incorporation, ammonium quaternization, or copolymerization.

Full text of DOI:10.1021/acs.macromol.0c01585

pubs.acs.org/Macromolecules
Article
19
Polymeric F MRI Contrast Agents Prepared by Ring-Opening
Metathesis Polymerization/Dihydroxylation
Iris K. Tennie and Andreas F. M. Kilbinger*
Cite This: https://dx.doi.org/10.1021/acs.macromol.0c01585
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The capability of ring-opening metathesis polymerization
(ROMP) to efficiently incorporate bulky monomers and conserve olefin
bonds during polymerization was exploited to design water-soluble
fluoropolymers, which were evaluated as potential quantitative 19F magnetic
resonance imaging (MRI) contrast agents. The fluoromonomeric units
comprised 3, 6, 9, or 18 magnetically equivalent fluorine atoms. Aqueous
solubility was achieved through dihydroxylation of the partially unsaturated
polymeric backbone and by tetraethylene glycol (TEG)-based linker
incorporation, ammonium quaternization, or copolymerization.
Emulsion formulation can be avoided by enhancing the
aqueous solubility by designing amphiphilic tetraethylene
glycol (TEG)-based fluorinated molecular probes,11 liposomal
formulations with hydrophilic fluorinated molecules,12 or
incorporating poly(ethylene glycol) (PEG) units on the
periphery of dendritic probes.13 Alternatively, sulfoxide-
containing2 statistical copolymers, block copolymers of poly-
(acrylic acid)14 or PEG-11415 as well as PEG-816 and PEG-
40017 units in hyperbranched polymers were implemented to
prevent aggregation of fluorinated units. All of these polymers
were prepared by either the reversible addition−fragmentation
chain transfer (RAFT) or atom transfer radical (ATRP)
polymerization, whereby mainly commercially available 2,2,2-
trifluoroethyl (meth)acrylate monomers2,14,16,17 were used as
fluorinated reporters. Despite the incorporation of water-
INTRODUCTION
■
Magnetic resonance imaging (MRI) plays an important role in
detection of various diseases. Metal-based contrast agents, such
as gadolinium(III) complexes, are often used to lower the
1
1
detection limit in H MRI and enhance the image contrast.
Especially in the unchelated form, they have, however, been
associated with safety concerns, including prolonged accumu-
lation within the body and nephrogenic systemic fibrosis.2
With excellent sensitivity second only to a 1H nucleus,
negligible background signal, and a wide chemical shift
range, it is possible to track 19F MRI contrast agents in a
quantitative manner.3
1
While the H MRI image is mostly composed of the NMR
signals of water protons, the 19F MRI image directly depicts
the fluorinated contrast agent. The image intensity is thus
dependent on the concentration of the NMR detectable 19F
nuclear spins as well as on the longitudinal T1 and transverse
T2 relaxation times of the fluorinated moiety. These are
extremely sensitive to the internal mobility and spatial
arrangement of the nuclear spins as well as to the dipole−
dipole and the Zeeman interaction with the applied magnetic
field. To achieve an optimal signal intensity, the T1/T2 ratio
should be near one. This requires decreasing T1 and increasing
T2 as much as possible when designing fluorinated contrast
agents. Aggregation of the nuclei, on the other hand, leads to a
reduced T2, severe line broadening, and diminution of the
NMR signal.4,5
solubilizing segments most of these copolymers exhibit low 19
F
T2 values and high T1/T2 ratios in aqueous solutions. It is only
very recently that water-soluble 2,2,2-trifluoroethyl acrylamide-
based homofluoropolymers featuring polar sulfoxide groups
have been prepared by Whittaker and colleagues.18
Ring-opening metathesis polymerization (ROMP), on the
other hand, is known as one of the fastest, most versatile, and
functional-group-tolerant polymerization approaches that is
also capable of efficiently polymerizing bulky monomers.
Polydispersities are low and the polymerization degrees can be
precisely controlled by altering the monomer to catalyst ratio,
whereby monomers comprise mono- or oligocyclic olefins.19
One of the strategies to prevent aggregation is to deliver
highly fluorinated probes such as perfluorooctyl bromide,6
perfluoropolyether (PFPE),7 perfluoro-15-crown-5-ether,8 and
PERFECTA (36 equivalent 19F atoms)9 as nanoemulsions
with a droplet size in the range of 100−200 nm. Such
emulsions are very lipophilic with long retention times in
internal organs and unstable especially in blood.10
Received: July 25, 2020
Revised: November 2, 2020
https://dx.doi.org/10.1021/acs.macromol.0c01585
© XXXX American Chemical Society
Macromolecules XXXX, XXX, XXX−XXX
A

Macromolecules
pubs.acs.org/Macromolecules
Article
reaction mixture was stirred for 36 h at 60 °C. Acetone was then
removed using a rotary evaporator, H2O was added, and the mixture
was extracted with dichloromethane (DCM). The resulting oil was
purified by column chromatography to provide the colorless product
(2.8 g, 78%). 1H NMR (300 MHz, CDCl3) δ 6.26 (t, J = 1.8 Hz, 2H),
3.86−3.73 (m, 4H), 3.73−3.63 (m, 2H), 3.30−3.19 (m, 2H), 2.67 (d,
J = 1.2 Hz, 2H), 1.52−1.44 (m, 1H), 1.31 (d, J = 9.9 Hz, 1H). 19F
NMR (282 MHz, CDCl3) δ −74.22 (s). 13C NMR (75 MHz, CDCl3)
δ 177.89 (s), 137.78 (s), 123.76 (q, J = 279.5 Hz), 79.10−73.55 (m),
68.06 (s), 67.78 (q, J = 34.2 Hz), 47.79 (s), 45.24 (s), 42.55 (s),
37.40 (s). HRMS (ESI) m/z C13H15F3NO3 [M + H]+: calculated
290.1036, found 290.1003; C13H14F3NO3Na [M + Na]+: calculated
312.0824, found 312.0821.
Among these, a range of norbornene-imides have been
successfully applied to prepare fluorine-18 nanoprobes for
positron emission tomography20 as well as Gd3+-DOTA-
based21 and nitroxide-based22 1H MRI contrast agents.
Another very useful property of ROMP is the conservation
of main chain olefin bonds during the polymerization.
Dihydroxylation of such double bonds catalyzed by osmium
tetroxide has been shown to improve aqueous solubility.23−25
Although this feature offers great possibilities to completely
alter the polarity of the polymeric backbone, it has mostly been
overlooked and not fully explored.
In this work, water-soluble 19F MRI contrast agents have
been prepared by dihydroxylation of the main chain olefins in
fluorinated ROMP polymers. This was achieved without the
need for additional solubilizing segments. Polymers with a
higher fluorine content (>21 wt %), on the other hand, did
require additional modifications such as quaternization of the
tertiary amines or copolymerization with hydrophilic TEG-
based monomers. In copolymers, a linear dependence of the
19F NMR signal intensity on the polymer content was observed
in a wide range of molecular masses.
Perfluoro-tert-butoxyethyl-cis-norbornene-exo-2,3-dicar-
boxiimide (M2). N-(Hydroxyethyl)-cis-norbornene-exo-2,3-dicar-
boxiimide (S2, 486 mg, 2.35 mmol) synthesized according to a
previous procedure26 and triphenylphosphine (738 mg, 2.81 mmol)
were dissolved in dry tetrahydrofuran (THF) (10 mL). Next, DIAD
(446 μL, 2.81 mmol) predissolved in 8 mL of THF was added
dropwise to the reaction mixture over 5 min. Then, perfluoro-tert-
butanol in 9 mL of THF was added at 0 °C and stirred overnight at
RT. The mixture was placed in the fridge for a few hours and the
residual precipitate was removed. The crude material was purified by
column chromatography with hexane/ethyl acetate (4:1, v/v) and the
1
product was obtained as a white crystalline solid (800 mg, 83%). H
EXPERIMENTAL SECTION
NMR (300 MHz, CDCl3) δ 6.22 (t, J = 1.8 Hz, 2H), 4.13 (t, J = 5.3
Hz, 2H), 3.74 (t, J = 5.3 Hz, 2H), 3.27−3.14 (m, 2H), 2.63 (d, J = 1.3
Hz, 2H), 1.49−1.38 (d, 1H), 1.19 (d, J = 10.4 Hz, 1H). 19F NMR
(282 MHz, CDCl3) δ −70.66 (s). 13C NMR (75 MHz, CDCl3) δ
177.43 (s), 137.71 (s), 120.08 (q, J = 293.3 Hz), 79.60 (td, J = 59.8,
29.9 Hz), 78.12-75.17 (m), 65.71 (d, J = 1.6 Hz), 47.74 (s), 45.14 (s),
42.51 (s), 37.87 (s). HRMS (ESI) m/z C15H13F9NO3 [M + H]+:
calculated 426.0752, found 426.0743; C15H12F9NO3Na [M + Na]+:
calculated 448.0572, found 448.0562.
■
Materials. 2,2,2-trifluoroethanol (Fluorochem, 99%), ethylene
carbonate (Alfa Aesar, 99%), diisopropyl azodicarboxylate (DIAD;
Fluorochem, 99%), di-tert-butylazodicarboxylate (DBAD; Fluoro-
chem, 98%), triethylamine (Fluorochem, 99%), N-(3-aminopropyl)-
diethanolamine (Fluorochem, 95%), perfluoro-tert-butanol (Fluoro-
chem, 97%), methanesulfonyl chloride (Sigma-Aldrich, 99%), ethyl
vinyl ether (Sigma-Aldrich, 99%), ethanolamine, potassium osmate-
(VI) dihydrate (Sigma-Aldrich), 4-methylmorpholine N-oxide
(Sigma−Aldrich, 97%), Grubbs’ second generation catalyst (G2;
Sigma−Aldrich), cis-norbornene-exo-2,3-dicarboxylic anhydride (Car-
bosynth), 1-amino-3,6,9-trioxaundecanyl-11-ol (amino-TEG-alcohol;
Combi-Blocks, 95%), MeO-PEG4-NH2 (Combi-Blocks, 97%), and
deuterated solvents (CDCl3, CD2Cl2, D2O, CD3COCD3, CD3OD,
and DMSO-d6; Cambridge Isotope Laboratories) were used as
received without further purification. Grubbs’ third generation catalyst
(G3) was synthesized from G2 by reaction with 5-bromopyridine
(Fluorochem, 97%) at room-temperature (RT) and purified by
filtration with dry pentane. N-(Hydroxyethyl)-cis-norbornene-exo-2,3-
dicarboxiimide (S2)26 and exo-N-methyl-norbornene-2,3-dicarboxii-
mide (M0)27 were prepared, as reported previously. Flash
chromatography was performed on silica gel (SiliCycle, 230−400
mesh, particle size 32−63 μm, 60 Å). Residual ruthenium and
osmium amounts were removed from the polymers by SiliaMetS
DMT (SiliCycle), which is a silica-bound 2,4,6-trimercaptotriazine.
SnakeSkin dialysis tubing, 3.5K MWCO, 22 mm, was purchased from
Thermo Fisher Scientific.
General Synthesis Procedure for Bis(2-hydroxyethyl)-
aminopropylamino- (S3), OH-TEG-amino- (S4), and Methyl-
TEG-amino-cis-norbornene-exo-2,3-dicarboxiimide (M7). cis-
Norbornene-exo-2,3-dicarboxylic anhydride (1.1 equiv) and the
amine (1.0 equiv) were charged into a reaction flask fitted with a
reflux condenser. Next, anhydrous toluene and triethylamine (0.1
equiv) were added and stirred overnight at 110 °C. The reaction
mixture was filtered and the crude product was purified by flash
chromatography using ethyl acetate/methanol as the eluent to obtain
the title compounds as colorless oils (S3: 94%, S4: 82%, M7: 88%).
S3: 1H NMR (300 MHz, CDCl3) δ 6.25 (s, 2H), 3.62−3.45 (m, 6H),
3.23 (s, 4H), 2.66 (d, J = 1.1 Hz, 2H), 2.58 (s, 4H), 2.50 (s, 2H), 1.71
(q, J = 6.9 Hz, 2H), 1.52−1.45 (m, 1H), 1.16 (d, J = 9.8 Hz, 1H). 13C
NMR (75 MHz, CDCl3) δ 178.38 (s), 137.78 (s), 59.65 (s), 56.26
(s), 52.00 (s), 47.81 (s), 45.11 (s), 42.74 (s), 36.51 (s), 25.44 (s).
HRMS (ESI) m/z C16H25N2O4 [M + H]+: calculated 309.1814,
found 309.1810; C16H24N2O4Na [M + Na]+: calculated 331.1634,
1
found 331.1626. S4: H NMR (400 MHz, CDCl3) δ 6.22 (s, 2H),
3.70−3.43 (m, 16H), 3.20 (s, 2H), 2.62 (s, 2H), 1.42 (d, J = 9.8 Hz,
1H), 1.29 (d, J = 9.8 Hz, 1H). 13C NMR (75 MHz, CDCl3) δ 178.04
(s), 137.82 (s), 72.54 (s), 70.83−70.25 (m), 69.87 (s), 66.93 (s),
61.69 (s), 47.81 (s), 45.25 (s), 42.71 (s), 37.76 (s). HRMS (ESI) m/z
C17H25NO6Na [M + Na]+: calculated 362.1580, found 362.1574. M7:
1H NMR (400 MHz, CDCl3) δ 6.27 (t, J = 1.7 Hz, 2H), 6.27 (t, J =
1.7 Hz, 2H), 3.72−3.48 (m, 16H), 3.36 (s, 3H), 3.28−3.17 (m, 2H),
2.66 (d, J = 1.2 Hz, 2H), 1.47 (d, J = 9.9 Hz, 1H), 1.35 (d, J = 9.9 Hz,
1H). 13C NMR (75 MHz, CDCl3) δ 178.11 (s), 137.94 (s), 77.48 (s),
77.33−77.03 (m), 76.84 (s), 72.05 (s), 70.81-70.53 (m), 70.00 (s),
67.01 (s), 59.13 (s), 47.94 (s), 45.39 (s), 42.83 (s), 37.87 (s). HRMS
(ESI) m/z C18H28NO6 [M + H]+: calculated 354.1917, found
354.1914; C18H27NO6Na [M + Na]+: calculated 376.1736, found
376.1729.
2-(2,2,2-Trifluoroethoxy)ethyl-cis-norbornene-exo-2,3-di-
carboxiimide (M1). S1.1 was obtained by a procedure adapted from
ref 28 whereby a round-bottomed flask was charged with 2,2,2-
trifluoroethanol (9.75 g, 111 mmol), NaOH (0.44 g, 11 mmol), and
ethylene carbonate (9.75 g, 111 mmol). The mixture was refluxed at
110 °C overnight and the product (14 g, 87%) was collected by
distillation. Next, PBr3 was slowly added to the colorless liquid at 0 °C
and stirred for 1 h. The ice bath was then removed and the reaction
mixture stirred for 3 h. The resulting crude product was placed on ice
and saturated NaHCO3 was very carefully added. The lower layer was
collected and distilled to yield the colorless product S1.2 (8 g, 95%).
1H NMR (400 MHz, CDCl3) δ 4.00−3.86 (m, 4H), 3.48 (td, J = 6.1,
2.0 Hz, 2H). 19F NMR (377 MHz, CDCl3) δ −74.38 (t, J = 8.6 Hz).
13C NMR (101 MHz, CDCl3) δ 123.91 (q, J = 279.6 Hz), 72.39 (s),
68.57 (q, J = 34.3 Hz), 29.42 (s). S1.3 was synthesized as reported by
Mansfeld and others.29 To prepare M1, S1.2 (2.8 g, 13.5 mmol) was
stirred with K2CO3 (138.2, 7.4 mmol) and acetone (20 mL) for 30
min at 0 °C before S1.3 (2 g, 12.26 mmol) was added and the
Bis(2-(2,2,2-trifluoroethoxy)ethyl)aminopropylamino-cis-
norbornene-exo-2,3-dicarboxiimide (M3). Methanesulfonyl
chloride was added dropwise to a solution of S3 (2.0 g, 6.49
mmol) in dry DCM at 0 °C. After 15 min of stirring, the ice bath was
B
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
removed and the mixture was stirred for further 4 h. Then, saturated
sodium bicarbonate was added and the DCM phase was washed three
times with water. The solvent was removed on a rotavap, and the
mixture was dissolved in dry THF (10 mL) and placed on ice. A
mixture of 2,2,2-trifluoroethanol and potassium tert-butoxide was then
added dropwise, the reaction was brought to 70 °C, and stirred
overnight. Next, the solvent was removed, and the crude product was
redissolved in ethyl acetate, washed three times with water, and
purified by flash chromatography with a hexane/ethyl acetate (1:1, v/
v) mixture to yield a yellowish oil (1.4 g, 70%). 1H NMR (300 MHz,
CDCl3) δ 6.28 (s, 2H), 3.84 (q, J = 8.8 Hz, 4H), 3.67 (t, J = 5.5 Hz,
4H), 3.50 (t, J = 7.4 Hz, 2H), 3.26 (s, 2H), 2.73 (t, J = 5.6 Hz, 4H),
2.66 (s, 2H), 2.56 (t, J = 6.8 Hz, 2H), 1.68 (p, J = 7.1 Hz, 2H), 1.50
(d, J = 9.8 Hz, 1H), 1.20 (d, J = 9.9 Hz, 1H). 19F NMR (282 MHz,
CDCl3) δ −74.26 (s). 13C NMR (75 MHz, CDCl3) δ 178.00 (s),
137.79 (s), 124.02 (q, J = 279.6 Hz), 71.41 (s), 68.47 (q, J = 33.9
Hz), 54.07 (s), 52.94 (s), 47.77 (s), 45.14 (s), 42.67 (s), 36.75 (s),
25.76 (s). HRMS (ESI) m/z C20H27F6N2O4 [M + H]+: calculated
473.1875, found 473.1856; C20H26F6N2O4Na [M + Na]+: calculated
495.1695, found 495.1674.
Bis(2-(perfluoro-tert-butoxy)ethyl)aminopropylamino-cis-
norbornene-exo-2,3-dicarboxiimide (M4). A mixture of S3 (2.0
g, 6.5 mmol), triphenylphosphine (4.9 g, 19 mmol), and perfluoro-
tert-butyl alcohol (2.4 mL, 18 mmol) was dissolved in anhydrous
diethyl ether under an atmosphere of argon. To this solution at 0 °C
was added di-tert-butylazodicarboxylate (DBAD, 4.2 g, 18 mmol) in
one portion. DBAD was used instead of DIAD as it was easier to
remove from the crude product after the Mitsunobu reaction. The
reaction mixture was warmed to RT and stirred for 24 h. After the
completion of the reaction as indicated by TLC, the suspension was
placed in a fridge overnight and the resulting precipitate was filtered
away. A solution of hydrogen chloride (6 mL) was added to the
mixture and was stirred for 1 h. The precipitate was washed with 4 M
HCl and diethyl ether. The product was purified by flash
chromatography with hexane/ethyl acetate (4:1, v/v) to yield a
colorless oil (800 mg, 81%). 1H NMR (400 MHz, CDCl3) δ 6.30 (t, J
= 1.8 Hz, 2H), 4.06 (t, J = 5.6 Hz, 4H), 3.55−3.45 (m, 2H), 3.32−
3.26 (m, 2H), 2.87 (t, J = 5.7 Hz, 4H), 2.68 (d, J = 1.2 Hz, 2H), 2.60
(t, J = 7.0 Hz, 2H), 1.73−1.63 (m, 2H), 1.55−1.51 (m, 2H), 1.21 (d,
J = 9.8 Hz, 2H). 19F NMR (377 MHz, CDCl3) δ −70.60 (s). 13C
NMR (75 MHz, CDCl3) δ (101 MHz, CDCl3) δ 177.94 (s), 137.81
(s), 120.33 (q, J = 293.6 Hz), 79.72 (dd, J = 59.3, 29.6 Hz), 68.71 (s),
53.68 (s), 52.88 (s), 47.79 (s), 45.17 (s), 42.62 (s), 36.36 (s), 26.01
(s). HRMS (ESI) m/z C24H23F18N2O4 [M + H]+: calculated
745.1370, found 745.1351; C20H26F6N2O4Na [M + Na]+: calculated
767.1190, found 767.1174.
Perfluoro-tert-butoxy-TEG-amino-cis-norbornene-exo-2,3-
dicarboxiimide (M6). S4 (803 mg, 2.37 mmol) and triphenylphos-
phine (745 mg, 2.84 mmol) were dissolved in dry THF (10 mL).
Next, DIAD (450 μL, 2.84 mmol) predissolved in 8 mL of THF was
added dropwise over 5 min. Then, perfluoro-tert-butanol in 9 mL of
THF was added in one portion at 0 °C and stirred overnight at RT.
The reaction mixture was placed in the fridge for a few hours and the
residual precipitate was removed. The crude product was purified by
flash chromatography using hexane/ethyl acetate (4:1, v/v) and the
1
product was obtained as a white solid (0.95 g, 72%). H NMR (400
MHz, CDCl3) δ 6.28 (d, J = 1.7 Hz, 1H), 4.15 (s, 1H), 3.77−3.53 (m,
8H), 3.26 (d, J = 1.5 Hz, 1H), 2.67 (s, 1H), 1.47 (d, J = 1.3 Hz, 1H),
1.36 (d, J = 9.8 Hz, 1H). 19F NMR (377 MHz, CDCl3) δ −70.37 (s).
13C NMR (75 MHz, CDCl3) δ 178.45 (s), 138.29 (s), 120.80 (m, J =
293.3 Hz), 71.52 (s), 71.10 (d, J = 3.2 Hz), 70.37 (s), 69.82 (d, J =
15.0 Hz), 67.37 (s), 48.29 (s), 45.74 (s), 43.16 (s), 38.21 (s). HRMS
(ESI) m/z C21H24F9NO6 [M + H]+: calculated 558.1538, found
558.1543; C21H24F9NO6Na [M + Na]+: calculated 580.1358, found
580.1351.
Ring-Opening Metathesis Polymerization (P#). Two Schlenk
flasks loaded with either a monomer M# (100 mg) or G3 catalyst
(amount depended on the desired molecular weight of the resulting
polymer) were evacuated and backfilled with argon three times. The
deoxygenated G3 was then dissolved in dry DCM (1 mL, degassed via
three freeze−pump−thaw cycles) and added in one portion to the
flask containing the monomer dissolved in dry, degassed DCM (2
mL). After 30 min of stirring at RT, the active species were quenched
with ethyl vinyl ether (0.5 mL) and the mixture was stirred for
another 30 min. Polymers were precipitated into methanol or hexane
multiple times. Residual ruthenium was removed by SiliaMetS DMT,
which was in turn discarded by centrifugation and filtration. Nearly
quantitative yields were obtained for the polymerization.
Ring-Opening Metathesis Polymerization of Statistical and
Block Copolymers (P#-stat-P7, P#-b-P7). While the first Schlenk
flask was loaded with the G3 catalyst (amount depended on the
desired molecular weight), the second Schlenk flask contained either
only the monomer M# or a mixture of monomers M# and M7,
depending on whether the desired copolymers were either block or
statistical, respectively. A monomer ratio M#:M7 of 1:1 was used for
all copolymers, except for P4-stat-P7 and P4-b-P7, in which case a
ratio of 1:3 (M4:M7) was implemented. The Schlenk flasks were
evacuated and backfilled with argon three times. The deoxygenated
G3 was then dissolved in dry DCM (1 mL, degassed via three freeze−
pump−thaw cycles) and added in one portion to the flask containing
the monomer dissolved in dry, degassed DCM (2 mL). In the case of
block copolymers, monomer M7 was added after 30 min of stirring at
RT. The mixture was stirred for another 30 min and the active species
were quenched with ethyl vinyl ether (0.5 mL), as described above.
The statistical copolymers, on the other hand, were stirred at RT for
only 45 min before the catalyst was quenched with ethyl vinyl ether.
Polymers were precipitated into methanol or hexane multiple times.
Residual ruthenium was removed by SiliaMetS DMT, which was in
turn discarded by centrifugation and filtration. Nearly quantitative
yields were obtained for the polymerization.
Formation of Quaternary Ammonium Polymers (P3-QA and
P4-QA). Polymers (P3 or P4, 1 equiv by monomer) were dissolved in
anhydrous acetone. After the addition of K2CO3 (1.5 equiv) and
methyl iodide (20 equiv), the reaction mixture was stirred overnight
at 50 °C. The polymers were filtered and dialyzed against acetone.
Dihydroxylation of Polymeric Olefins (P#-OH, P#-stat-P7-
OH, P#-b-P7-OH, and P#-QA-OH). Method A: Polymers (1 equiv
by monomer, 100 mg) were dispersed in acetone (4 mL) and stirred
at RT. 4-Methylmorpholine N-oxide (2.1 equiv) and K2OsO4·2H2O
(0.01 equiv) predissolved in H2O (0.4 mL) were added consecutively
and the mixture was stirred for 18 h. Next, osmium was removed by
stirring with SiliaMetS DMT for 2 h. The mixture was then filtered
directly into a dialysis bag and dialyzed against water for three days.
The remaining silica was removed by centrifugation and filtration.
Method B: CeCl3·7H2O (0.1 equiv) predissolved in H2O (0.1 mL)
was added to a suspension of NaIO4 in H2O (0.8 mL). The reaction
2,2,2-Trifluoroethoxy-TEG-amino-cis-norbornene-exo-2,3-
dicarboxiimide (M5). Methanesulfonyl chloride was added
dropwise to a solution of S4 (212 mg, 0.6 mmol) in dry DCM at 0
°C. After 15 min of stirring, the ice bath was removed and the mixture
was stirred for further 1 h. Then, the solvent was removed on a
rotavap, the product was redispersed in ethyl acetate and filtered, and
the filtrate was collected. The product was redissolved in dry THF (10
mL) and placed on ice. A mixture of 2,2,2-trifluoroethanol and
potassium tert-butoxide was then added dropwise. The reaction was
brought to 70 °C and stirred overnight. Next, the solvent was
removed, and the crude material was redissolved in ethyl acetate,
washed three times with water, and purified by flash chromatography
with a hexane/ethyl acetate (1:1, v/v) mixture to yield a yellowish oil
1
(160 mg, 61%). H NMR (400 MHz, CDCl3) δ 6.28 (t, J = 1.7 Hz,
1H), 3.90 (q, J = 8.8 Hz, 1H), 3.78 (dd, J = 5.6, 3.6 Hz, 1H), 3.74−
3.53 (m, 6H), 3.28−3.24 (m, 1H), 2.67 (d, J = 1.0 Hz, 1H), 1.47 (dd,
J = 6.3, 4.9 Hz, 1H), 1.36 (d, J = 9.9 Hz, 1H). 19F NMR (377 MHz,
CDCl3) δ −74.27 (s). 13C NMR (75 MHz, CDCl3) δ 178.00 (s),
137.84 (s), 127.07−121.53 (m), 71.95 (s), 70.64 (dd, J = 7.5, 6.0 Hz),
69.91 (s), 68.77 (d, J = 34.0 Hz), 66.92 (s), 47.83 (s), 45.29 (s),
42.72 (s), 37.76 (s). HRMS (ESI) m/z C19H27F3NO6 [M + H]+:
calculated 422.1790, found 422.1784; C19H26F3NO6Na [M + Na]+:
calculated 444.1610, found 444.1600.
C
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
a
Scheme 1. Synthesis (A) and Polymerization (B) of Fluorinated Monomers (M1−M6)
a
Reagents and conditions:(a) CF3CH2OH, NaOH, 150 °C, 95%; (b) PBr3, 0 °C to RT; (c) K2CO3, acetone, RT to 60 °C, 78% (over two steps);
(d) (CF3)3COH, DIAD, PPh3, THF, 0−40 °C, 83%; (e3) (i) CH3SO2Cl, N3Et, DCM, 0 °C; (ii) CF3CH2OH, (CH3)3COK, THF, 0−70 °C, 70%;
(e4) (CF3)3COH, DBAD, PPh3, Et2O, 0−40 °C, 81%; (f5) (i) CH3SO2Cl, N3Et, DCM, 0 °C; (ii) CF3CH2OH, (CH3)3COK, THF, 0 −70 °C,
61%; (f6) (CF3)3COH, DIAD, PPh3, THF, 0−40 °C, 72%; (g) (i) DCM, RT; (ii) C2H5OCH5CH2; and (h) NMO, K2OsO4·2H2O, acetone, H2O,
RT.
mixture was stirred at RT and placed on ice as soon as the suspension
turned yellow. RuCl3·H2O (0.01 equiv) predissolved in H2O (0.1
mL) and polymers (1 equiv by monomers) predissolved in
acetonitrile (3 mL) and ethyl acetate (3 mL) were added
consecutively. The reaction was vigorously stirred for further 60
min on ice, at which point a mixture of Na2SO3 and Na2SO4 was
added. After stirring for 30 min, the precipitate was removed by
centrifugation and filtration. The supernatant was dialyzed against
water for three days. Method C: The procedure for syn
dihydroxylation was adapted from ref 30. 5 mL of glacial acetic acid
was added to a mixture of polymers (200 mg, 1 equiv by monomer),
NaIO4 (0.6 equiv) and LiBr (0.4 equiv). The reaction mixture was
then heated to 95 °C and stirred for 12 h. Next, the polymers were
extracted into the ethyl acetate phase and washed three times with
aqueous NaHCO3. The solvent was evaporated on a rotavap, the
polymers dissolved in methanol (10 mL), the pH of the reaction
mixture neutralized, and the polymers stirred with K2CO3 (10 equiv)
at 50 °C. After 12 h of stirring, the resulting polymers were dialyzed
against methanol for one day and against ultrapure water for further 2
days.
NMR spectrometer in phase-sensitive mode. A total of 256 time
increments were collected and linearly predicted to 1024, with 4
transients per increment and a relaxation delay of 1.5 s. Diffusion
ordered spectra (DOSY) were also obtained at 25 °C on a Bruker
Avance 400 NMR spectrometer with a standard Bruker pulse
program, stegp1s. Diffusion time, the number of gradient steps,
relaxation, and recovery delay were set to 1000 ms, 16, 3 s, and 1 μs,
respectively. Weight-average molecular weights were determined by
DOSY as described by Grubbs and colleagues,31 whereby PEG
standards were used to obtain a calibration curve. TopSpin 3.2
software was used to process the HMQC and DOSY spectra.
19F NMR Relaxation Time Measurements. 19F NMR spectra of
solutions were recorded without proton decoupling on a Bruker
Avance 300 NMR spectrometer at a fluorine frequency of 282 MHz.
The samples were dissolved in D2O with an average concentration of
2 mg/mL. The magnetic field was locked to D2O, which was fully
encapsulated within an internal capillary tube and all of the spectra
were acquired at 25 °C. 19F NMR longitudinal (T1) and transverse
(T2) relaxation times were measured using the inversion recovery and
Carr−Purcell−Meiboom−Gill (CPMG) techniques, respectively.
Shimming was performed prior to each experiment to minimize the
B0 field inhomogeneity. 19F NMR spectra of solutions for relaxation
time measurements were recorded using a 90° pulse of 9 μs, an
acquisition time of 0.98 s, and a repetition delay of 5−10 s. The
spectrum width was 50 kHz and 4k data points were collected.
Typically, two acquisitions were accumulated to improve the signal-
1H, 19F, and 13C NMR. 1H, 19F, and 13C NMR spectra were
recorded at 25 °C on a Bruker Avance 300 (1H NMR 300 MHz, 19F
NMR 282 MHz, 13C NMR 75 MHz) or Bruker Avance 400 (1H
NMR 400 MHz, 19F NMR 376 MHz, 13C NMR 100 MHz) NMR
spectrometer. Decoupled heteronuclear multiple-quantum correlation
(HMQC) spectra were obtained at 25 °C on a Bruker Avance 400
D
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
Figure 1. MALDI-ToF mass spectrum of P2-OH. The detected masses in the inset correspond to the increasing amount of dihydroxyl groups
within a polymer chain. If the C,C-bonds had been cleaved, the masses would have been 18 m/z apart, as shown in the Supporting Information
(Figure S165).
to-noise ratio. Bruker’s TopSpin 3.2 software obtains a 1D spectrum
for each τ value stored in a 2D data set, which was baseline corrected
and phased for quantitative measurements. Integration, fitting, and T1
or T2 calculations were performed with Bruker’s Dynamics Center
software. For 19F NMR signal-to-noise ratio measurements, the
solution spectra at 282 or 376 MHz were measured under the
following conditions: 16 scans, relaxation delay 1 s, 90° pulse width 9
μs or 18 μs, and an acquisition time of 0.98 or 0.73 s, respectively.
High-Resolution Accurate Mass Spectrometry (HRMS).
High-resolution and accurate mass spectra (HRMS) were acquired
by electrospray ionization (ESI) on a Thermo Scientific LTQ
Orbitrap XL equipped with a nano-electrospray ion source and a
resolution of 105 at m/z 400.
Matrix-Assisted Laser Desorption Ionization Time-of-Flight
Mass Spectrometry (MALDI-ToF). MALDI-ToF was performed on
a Bruker ultrafleXtreme instrument using 2-[(2E)-3-(4-tert-butyl-
phenyl)-2-methylprop-2-enylidene] malononitrile (DCTB) as a
matrix, polystyrene as a calibrant, and silver trifluoroacetate as an
ionizing salt. Spectra were recorded with an accelerating voltage of 20
kV in reflector positive ion mode. The solutions of matrix/sample/
cationizing agent were combined in a 10:1:1 ratio (v/v/v). The
resulting solution (1 μL) was deposited onto the plate, dried, and
examined. Ions were detected in a m/z range of 2 × 103 to 2 × 104.
Gel Permeation Chromatography (GPC). The average relative
molar mass (Mn, GPC) and molecular weight distribution (Đ = Mw/
Mn) values of polymers were determined by GPC in dimethylforma-
mide (DMF) (2 mg/mL) at RT and a flow rate of 1 mL/min. The
elution curves were calibrated with 10 monodisperse polystyrene
standards (Malvern Polycal PS standards, MW range from 103 to 3 ×
106 g/mol). For measurements in DMF, the GPC instrument was
supplied with a Viscotek GPCmax VE2001 GPC solvent/sample
module, a Viscotek UV detector 2600, a Viscotek VE3580 RI
detector, and two Viscotek T6000 M columns (7.8 Å, 300 mm, 103−
107 Da). All samples were dissolved in HPLC grade solvents, shaken
overnight, and filtered through a PTFE syringe membrane filter (0.45
μm pore size, VWR) prior to GPC measurements. Mn values of
polymers were not corrected for potential variations in the
hydrodynamic diameters in DMF compared to the polystyrene
standards.
Dynamic Light Scattering (DLS). DLS measurements were
performed on a Beckman Coulter DelsaMax Series instrument. The
scattering angle used was 90° and the temperature was fixed at 25 °C.
Prior to measurements, the samples were dried and redispersed in
ultrapure H2O.
RESULTS AND DISCUSSION
■
Fluorinated monomers based on exo-norbornene-imides that
contain either 3 (M1, M5), 6 (M3), 9 (M2, M6), or 18 (M4)
identical 19F nuclei were designed in the form of one or more
terminal trifluromethyl groups (Scheme 1). In M1 and M2, the
fluorinated moieties were connected through a shorter alkyl
linker to the imides, whereas in M5 and M6 a longer TEG-
based linker was chosen. The pendants of terminal
trifluromethyl groups in M3 and M4 are somewhat
intermediary in length and incorporate a tertiary amine
functionality. The corresponding homopolymers (P1-P6)
were obtained by ROMP with Grubbs’ 3rd generation catalyst
(G3). Such ROMP polymers are intrinsically quite rigid due to
their partially unsaturated backbone. Addition of a fluorinated
moiety to the monomeric units further decreases their water
solubility. In fact, even when olefin bonds are hydrogenated or
when fluorinated monomers are copolymerized with TEG-
substituted monomers, they remain water-insoluble.
Initially, LiBr-catalyzed NaIO4-facilitated dihydroxylation of
polymers (P1−P4) under acidic conditions30 was attempted.
All of the resulting polymers except P2-OH-C were soluble in
water and almost complete dihydroxylation could be shown by
the missing olefin cross-peaks in the HMQC NMR spectra
(Figures S177 and S178). Since molecular weight determi-
nation by GPC and DOSY was unfeasible due to insolubility in
organic solvents and aggregation in water, respectively, the
question of overoxidation-related C,C-bond cleavage remained
unanswered.
E
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
f
Table 1. Physical and NMR Properties of Bishydroxylated Polymers
a
c
c
d
polymer
Mw,NMR [kDa] Mn,GPC [kDa]
Đ
Rh [nm]
19F T1 [ms]
19F T2 [ms]
19F T1/T2 fluorine content [%] 19F NMR SNR
b
e
P1-OH
P3-OH
P3-QA-OH
P5-OH
P6-OH
7.1
11.7
5.2
6.9
11.2
11.2
6.3
1.18
1.10
1.10
1.14
1.16
81.8
19.8
1.9
1.5
10.3
1070 237
1245 164
948 42
1716 96
1626 142
239 63
401 127
381 67
1071 49
4.5
3.1
2.5
1.6
12.2
17
22
17
12
28
12
15
39
31
11
b
e
3.2
5.2
b
e
5.6
133 45
a
b
Molecular weights of bishydroxylated polymers estimated by 1H DOSY NMR in D2O or by GPC in DMF. Mw,NMR was calculated from the PEG
c
standard calibration curve using the experimental value of the diffusion coefficient. Molecular weights and poly dispersity indices (Đ) of their
d
parent non-dihydroxylated polymers. Hydrodynamic radii of bishydroxylated polymers calculated by the Stokes−Einstein equation from the
e
diffusion coefficients of the polymers determined by 1H DOSY NMR in D2O or measured by DLS in H2O. 1H DOSY and 19F NMR SNRs were
f
measured at 400 MHz and 376 MHz, respectively, at a polymer concentration of 0.5 mg/mL. For 19F NMR relaxation time measurements (282
MHz, 298 K), the polymers were dissolved in water at a concentration of 2 mg/mL.
Scheme 2. Chemical Structures of Fluorinated Bishydroxylated Homopolymers (P#-OH), Block Copolymers (P#-b-P7-OH),
and Statistical Copolymers (P#-stat-P7-OH)
Thus, we embarked on an investigation of RuO4-catalyzed
Lewis acid-facilitated dihydroxylation32 with metathesis-active
ruthenium complexes such as G3. This would have been an
excellent method to perform ROMP followed by dihydrox-
ylation all in one-pot. Unfortunately, the dihydroxylation was
only partial and accompanied by cleavage of the olefin bonds
as shown by MALDI-ToF (Figure S159−S160) and NMR
(Figures S161−S162). The dihydroxylation efficiency im-
proved when G3 was replaced with Grubbs’ 2nd generation
catalyst (Figures S163−S164) and became complete when
The latter problem was avoided with the classical Upjohn
dihydroxylation, whereby N-methylmorpholine N-oxide was
used to prepare OsO4 in situ, while residual ruthenium and
osmium were removed by silica-bound 2,4,6-trimercaptotria-
zine. The resulting polymers (P1-OH−P6-OH, Scheme 2)
were slightly soluble in DMF and their molecular weights
could be estimated by GPC (Figures S179−S183). The
MALDI-ToF mass spectrum of P2-OH also indicated that
after 18 h the dihydroxylation was almost complete and that
the C,C-bond cleavage did not occur (Figure 1). Both P2-OH
and P4-OH were insoluble in water due to the perfluoro-t-
butyl ether and high fluorine contents of 37 and 43%,
respectively.
33
RuCl3 was implemented as a source of RuO4 (Figures
S165−S167). Despite efforts to accelerate hydrolysis of
34
Similarly, very weak 19F NMR signals of P1-OH, P3-OH,
and P6-OH (with fluorine contents of 17, 22, and 28%,
ruthenate esters with CeCl3, C,C-bond cleavage could not
be prevented.
F
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
d
Table 2. Physical and NMR Properties of Bishydroxylated Statistical and Block Copolymers
a
b
b
c
polymer
Mw,NMR [kDa] Mn,GPC [kDa]
Đ
Rh [nm] 19F T1 [ms] 19F T2 [ms] 19F T1/T2 fluorine content [%] 19F NMR SNR
P1-stat-P7-OH
P1-b-P7-OH
P2-stat-P7-OH
P2-b-P7-OH
P3-stat-P7-OH
P3-b-P7-OH
P4-stat-P7-OH
P4-b-P7-OH
P5-stat-P7-OH
P5-b-P7-OH
P6-stat-P7-OH
P6-b-P7-OH
5.0
4.7
9.6
16.4
5.6
4.3
12.6
15.5
4.3
5.3
5.5
6.5
6.2
5.4
5.1
6.2
5.5
5.3
6.5
5.0
5.0
5.2
5.6
1.27
1.21
1.21
1.17
1.27
1.44
1.19
1.19
1.12
1.14
1.07
1.16
1.9
3.1
3.2
1.9
2.0
1.7
4.5
5.6
1.7
2.0
2.0
658 21
696 19
478 26
467 20
855 73
361 66
471 37
451 50
1685 51
1420 66
561 45
263 24
289 17
66 15
64 21
352 25
51 19
124 29
82 27
1215 86
979 72
2.4
2.5
7.2
7.3
2.4
7.1
3.8
5.5
1.4
1.5
10.3
6.2
8
8
30
30
21
21
20
14
28
15
32
29
31
34
20
20
13
13
17
17
7
7
17
17
54
88
9
8
20.0
10.6
1
547
9
a
Molecular weights of bishydroxylated polymers estimated by H DOSY NMR in D2O were calculated from the PEG standard calibration curve
b
using the experimental value of the diffusion coefficient. Molecular weights and poly dispersity indices (Đ) of their parent non-dihydroxylated
c
polymers. Hydrodynamic radii of bishydroxylated polymers calculated by the Stokes−Einstein equation from the diffusion coefficients of the
polymers determined by 1H DOSY NMR in D2O. 1H DOSY and 19F NMR SNRs were measured at 400 and 376 MHz, respectively, at a polymer
d
concentration of 0.5 mg/mL. For 19F NMR relaxation time measurements (282 MHz, 298 K), the polymers were dissolved in water at a
concentration of 2 mg/mL.
Figure 2. Comparison of (a) 19F NMR T1 and (b) T2 relaxation times, (c) 19F NMR T1/T2 ratios and (d) 19F NMR SNR values of dihydroxylated
statistical (P(1-4)-stat-P7-OH) and block (P(1-4)-b-P7-OH) copolymers of various molecular weights in water. The 19F NMR signal of P(1−4)-
b-P7-OH block copolymers with a molecular weight larger than 35 kDa was not detected and thus the 19F NMR relaxation properties were not
measured.
respectively) in water could only be obtained at low molecular
weights (i.e., at 7, 12, and 5 kDa, respectively). Further
increase of the molecular weights completely attenuated the
substantially when tertiary amines were transformed into
quaternary ammonium salts (P3-QA-OH, Table 1) by
methylation with methyl iodide and slightly increased when
the molecular weight of the parent olefin polymer was
increased from 11 to 32 kDa. In both cases, the 19F NMR
19F NMR signal due to aggregation of fluorinated moieties. 19
F
NMR signal-to-noise ratios (SNRs) of P3-OH improved
G
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
Figure 3. Comparison of water-soluble polymers and copolymers of different molecular weights. (a) 19F NMR T1 and (b) T2 relaxation times, (c)
19F NMR T1/T2 ratios and (d) 19F NMR SNR values of the dihydroxylated homofluoropolymer P5-OH, statistical (P(5-6)-stat-P7-OH) and block
(P(5-6)-b-P7-OH) copolymers of various molecular weights in water.
signal was linearly dependent on the polymer concentration
(Figure S233). Quaternization of P4-OH, on the other hand,
did not contribute to its water solubility and the 19F NMR
signal could not be detected in water.
per monomer), the ratio of TEG-ylated monomers had to be
increased to three so as not to breach the 21% fluorine content
limit. All of the resulting statistical copolymers (P#-stat-P7-
OH, Figure 2) indeed exhibited higher 19F SNRs and improved
19F NMR T1/T2 ratios compared to their homofluoropoly-
meric counterparts (Tables 1 and 2). In addition, all of the
obtained 19F SNRs were linearly dependent on the polymer
concentration in water. This was also true for copolymers with
higher molecular weights (40−60 kDa). While the 19F SNRs of
P2-stat-P7-OH, P3-stat-P7-OH, and P4-stat-P7-OH de-
creased with the molecular weight, the 19F SNRs of P1-stat-
P7-OH and P6-stat-P7-OH remained constant and the 19F
SNR of P5-stat-P7-OH increased (cf. Figures S230−S240 and
Table S6).
A comparison between trifluoromethyl- (i.e., P1-stat-P7-
OH) and perfluoro-t-butyl- (i.e., P2-stat-P7-OH) function-
alized copolymers again demonstrates that it is possible to
achieve 3-fold lower 19F NMR T1/T2 and 43% higher signal-to-
noise ratios despite a 60% lower fluorine content (Table 1 and
Figure 2). However, highly fluorinated units may still prove
beneficial if larger amounts of non-fluorinated components are
required especially at lower molecular weights. For example,
similar 19F NMR T1/T2 and signal-to-noise ratios were
obtained (Table 1) when the TEG (M7) to bisperfluoro-t-
butylated monomer (M4, 18 equivalent fluorines) ratio was
increased 3-fold in P4-stat-P7-OH compared to P3-stat-P7-
Compared to P1-OH, P5-OH yielded an excellent 19F NMR
signal in water and the 19F NMR T1/T2 ratio improved
substantially from 4.5 to 1.6 (Table 1), which is excellent for
homofluoropolymers. While the low 19F NMR T2 value of P1-
OH was a consequence of shorter linkages between
trifluoromethyl groups and the polymeric backbone, 4.5
times longer 19F T2 of P5-OH was a result of longer
tetraethylene glycol linkers separating the fluorine moieties
from the polymer main chain and thus faster internal motions.
The glycol side chains also allowed for greater exposure to the
solvent. Despite 19F T1/T2 and SNR values increasing from 1.6
to 2.2 (Table S5) and decreasing by almost 2-fold (Table S6),
respectively, when the molecular weight of their parent olefin
polymers was increased from 13 to 56 kDa, the 19F NMR
SNRs of the highest molecular weight P5-OH were still
linearly dependent on the polymer concentration (Figure
S235).
The question arose whether 19F NMR relaxation times and
SNRs of bishydroxylated polymers (P#-OH, Figure 2) would
improve at higher molecular weights with random incorpo-
ration of hydrophilic TEG-ylated monomers in a 1:1 ratio. In
the case of copolymers based on M4 (18 equivalent fluorines
H
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
OH, which contained TEG and bis-trifluoromethyl units (M3,
6 equivalent fluorines) in a one-to-one ratio.
group with a longer chain length despite an increase in 19F
NMR T1/T2 ratios.36 Out of all examined copolymers, P5-stat-
P7-OH was also the one with the smallest decrease in 19F
NMR T1 (15%) and T2 (30%) values when the molecular
weight was increased from 5 to 50 kDa.
While the 19F NMR signal was still detectable in statistical
fluorinated copolymers (P(1−4)-stat-P7-OH) with molecular
weights between 40 and 60 kDa, the 19F resonance peak
completely disappeared in block copolymers (P(1−4)-b-P7-
OH, Scheme 2) of comparable size (Figure 2). At higher
molecular weights, the fluorinated block probably self-
assembles into a fluorous phase similar to the homofluoropol-
ymers, whereas the phase segregation of the fluorinated
moieties in statistical copolymers is prevented by randomly
distributed hydrophilic TEG chains. This was also shown in
the case of a fluorinated TEG-based amphiphile by Yu and
others.35 At lower molecular weights, the differences in 19F
NMR relaxation properties between statistical and block
copolymers were not found to be profound. This is true
especially in M1-, M2-, and M5-based statistical and block
copolymers with a molecular weight of 5−6 kDa, where 19F
SNR and T1/T2 ratios were within error bars.
The 19F NMR T1 and T2 values of P6-stat-P7-OH and P6-b-
P7-OH by contrast both increased by 11% and decreased by
about 47%, respectively, with the 10-fold rise of molecular
weight (Table S4) due to very fast internal rotations of 19F
spins. In large macromolecules, these are known to result in
longer longitudinal relaxation times with increasing molecular
weights, whereas moderate rotations observed in M1-5-based
copolymers improved the efficiency of spin−lattice relaxation
and therefore shortened the T1 values.37 Although P6-b-P7-
OH block copolymers showed lower 19F NMR T1/T2 ratios
than their statistical counterparts (Table S5 and Figure 3), P6-
stat-P7-OH also exhibited better 19F NMR SNRs with
increasing molecular weights. Nevertheless, regarding the 19F
NMR relaxation properties especially at higher molecular
weights, the examined fluorinated statistical copolymers
performed better than the block copolymers, which were in
turn superior to the homopolymers.
Since P5-OH was the only homofluoropolymer with the 19F
NMR signal still detectable at high molecular weights, it is of
no surprise that the corresponding block copolymers
performed better at higher molecular weights than the P(1−
4)-based block copolymers. The 19F SNRs of P5-b-P7-OH
remained constant when the molecular weight of the parent
olefin polymers was increased from 5 to 10 kDa. A decrease in
19F NMR SNR similar to P5-OH was only observed when the
molecular weight was increased to 50 kDa (Figure 3). This is
again most likely a consequence of strong dipolar couplings of
the near-neighboring 19F nuclei.
CONCLUSIONS
■
In summary, ring-opening metathesis polymerization quickly
and efficiently provided narrowly dispersed fluoropolymers
bearing monomeric units of 3, 6, 9, or 18 magnetically
equivalent fluorine atoms. We found that dihydroxylation of
the partially unsaturated polymeric backbone substantially
improves their water solubility and 19F NMR signal-to-noise
ratio. This applies when the fluorine content does not exceed
the 21 wt % limit and when the molecular weight is low. In the
investigated homofluoropolymers, trifluoromethyl units thus
exhibit better 19F NMR signal-to-noise ratios than the
perfluoro-t-butyl ones. The 19F magnetic resonance properties
further improve when longer water-solubilizing linkers between
the fluorinated moieties and polymer main chain are
incorporated as opposed to smaller, aliphatic connectors. A
similar effect is achieved either by ammonium quaternization
of linkers bearing tertiary amines or by copolymerization with
tetraethylene glycol-substituted monomers, especially when the
fluorine content is high. All of these polymers exhibited 19F
NMR signal-to-noise ratios linearly dependent on the polymer
concentration. Furthermore, statistical copolymers composed
of methyl- and trifluoromethyl-functionalized TEG-based
monomers yielded a linear dependence of the 19F NMR
intensity on the fluorine content over a molecular weight range
of 5−50 kDa and could thus find application as quantitative
polymeric 19F MRI contrast agents.
The 19F SNRs of P5-stat-P7-OH, on the other hand,
increased slightly with molecular weight (cf. Figures S236−
S238) despite the subtle growth of 19F NMR T1/T2 ratios from
1.4 to 1.7 (Table S5 and Figure 3). The rise of SNR was in a
linear relationship with both the increasing fluorine content
and the molecular weight (Figure 4). Such a linear dependence
of the 19F NMR intensity on the fluorine content in a wide
range of molecular masses may be used for quantitative MRI.
Whittaker and colleagues even observed rising 19F NMR SNRs
of discrete oligo(acrylic acid)s containing one terminal CF3
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.macromol.0c01585.
Full experimental procedures; NMR and HRMS spectra
of monomers; NMR spectra of polymers; MALDI-ToF/
MS spectra of polymers; dihydroxylation experiments;
GPC elugrams of polymers; and NMR relaxation times
of polymers (PDF)
Figure 4. Linear relationship of 19F NMR signal-to-noise ratios (282
and 376 MHz, 25 °C, ns = 16, lb = 0.3 Hz) of P5-stat-P7-OH
copolymers with molecular weights of (a) 5 kDa, (b) 10 kDa, and (c)
50 kDa in D2O.
I
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
Amphiphile as 19F MRI/Fluorescence Dual-Imaging Agent by
Tuning the Self-Assembly. J. Org. Chem. 2015, 80, 6360−6366.
(14) Peng, H.; Blakey, I.; Dargaville, B.; Rasoul, F.; Rose, S.;
Whittaker, A. K. Synthesis and Evaluation of Partly Fluorinated Block
Copolymers as MRI Imaging Agents. Biomacromolecules 2009, 10,
374−381.
AUTHOR INFORMATION
■
Corresponding Author
Andreas F. M. Kilbinger − Department of Chemistry,
University of Fribourg, CH-1700 Fribourg, Switzerland;
orcid.org/0000-0002-2929-7499;
(15) Huang, X.; Huang, G.; Zhang, S.; Sagiyama, K.; Togao, O.; Ma,
X.; Wang, Y.; Li, Y.; Soesbe, T.; Sumer, B.; Takahashi, M.; Sherry, D.
A.; Jinming, G. Multi-Chromatic pH-Activatable 19F-MRI Nanop-
robes with Binary ON/OFF pH Transitions and Chemical-Shift
Barcodes. Angew. Chem., Int. Ed. 2013, 52, 8074−8078.
(16) Rolfe, B.; Blakey, I.; Squires, O.; Peng, H.; Boase, N.;
Alexander, C.; Parsons, P.; Boyle, G.; Whittaker, A.; Thurecht, K.
Multimodal Polymer Nanoparticles with Combined 19F Magnetic
Resonance and Optical Detection for Tunable, Targeted, Multimodal
Imaging in Vivo. J. Am. Chem. Soc. 2014, 136, 2413−2419.
(17) Thurecht, K.; Blakey, I.; Peng, H.; Squires, O.; Hsu, S.;
Alexander, C.; Whittaker, A. Functional Hyperbranched Polymers:
Toward Targeted in Vivo 19F Magnetic Resonance Imaging Using
Designed Macromolecules. J. Am. Chem. Soc. 2010, 132, 5335−5337.
(18) Fu, C.; et al. Low-Fouling Fluoropolymers for Bioconjugation
and In Vivo Tracking. Angew. Chem. 2020, 132, 4759−4765.
(19) Hilf, S.; Kilbinger, A. Functional End Groups for Polymers
Prepared Using Ring-Opening Metathesis Polymerization. Nat. Chem.
2009, 1, 537.
Email: andreas.kilbinger@unifr.ch
Author
Iris K. Tennie − Department of Chemistry, University of
Fribourg, CH-1700 Fribourg, Switzerland
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.macromol.0c01585
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the NCCR Bio-Inspired Materials for
funding/financial support.
REFERENCES
■
(20) Matson, J.; Grubbs, R. Synthesis of Fluorine-18 Functionalized
Nanoparticles for Use as in Vivo Molecular Imaging Agents. J. Am.
Chem. Soc. 2008, 130, 6731−6733.
(1) Lauffer, R. B. Paramagnetic Metal Complexes as Water Proton
Relaxation Agents for NMR Imaging: Theory and Design. Chem. Rev.
1987, 87, 901−927.
(2) Fu, C.; Zhang, C.; Peng, H.; Han, F.; Baker, C.; Wu, Y.; Ta, H.;
Whittaker, A. K. Enhanced Performance of Polymeric 19F MRI
Contrast Agents through Incorporation of Highly Water-Soluble
Monomer MSEA. Macromolecules 2018, 51, 5875−5882.
(3) Knight, J. C.; Edwards, P. G.; Paisey, S. J. Fluorinated Contrast
Agents for Magnetic Resonance Imaging; a Review of Recent
Developments. RSC Adv. 2011, 1, 1415.
(4) Wang, K.; Peng, H.; Thurecht, K. J.; Puttick, S.; Whittaker, A. K.
pH-responsive star polymer nanoparticles: potential 19 F MRI
contrast agents for tumour-selective imaging. Polym. Chem. 2013, 4,
4480−4489.
(5) Jirak, D.; Galisova, A.; Kolouchova, K.; Babuka, D.; Hruby, M.
Fluorine Polymer Probes for Magnetic Resonance Imaging: Quo
Vadis? Magn. Reson. Mater. Phys., Biol. Med. 2019, 32, 173−185.
(6) Diou, O.; Tsapis, N.; Giraudeau, C.; Valette, J.; Gueutin, C.;
Bourasset, F.; Zanna, S.; Vauthier, C.; Fattal, E. Long-Circulating
Perfluorooctyl Bromide Nanocapsules for Tumor Imaging by 19F
MRI. Biomaterials 2012, 33, 5593−5602.
(7) Janjic, J.; Srinivas, M.; Kadayakkara, D.; Ahrens, E. Self-
Delivering Nanoemulsions for Dual Fluorine-19 MRI and Fluo-
rescence Detection. J. Am. Chem. Soc. 2008, 130, 2832−2841.
(8) Dardzinski, B.; Sotak, C. Rapid Tissue Oxygen Tension Mapping
using 19F Inversion-Recovery Echo-Planar Imaging of Perfluoro-15-
crown-5-ether. Magn. Reson. Med. 1994, 32, 88−97.
(9) Tirotta, I.; Mastropietro, A.; Cordiglieri, C.; Gazzera, L.; Baggi,
F.; Baselli, G.; Bruzzone, M.; Zucca, I.; Cavallo, G.; Terraneo, G.;
Baldelli Bombelli, F. A Superfluorinated Molecular Probe for Highly
Sensitive in Vivo 19F-MRI. J. Am. Chem. Soc. 2014, 136, 8524−8527.
(10) Neubauer, A. M.; Myerson, J.; Caruthers, S. D.; Hockett, F. D.;
Winter, P. M.; Chen, J.; Gaffney, P. J.; Robertson, J. D.; Lanza, G. M.;
Wickline, S. A. Gadolinium-Modulated 19F Signals from Perfluor-
ocarbon Nanoparticles as a New Strategy for Molecular Imaging.
Magn. Reson. Med. 2008, 60, 1066−1072.
(11) Jiang, Z.; Liu, X.; Jeong, E.; Yu, Y. Symmetry-Guided Design
and Fluorous Synthesis of a Stable and Rapidly Excreted Imaging
Tracer for 19F MRI. Angew. Chem., Int. Ed. 2009, 121, 4849−4852.
(12) Tanifum, E.; Patel, C.; Liaw, M.; Pautler, R.; Annapragada, A.
Hydrophilic Fluorinated Molecules for Spectral 19F MRI. Sci. Rep.
2018, 8, No. 2889.
(21) Randolph, L.; LeGuyader, C.; Hahn, M.; Andolina, C.;
Patterson, J.; Mattrey, R.; Millstone, J.; Botta, M.; Scadeng, M.;
Gianneschi, N. Polymeric Gd-DOTA Amphiphiles form Spherical and
Fibril-Shaped Nanoparticle MRI Contrast Agents. Chem. Sci. 2016, 7,
4230−4236.
(22) Sowers, M.; McCombs, J.; Wang, Y.; Paletta, J.; Morton, S.;
Dreaden, E.; Boska, M.; Ottaviani, M.; Hammond, P.; Rajca, A.;
Johnson, J. Redox-Responsive Branched-Bottlebrush Polymers for in
Vivo MRI and Fluorescence Imaging. Nat. Commun. 2014, 5,
No. 5460.
(23) Meier, S.; Reisinger, H.; Haag, R.; Mecking, S.; Mulhaupt, R.;
̈
Stelzer, F. Carbohydrate Analogue Polymers by Ring Opening
Metathesis Polymerisation (ROMP) and Subsequent Catalytic
Dihydroxylation. Chem. Commun. 2001, 9, 855−856.
(24) Carrillo, A.; Gujraty, K. V.; Rai, P. R.; Kane, R. S. Design of
Water-Soluble, Thiol-Reactive Polymers of Controlled Molecular
Weight: a Novel Multivalent Scaffold. Nanotechnology 2005, 16,
S416−S421.
(25) Miki, K.; Kimura, A.; Oride, K.; Kuramochi, Y.; Matsuoka, H.;
Harada, H.; Hiraoka, M.; Ohe, K. High-Contrast Fluorescence
Imaging of Tumors In Vivo Using Nanoparticles of Amphiphilic
Brush-Like Copolymers Produced by ROMP. Angew. Chem., Int. Ed.
2011, 50, 6567−6570.
(26) Qiao, Y.; Ping, J.; Tian, H.; Zhang, Q.; Zhou, S.; Shen, Z.;
Zheng, S.; Fan, X. Synthesis and Phase Behavior of a Polynorbornene-
Based Molecular Brush with Dual ”Jacketing” Effects. J. Polym. Sci.,
Part A: Polym. Chem. 2015, 53, 2116−2123.
(27) Walton, H. M. Potential Antimicrobial Agents. III. 4-
Methylamino-2, 4-alkadienoic Acid γ-Lactams1. J. Org. Chem. 1956,
22, 315−318.
(28) Mei, X.; Yue, Z.; Tufts, J.; Dunya, H.; Mandal, B. Synthesis of
New Fluorine-Containing Room Temperature Ionic Liquids and their
Physical and Electrochemical Properties. J. Fluorine Chem. 2018, 212,
26−37.
(29) Mansfeld, F.; Feng, G.; Otto, S. Photo-Induced Molecular-
Recognition-Mediated Adhesion of Giant Vesicles. Org. Biomol. Chem.
2009, 7, 4289−4295.
(30) Emmanuvel, L.; Shaikh, T.; Sudalai, A. NaIO4/LiBr-Mediated
Diastereoselective Dihydroxylation of Olefins: A Catalytic Approach
to the Prevost-Woodward reaction. Org. Lett. 2005, 7, 5071−5074.
(13) Bo, S.; Song, C.; Li, Y.; Yu, W.; Chen, S.; Zhou, X.; Yang, Z.;
Zheng, X.; Jiang, Z.-X. Design and Synthesis of Fluorinated
J
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
pubs.acs.org/Macromolecules
Article
(31) Li, W.; Chung, H.; Daeffler, C.; Johnson, J. A.; Grubbs, R. H.
Application of 1H DOSY for facile measurement of polymer
molecular weights. Macromolecules 2012, 45, 9595−9603.
(32) Scholte, A.; An, M.; Snapper, M. Ruthenium-Catalyzed
Tandem Olefin Metathesis-Oxidations. Org. Lett. 2006, 8, 4759−
4762.
(33) Plietker, B.; Niggemann, M. An Improved Protocol for the
RuO4-Catalyzed Dihydroxylation of Olefins. Org. Lett. 2003, 5,
3353−3356.
(34) Plietker, B.; Niggemann, M. RuCl3/CeCl3/NaIO4: A New
Bimetallic Oxidation System for the Mild and Efficient Dihydrox-
ylation of Uncreactive Olefins. J. Org. Chem. 2005, 70, 2402−2405.
(35) Taraban, M.; Yu, L.; Feng, Y.; Jouravleva, E.; Anisimov, M.;
Jiang, Z.; Yu, Y. Conformational Transition of a Non-Associative
Fluorinated Amphiphile in Aqueous Solution. RSC Adv. 2014, 4,
54565−54575.
(36) Zhang, C.; Kim, D. S.; Lawrence, J.; Hawker, C. J.; Whittaker,
A. K. Elucidating the Impact of Molecular Structure on the 19F NMR
Dynamics and MRI Performance of Fluorinated Oligomers. ACS
Macro Lett. 2018, 7, 921−926.
(37) Hull, W. E.; Sykes, B. D. Fluorotyrosine Alkaline Phosphatase:
Internal Mobility of Individual Tyrosines and the Role of Chemical
Shift Anisotropy as a 19F Nuclear Spin Relaxation Mechanism in
Proteins. J. Mol. Biol. 1975, 98, 121−153.
K
https://dx.doi.org/10.1021/acs.macromol.0c01585
Macromolecules XXXX, XXX, XXX−XXX