Thermoresponsive poly(N -C3 glycine)s

Article abstract of DOI:10.1021/ma302412v

Ring-opening polymerization of N-substituted glycine N-carboxyanhydrides (NCAs) was applied to prepare a series of well-defined poly(N-C3 glycine)s (C3 = n-propyl, allyl, propargyl, and isopropyl), polypeptoids, with molecular weights in the range of 1.8-6.6 kg mol^-1. Poly(N-isopropyl glycine), a previously unreported polypeptoid, could be obtained by bulk polymerization of the corresponding NCA in the melt. The samples were characterized by spectroscopy (NMR and FT-IR), size exclusion chromatography (SEC), and matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI-ToF MS). The polymers could be dispersed in water up to 20-40 g L ^-1; the poly(N-propargyl glycine) was not soluble in water. Turbidity measurements of the three water-soluble polypeptoids illustrated cloud point temperatures dependent on structural and electronic properties of the side chain. The cloud point temperatures were found to increase in the order C3 = n-propyl (15-25 C) < allyl (27-54 C) < isopropyl (47-58 C). Long-term annealing of the aqueous solution of poly(N-{n-propyl} glycine) and poly(N-allyl glycine) above the cloud point temperature resulted in the formation of crystalline microparticles with melting points of 188-198 and 157-165 C (differential scanning calorimetry, DSC), respectively, and rose bud type morphology (scanning electron microscopy, SEM).

Full text of DOI:10.1021/ma302412v

Article
pubs.acs.org/Macromolecules
Thermoresponsive Poly(N‑C3 glycine)s
Joshua W. Robinson,†,‡ Christian Secker,†,‡ Steffen Weidner,§ and Helmut Schlaad†,*
†Department of Colloid Chemistry, Max Planck Institute of Colloids and Interfaces, Research Campus Golm, 14424 Potsdam,
Germany
§1.3 Structure Analysis, Federal Institute for Material Research and Testing (BAM), Richard-Willstatter-Strasse 11, 12489 Berlin,
̈
Germany
ABSTRACT: Ring-opening polymerization of N-substituted glycine N-
carboxyanhydrides (NCAs) was applied to prepare a series of well-
defined poly(N-C3 glycine)s (C3 = n-propyl, allyl, propargyl, and
isopropyl), polypeptoids, with molecular weights in the range of 1.8−6.6
kg mol−1. Poly(N-isopropyl glycine), a previously unreported poly-
peptoid, could be obtained by bulk polymerization of the corresponding
NCA in the melt. The samples were characterized by spectroscopy
(NMR and FT-IR), size exclusion chromatography (SEC), and matrix-
assisted laser desorption/ionization time-of-flight mass spectroscopy
(MALDI−ToF MS). The polymers could be dispersed in water up to
20−40 g L−1; the poly(N-propargyl glycine) was not soluble in water. Turbidity measurements of the three water-soluble
polypeptoids illustrated cloud point temperatures dependent on structural and electronic properties of the side chain. The cloud
point temperatures were found to increase in the order C3 = n-propyl (15−25 °C) < allyl (27−54 °C) < isopropyl (47−58 °C).
Long-term annealing of the aqueous solution of poly(N-{n-propyl} glycine) and poly(N-allyl glycine) above the cloud point
temperature resulted in the formation of crystalline microparticles with melting points of 188−198 and 157−165 °C (differential
scanning calorimetry, DSC), respectively, and rose bud type morphology (scanning electron microscopy, SEM).
Chart 1. Selected [C6H11NO] Structural Isomers: Poly(N-
isopropylacrylamide), Poly(2-isopropyl-2-oxazoline),
Polyleucine, and Poly(N-isobutyl glycine) (Left to Right)
INTRODUCTION
■
“Smart” polymeric materials that react to external stimuli (i.e.,
pH, temperature, light, magnetic and electric fields, etc.) have
received much attention over the years because of their
commercial utility (self-healing, shape memory, piezoelectric,
etc.) and/or biomedical (drug delivery, biosensing, etc.)
applications.1 In addition, research with stimuli-responsive
materials provides insight and understanding into how the
macromolecular systems’ morphology, conformation, and/or
mechanic properties are influenced by noncovalent interactions
(e.g., van der Waals, hydrogen bonding, and columbic/
electrostatic forces). Thermo-responsive polymers based on
polyethers, -esters, and -amides have been extensively
investigated toward the aforementioned biomedical applica-
tions.2−4 In particular the polyacrylamides and poly(2-
oxazoline)s have demonstrated phenomenal tunable and
precision controlled thermally induced biphasic separations,
under aqueous conditions, as homo/copolymers, colloidal
constructs, and networks (i.e., hydrogels).5−10 In particular,
poly(N-isopropylacrylamide) (PNIPAM) and poly(2-isopropyl-
2-oxazoline) (PIPOX),11,12 structural or constitutional isomers
of each other (see Chart 1), have lower critical solution
temperatures (LCST; 32 and 36 °C, respectively) near the
physiological temperature range. For the most part, these
polyamides readily undergo dissolution upon cooling demon-
strating reversibility. However, PIPOX undergoes an irrever-
sible biphasic separation upon long-term annealing above the
cloud point temperature via a crystallization process occurring
within the 2-phase region.13−15 Such crystallization phenom-
enon in solution has been observed for poly(2-oxazoline)s, yet
for no other polymer.16
The thermo-responsive polymers PNIPAM and PIPOX
(Chart 1) are constitutional isomers with either 2° or 3°
amide side chains. Other isomers are the peptide-based main-
chain polyamides polyleucine (PL, 2° amide) and poly(N-
isobutyl glycine) (PNIBG, 3° amide), neither of which
demonstrated solubility in water or LCST behavior.17,18
Although the insolubility of PL can be accredited to hydrogen
bonding,19 the solubility of PNIBG on a purely hydrophobic/
hydrophilic ratio argument should have been consistent with
that of PIPOX. However, the placement of the 3° amide
(hydrogen acceptor) may play a significant role in solubility on
an availability argument (hydrophobic pocket). Luxenhofer et
al.18 described some solubility investigations that confirmed the
water solubility of poly(N-alkyl glycine)s, also referred to as
Received: November 22, 2012
Revised: January 16, 2013
Published: January 31, 2013
© 2013 American Chemical Society
580
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
Scheme 1. Three-Step Synthesis (i−iii) and Ring-Opening Polymerization (iv, v: Quenching with Ac2O) of N-C3 Glycine NCA,
C3 = n-Propyl, Allyl, Propargyl, and Isopropyl
polypeptoids,20,21 with methyl (= polysarcosine)22 and ethyl
side chains. Polypeptoids with C3 chains (n-propyl) had limited
tion under reduced pressure and the crude material was purified by
recrystallization techniques utilizing MeOH and Et2O. The resulting
N-isopropylaminoacetic acid was collected as a white crystalline
solubility and those with C4 chains (n- and isobutyl) were
material (15.23 g, 0.130 mol, 48%) with a melting point of 198 °C. 1H
insoluble in water. Shortly after, Zhang et al.23 reported the
NMR (DMSO-d6, 400 MHz): δ 13.74 (br s, 1H), 9.10 (s, 2H), 3.82 (s,
2H), 3.34−3.25 (m, 1H), 1.23 (d, J = 6 Hz, 6H). (ii) A suspension of
2-(isopropylamino)acetic acid hydrochloride (15.46 g, 0.132 mol) was
solubility and cloud point temperatures (Tcp) of statistical
copolymers of poly(N-{ethyl/n-butyl} glycine) in aqueous
solutions. From these studies, we rationalized that the solubility
and suspected thermo-responsive behavior of poly(N-C3
glycine)s (C3 = n-propyl and also isopropyl, allyl, and
propargyl) warranted further investigations. Notably, to the
best of our knowledge, the poly(N-isopropyl glycine) has not
yet been reported. We envisioned comparable results, in
regards to LCST behavior and crystallinity, as described for the
poly(2-oxazoline)s. Furthermore, the unique steric and
electronic variations of the side chains would provide additional
insight into side-chain influences on the LCST behavior.
prepared in toluene (182 mL) and cooled via an ice bath. Once
cooled, a 1 M solution of NaOH (398 mL) was added and stirred for
15 min. Benzyl chloroformate (19.13 mL, 0.134 mol) was injected into
the stirring reaction. After 4 h, stirring was halted to allow for phase
separation. The aqueous layer was extracted from the organic layer and
the pH was adjusted to 1−2 via concentrated HCl. The layers were
then recombined and shaken in a separatory funnel. The organics were
extracted in EtOAc (3 × 300 mL) and subsequentially dried over
MgSO4. Once filtered, the organic layers were concentrated down, by
reduced vacuum techniques, to yield N-carboxybenzyl(Cbz)-N-
isopropylaminoacetic acid as a viscous yellow oil (31.51 g, 0.125
mol, 95%). 1H NMR (CDCl3, 400 MHz): δ 9.18 (br s, 2H), 7.45−7.22
(m, ∼10H), 5.18, 5.13 (ss, 2H + 2H), 4.60−4.43−4.27 (mm, 1H +
1H), 3.96, 3.90 (ss, 2H + 2H), 1.13 (d, J = 7 Hz, 12H). GC−MS
(MSD): Rt = 10.8 min; (EI) m/z 251.2 (M+, [C13H17NO4]+ = 251.12),
116.1 (M+ − Cbz, [C5H10NO2]+ = 116.07), 91.1 (benzyl, [C7H7]+ =
91.05), 43.1 (isopropyl, [C3H7]+ = 43.05). (iii) N-Cbz-N-isopropyl-
aminoacetic acid (18.73 g, 74 mmol) was combined with acetic
anhydride (14.1 mL, 0.149 mol) in a 2-neck round-bottom flask, fitted
with a water cooled condenser, under positive argon pressure. Acetyl
chloride (10.74 mL, 0.152 mol) was injected into this stirring solution
and the reaction was subsequentially refluxed for 5 h. The volatile
organics were removed via rotoevaporation techniques to afford a
crude yellow oil. Fraction distillation, under vacuum (0.03 mbar) with
a heated condenser (∼50 °C), afforded a crystalline material with
EXPERIMENTAL SECTION
■
Materials. Unless otherwise stated, all reagents were purchased and
used as is from commercial sources (Sigma-Aldrich, Acros, or Alfa
Aesar). Anhydrous solvents (N,N-dimethylformamide (DMF), N,N-
dimethylacetamide (DMA), N-methyl-2-pyrrolidone (NMP), and
benzonitrile (PhCN)) were purchased in bottles with a septum over
molecular sieves; dichloromethane and acetic anhydride (Ac2O) were
distilled from calcium hydride, tetrahydrofuran (THF) from sodium.
All other solvents (1,4-dioxane, ethyl acetate (EtOAc), methanol
(MeOH), etc.) were used as is from their respective commercial
sources. Freshly distilled solvents were stored over activated molecular
sieves (3 Å; 3−5 mm beads) and sealed under argon (flowed through a
calcium chloride drying tube). Thin layer chromatography (TLC; 0.2
1
yellow discoloration. H NMR analysis suggests that the byproduct,
mm silica gel with fluorescent indicator; Polygram SIL G/UV254
)
benzyl chloride, was present as an impurity. Removal of this impurity
was achieved via a liquid-melt-extraction technique utilizing heptanes.
N-isopropyl glycine NCA was collected as a white crystalline material
plates with visualizing agents (UV, iodine chamber, or KMnO4 stain)
were utilized to monitor intermediate steps. Flash chromatography
techniques were utilized with nitrogen pressure to push eluent/sample
mixture through silica gel (pore size 60 Å; Fluka). All laboratory
equipment was cleaned and oven-dried prior to use. The reaction
flasks, for polymerization, were prepared by flash drying with a heat
gun and reduced pressure (0.012 mbar).
1
(7.68 g, 53.3 mmol, 72%) with a melting point of 59 °C. H NMR
(CDCl3, 400 MHz): δ 4.34−4.22 (m, 1H), 4.03 (s, 2H), 1.26 (d, J =
6.8 Hz, 6H). 13C NMR (CDCl3, 100 MHz): δ 165.7, 151.2, 45.2, 44.7,
19.9. FT-IR (crystal): νmax 2981, 2940, 2888, 1842, 1755, 1452, 1416,
̃
1395, 1376, 1361, 1290, 1242, 1198, 1154, 1130, 1075, 966, 900, 876,
808, 750, 682, 620 cm−1. GC−MS (MSD): Rt = 12.7 min; (EI) m/z
143.1 (M+, [C6H9NO3]+ = 143.05), 128.0 (M+ − O + H,
[C6H9NO2]2+H = 128.07), 56.1 (M+ − isopropyl −2O,
[C2H2NO]3+ = 56.01), 43.1 (isopropyl, [C3H7]+ = 43.05).
Monomer Synthesis. The N-carboxyanhydrides (NCAs) of the
N-C3 glycines, C3 = n-propyl (G−-NCA), allyl (G=-NCA), propargyl
(G≡-NCA), and isopropyl (G⊥-NCA), were synthesized in three steps,
starting from glyoxylic acid or α-bromo ethylacetate and C3-amine, as
described elsewhere (Scheme 1).18,24 Exemplary procedure for the N-
isopropyl glycine NCA: (G⊥-NCA, 3-isopropyloxazolidine-2,5-dione):
(i) Glyoxylic acid monohydrate (59.92 g, 0.651 mol) was dissolved in
reagent grade CH2Cl2 (1.3 L) and mixed with isopropylamine (23.2
mL, 0.271 mol). The reaction was allowed to stir for 24 h at room
temperature, and then the volatile organics were removed by
rotoevaporation under reduced pressure. The remaining intermediate
material was dissolved and refluxed (∼24 h) in a 1 M hydrochloric acid
solution (1.3 L). The aqueous medium was removed by rotoevapora-
Polymer Synthesis. The majority of poly(N-C3 glycine)s, C3 = n-
propyl (PG− ), allyl (PG= ), and propargyl (PG≡ ), were synthesized
n
n
n
as described elsewhere.18 Poly(N-isopropyl glycine)s (PG⊥ ) were
n
prepared as follows (exemplary procedure for the PG⊥49): G⊥-NCA
was transferred into a flash dried Schlenk flask prepared under an
argon atmosphere. The sealed reactor flask was heated in an oil bath to
∼5−10 °C above the melting point of the NCA. Once the substance
was completely melted, the initiator solution (1 M benzylamine/NMP;
581
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
Figure 1. MALDI−ToF mass spectra of the series of synthesized poly(N-C3 glycine)s, C3 = n-propyl (PG− ) (matrix/cation: DCTB/K+), allyl
n
(PG= ) (DCTB/Na+), propargyl (PG≡ ) (dithranol/Na+), and isopropyl (PG⊥ ) (DCTB/K+) (left to right).
n
n
n
[NCA]0/[BnNH2]0 = 300) was injected into the reaction flask, and
the mixture was heated to 160 °C. With time the oil became more
viscous and around 2.5 h no further change was observed. The
reaction was removed from the oil bath and cooled to room
temperature. As crude 1H NMR and GC−MS analysis detected
unreacted monomer, the crude material was purified via dialysis
(MWCO 500 Da, regenerated cellulose) against H2O (∼3 d) then
MeOH (∼3 d). The remaining sample was concentrated down, via
rotoevaporization under reduced pressure, to a slightly yellow material.
1H NMR (DMF-d7, 600 MHz): δ 7.39−7.25 (m, 5H), 4.86−4.65 (br,
∼59H), 4.61−3.91 (m, ∼99H), 1.39−0.71 (m, ∼292H). 13C NMR
102 and 103 Å. Solutions containing ∼0.15 wt % polymer were filtered
through 0.45 μm filters; the injected volume was 100 μL. Calibration
was done with polystyrene standards (Polymer Standards Service PSS,
Mainz, Germany). The NTeqGPC V6.4 software (Wolfgang Schupp,
Ober-Hilbersheim, Germany) was used for data recording and
handling. Matrix-assisted laser desorption/ionization time-of-flight
mass spectrometry (MALDI−ToF MS): Measurements were carried
out with a Bruker Autoflex III Smartbeam MALDI (Bruker Daltonik),
equipped with a laser working at 356 nm. Acceleration voltage was 20
kV and 4 × 500 shots at different places of the spot were recorded.
Instrument software FlexControl and FlexAnalysis was used for data
handling. Samples were prepared by applying the dried droplet
method. 50 μL of matrix trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-
propenylidene]malononitrile, DCTB, or dithranol (10 mg mL−1 in
THF) were premixed with 20 μL of sample (2 mg mL−1 in THF) and
either sodium or potassium trifluoroacetate. One μL of that mixture
was dropped on the target. Thermal analysis: Differential scanning
calorimetry (DSC) was measured with a Mettler Toledo DSC1/
TC100 at heating/cooling rates of 1 or 20 °C min−1 under a nitrogen
atmosphere. Glass transition temperatures were determined from
second heating curves at 20 °C min−1. Thermogravimetric analysis
(TGA) was carried out on a Netzsch TG 209 F1 at 20 °C min−1 under
a nitrogen atmosphere. Turbidity measurements were conducted on a
T70+ UV−visible spectrometer fitted with a Peltier element (Tresser
Instruments, Germany) operating at λ = 660 nm. The aqueous
solutions in the cuvettes were heated/cooled at a rate of 1 °C min−1
without stirring. The temperature at which the heating scan
transmittance dropped to 80% was taken as the cloud point
temperature. Scanning electron microscopy (SEM) images were
taken with a Gemini Leo 1550 microscope operating at 3 kV. The
samples were loaded on carbon-coated stubs and sputtered with Au/
Pd 80/20 alloy prior to imaging.
(DMF-d7, 150 MHz): δ 170.8, 46.7, 44.9, 20.1; FT-IR (crystal) νmax
̃
2972, 2933, 2873, 1645, 1464, 1409, 1362, 1327, 1298, 1253, 1198,
1126, 1078, 996, 923, 846, 631, 589, 575 cm−1. MALDI−ToF−MS:
Mn = 5000 Da, Δm = 99.2 ([C5H9NO] = 99.06 Da), residual mass,
app
app
r.m. = 43.0. SEC (PS cal.): Mn = 5430 g mol−1, Mw = 7200 g
mol−1, polydispersity index PDI = 1.26.
Analytical Instrumentation and Methods. Melting points were
determined with a MEL-TEMP II apparatus, produced by Laboratory
Devices Inc. (USA), in open capillaries. Gas chromatography−mass
spectrometry (GC−MS): An Agilent Technologies GC 6890N MS
5975 equipped with an autosampler was utilized for analysis of
compounds, purity, and detection of the monomer for polymerization
monitoring. Samples were prepared in 2 mL septum-sealed vials at
concentrations between 1−10 mg mL−1. Enhanced ChemStation
software was utilized for the measuring program and analysis.
Monomer measuring program: An aliquot of 1 μL was injected into
the heating block (200 °C), split (50:50), and flowed (He; 0.757 bar)
through the heated (50−200 °C; 2 min hold then 8 °C min−1)
column. After a 2 min solvent delay, the data was captured (duration
of 21.75 min) by the aforementioned software and subsequentially
1
analyzed. Nuclear magenetic resonance (NMR) spectroscopy: H and
13C NMR measurements were conducted on Bruker DPX-400 or
Varian 600 MHz instruments at room temperature. Samples were
prepared either in CDCl3, DMSO-d6, or DMF-d7 (Aldrich) at
concentrations of ∼5−25 mg mL−1; signals were referenced to the
RESULTS AND DISCUSSION
■
Synthesis. The N-C3 glycine N-carboxyanydrides (NCAs),
C3 = n-propyl (G−-NCA), allyl (G=-NCA), propargyl (G≡-
NCA), and isopropyl (G⊥-NCA), were synthesized in three
steps as outlined in Scheme 1.18,24 The ring-opening polymer-
izations of the NCAs, except G⊥-NCA (see below), were
initiated with benzylamine ([NCA]0/[BnNH2]0 = n = 25, 50,
and 75) and conducted in dry benzonitrile solution (∼10 wt %)
for several days at room temperature.18 For convenience, these
polymerizations were carried out in Teflon sealable round-
bottom flasks over the bench at reduced pressures.25 A series of
1
signal of the solvent at δ H 7.26 (13C 77.0), 2.50 (39.5), and 8.02
(162.6) ppm, respectively. Fourier transform infrared (FT-IR)
spectroscopy: Measurements were completed on a Varian 1000 FT-
IR, Scimitar series, equipped with an interchangeable sample head.
Samples were loaded onto a diamond attenuated total reflectance
(ATR) accessory and the Varian Resolutions FTS 1000 software
allowed for measurements (transmittance; resolution =4; 32 scans)
and the recording of data. Size exclusion chromatography (SEC): SEC
with simultaneous UV (270 nm) and RI detection was performed in
NMP (+0.5 wt % LiBr) at +70 °C, flow rate: 0.8 mL min−1, using a
column set of two 300 × 8 mm2 PSS-GRAM (spherical polyester
particles with an average diameter of 7 μm) columns with porosities of
poly(N-C3 glycine)s PG− , PG= , and PG≡n was isolated in high
n
n
yields (>80%) and characterized by 1H NMR and FT-IR
spectroscopy, SEC (data not shown), and MALDI−ToF mass
582
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
Table 1. Molecular Characteristics of the Poly(N-C3 glycine)s (n, Number-Average Degree of Polymerization; Mn, Number-
Average Molecular Weight; Mnapp, Apparent Number-Average Molecular Weight; PDI, Apparent Polydispersity Index)
b
c
MALDI−ToF MS
Mn (kDa)
SEC
Mn (kg mol−1
)
a
app
sample
side chain R
end groups α/ω
NMR
n
n
PDI
PG−
n-propyl
n-propyl
n-propyl
allyl
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/Ac
BnNH/H
BnNH/H
22
37
55
21
35
45
21
37
51
17
27
50
2.37
3.85
5.64
2.63
4.51
6.63
2.07
3.80
6.46
1.78
2.81
22
37
55
24
44
68
21
39
67
16
27
49
2.5
4.1
6.2
3.8
6.9
10.1
3.1
5.4
9.9
2.7
3.9
5.4
1.16
1.19
1.19
1.05
1.04
1.03
1.05
1.04
1.04
1.53
1.14
1.26
22
PG−
37
PG−
55
PG=
24
PG=
allyl
44
PG=
allyl
68
PG≡
propargyl
propargyl
propargyl
isopropyl
isopropyl
isopropyl
21
PG≡
39
PG≡
67
PG⊥
16
PG⊥
27
PG⊥
5.00
49
a
b
1H NMR end group analysis, n = integral(−C(O)CH2N−) × 2.5/integral(−C6H5). n = (Mn − BnNH − (Ac or H))/Δm, Δm = mass of repeat
c
unit. Apparent molecular weight distribution determined from the RI trace using a polystyrene calibration curve.
spectrometry (Figure 1) to confirm the chemical structures, as
drawn in Scheme 1, and to access number-average molecular
weights (degrees of polymerization) and molecular weight
distributions (polydispersity indexes, PDI) − results are
summarized in Table 1. Notably, there is excellent agreement
between the number-average degrees of polymerization of
PG−n determined by 1H NMR and MALDI−ToF MS, however,
not so for PG= , and PG≡ (deviations are increasing with
NCA (as it was observed by Kricheldorf et al. for sarcosine
NCA in bulk at 120 °C).28
Solubility and Thermal Properties. The polymer
samples, which were precipitated into petroleum ether, diethyl
ether, or dialyzed against methanol and dried in vacuo at room
temperature, with lowest molecular weights (n ∼ 16−24) were
subjected to first solubility tests in deionized water. PG−
22
demonstrated limited solubility, as previously reported by
n
n
Luxenhofer,18 and also PG=24 and PG≡21 did not readily solvate
increasing chain length). The mass spectrometric results were
taken as the absolute measure of n.
in the aqueous environment. Just the PG⊥ could be directly
16
G⊥-NCA, on the other hand, could not (or very slowly) be
polymerized under these conditions. Pioneering work by
Ballard and Bamford support unfavorable kinetics (<107 dm3
mol−1 s−1) between isopropyl amine and G⊥-NCA.26 Inference
from this work suggested that the steric hindrance of the
propagating species and the low concentrations of the
monomer within a defined volume lower the statistical
possibility of a favorable bimolecular reaction to occur (i.e.,
nucleophilic attack at the C5 position). Therefore, the increase
in the concentration of monomer within a defined dimension
should readily overcome the aforementioned limitation. In as
much, polymerization occurred when the benzylamine was
directly injected into the molten G⊥-NCA (mp 59 °C)
([NCA]0/[BnNH2]0 = n = 25, 75, and 300) and then heated
to 160 °C for 2.5 h. Reactions were optionally quenched with
acetic anhydride and products purified by dialysis (see
dissolved in water, up to 40 g L−1 (PG⊥ is soluble in water as
n
well as in organic solvents like methanol, DMF, NMP, and
chloroform). The solubility of PG−22 and PG=24, but not PG≡
,
21
could subjectively be improved by changing the temperature of
the medium, agitation, and various weight concentrations.
Arguably, the variances in water solubility between the poly(N-
C3 glycine) samples could not simply be explained by a
hydrophilic/hydrophobic understanding. Instead we suspected
that the difference could be due to (semi)crystallinity of the
samples (see reports in the literature about the crystalline
structures of peptoids29 and related polyoxazolines13−16,30).
The thermal histories of the poly(N-C3 glycine)s were
analyzed by DSC; the exemplary first heating curves at 1 °C
min−1 of PG−37, PG=44, PG≡39, and PG⊥27 are shown in Figure 2
(gray lines). Endothermic melting transitions were observed for
the samples PG− (two melting peaks at Tm = 190−197 °C)
37
Experimental Section). The three PG⊥ samples exhibited the
and PG=44 (one melting peak at Tm = 160 °C), confirming their
n
expected chemical structures (1H NMR, FT-IR, MALDI−ToF
MS) and degrees of polymerization of n ∼ 17, 27, and 50 (1H
NMR) (deviations from expected values are due to incomplete
monomer conversions, estimated to be <40%). Although the
poly(N-isopropyl glycine) chains underwent fragmentation
during the MALDI−ToF mass spectrometric analysis (Figure
1, right), the experimental Mn values and calculated average
degrees of polymerization (n ∼ 16, 27, and 49) are in very good
semicrystalline structures. PG− has a considerably higher
37
crystallinity than PG=44, as estimated from the respective
melting enthalpies (areas under the endothermic peaks).
Notably, no crystallization exotherm was revealed before
melting of the PG−37, suggesting that this sample was highly,
if not fully, crystalline (see below). Melting transitions could
not be detected for PG≡39 and PG⊥27. Any melting transition of
PG≡39, however, could be buried under the strong exotherm
observed at higher temperature, T > 150 °C, which is attributed
to a chemical reaction (not yet identified) followed by
decomposition (sample turned red between 150−200 °C and
into a black solid >200 °C without considerable mass loss until
27
1
agreement with H NMR results. SEC analyses (data not
shown) indicated monomodal and fairly narrow molecular
weight distributions for PG⊥ and PG⊥49, but an apparently
27
multimodal distribution for PG⊥16 (PDI ∼ 1.5), which however
is not observed by MALDI−ToF MS. Notably, SEC did not
detect any low molecular weight or cyclic chains, thus excluding
the occurrence of a self-initiated thermal polymerization of G⊥-
∼360 °C, TGA). PG⊥ appears to be the only sample being
27
completely amorphous. The glass transition temperatures of the
polymers were determined from the second heating curves
583
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
55), PG=n (n = 24, 44, and 68), and PG⊥n (n = 16, 27, and 49)
were prepared at room temperature or reduced temperatures
(5−15 °C). All sample solutions were visually confirmed to be
optically clear prior to turbidity measurements. In addition,
water was utilized as the zero reference (100% transmittance)
for the UV−vis detector. The turbidity curves of the 0.1 and 1.0
wt % polymer solutions are presented in Figure 3. Also
included in Figure 3 are the corresponding plots (phase
diagrams) of the cloud point temperatures (TCP), determined
as the temperature at which in the heating scan the
transmittance dropped to ∼80%, versus the polymer concen-
tration (0.1, 0.21, 0.34, 0.48, and 1.0 wt %).
The phase transition appeared to be reversible for all
solutions at concentrations below 1.0 wt %. Occasionally,
translucent ribbons or white film/precipitates were observed in
the lower 1/3 portion of the cuvette after the heating/cooling
cycle. A change of concentration might affect the subsequent
phase transition, possibly resulting in a shift of the solution
clearing to a higher temperature than TCP (as for instance
recognized in the turbidity curve of PG− at 1.0 wt %, Figure
22
3). Rather broad transitions, especially at low concentrations,
and pronounced hysteresis between the heating and cooling
scans (ΔT ∼ 3−10 °C) were observed for the PG−n and PG⊥
n
solutions, whereas the PG= solutions revealed very sharp
n
transitions with almost no hysteresis (similar to PNIPAM).3
Seemingly, the interchain association and dissociation process is
delicately affected by the electronic properties of the side chain
(unsaturation versus saturation) and less by the degree of
branching. The allyl groups appear to be easier rehydrated than
the propyl chains. Also worth to be considered are the possible
different degrees of chain entanglement and the kinetics of
trans/cis rotamerization of the 3° amide (as observed for
PIPOX).15
Figure 2. DSC first heating curves (1 °C min−1) of poly(N-C3
glycine)s, C3 = n-propyl (PG−37), allyl (PG=44), propargyl (PG≡39),
and isopropyl (PG⊥27) as obtained after precipitation/dialysis and
drying (gray lines) and after methanol treatment and drying in vaccuo
(black lines).
(data not shown) measured at 20 °C min−1: Tg ∼ 64 °C
(PG=44), 77 °C (PG−37), 108 °C (PG≡39), and 121 °C (PG⊥27).
Attempting to remove crystalline domains,31 the samples
were dissolved in methanol (chosen for its hydrogen donating
character and low boiling point, ∼65 °C) and subsequentially
the solvent removed under reduced pressure (complete
The cloud point temperatures (not to be confused with the
LCSTs) were found to be TCP ∼ 15−25 °C (PG− ), 27−54 °C
n
(PG= ), and 47−58 °C (PG⊥ ) (Figure 3, left). The TCPs
n
n
appeared relatively unaffected (ΔTCP ∼ 5 °C) by polymer
concentration and chain length, at least in the examined
experimental range, except those of PG−22 and PG=24 solutions.
The TCPs of the PG= solutions were strongly increasing with
1
removal of methanol confirmed by H NMR). The DSC first
24
decreasing concentrations and were shifted 3−20 °C to higher
heating curves of these methanol-treated samples are shown in
Figure 2 (black lines). The endothermic melting peaks and
crystallinities of PG− and PG= were found to be greatly
temperature as compared to the solutions of the higher
homologues (PG= and PG=68) (similar to PIPOX).3 This
44
37
44
behavior might be attributable to a better hydration of the
shorter chains rather than to contributions of the end groups.
reduced or completely erased (the exotherm at 130 °C in the
DSC scan of PG− is attributable to crystallization of the
37
The opposite, however unusual, trend was noticed for the PG−
sample,13 not observed for PG=44), whereas no structural
n
series, the PG−22 exhibiting the lowest TCP of all samples (apart
change could be recognized for PG≡ (the sample is
39
from PG≡ ). Similar observations (i.e., increasing TCP with
considered to be amorphous). The amorphous sample PG−
n
37
then demonstrated improved solubility in water, up to 40 g L−1,
increasing degree of polymerization) were reported for
dendronized polymer with peripheral oligo(ethylene oxide)
chains32 and poly(methoxy diethylene glycol acrylate)33 and
were explained by a tentatively poor shielding of short
hydrophobic backbones from the water phase. Here, inversely,
hydrocarbon chains are tethered to a polar polyglycine chain,
which implies that the solvation is improved (i.e. TCP is higher)
when the backbone is less shielded. Accordingly, in respect to
the TCP values, the shielding of the backbone decreases in the
order n-propyl > allyl > isopropyl, which seems to correlate
and the PG= appeared to be soluble in water at low
44
temperatures (plus time and agitation) and concentrations
below 20 g L−1. The PG≡39 was poorly soluble in methanol and
unaffected by this treatment and remained insoluble in water.
Acetone (bp ∼56 °C) and THF (bp ∼66 °C) were explored
separately as alternative treatment techniques whereas 0.1 M
NaOH solutions were investigated as an alternative aqueous
medium. None of these alternative treatments or conditions
improved the solubility of PG≡39 in aqueous media. In general,
the solubility of the amorphous poly(N-C3 glycine)s in water
with the rotational flexibility of the C3 chain. Notably, PG≡ is
n
decreased in the following order: PG⊥ ∼ PG− > PG=
≫
not or poorly soluble in water or methanol (see above),
indicating that the polyglycine backbone is most efficiently
shielded by propargyl groups. It is assumed that the solvation or
hydration of the chains is hindered by favorable intrachain
n
n
n
PG≡ (insoluble).
n
Thermoresponsive Solution Behavior. Here, 0.1−1.0 wt
% (1−10 g L−1) aqueous solutions of PG− (n = 22, 37, and
n
584
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
Figure 3. Turbidity curves (heating → cooling at 1 °C min−1) of the 0.1 and 1.0 wt % aqueous solutions of poly(N-C3 glycine)s, C3 = n-propyl
(PG− , n = 22, 37, and 55), allyl (PG= , n = 24, 44, and 68), and isopropyl (PG⊥ , n = 16, 27, and 49) (left) and plots of the cloud point temperatures
n
n
n
versus concentration (right).
coordination of the amide groups to the acetylene units
through hydrogen bonding (CO...H−CC) and/or n→π*
interactions (CO...CC).34,35 For confirmation, the intra-
chain interactions and also the polyglycine chain conformations
need to be further analyzed, for instance by 2D-NMR and FT-
IR/Raman spectroscopy.
PG−37 and PG=44 were annealed at 75 °C for 60 h and 70 °C for
24 h, respectively (annealing temperature was about 40 degrees
above TCP). The formed precipitates were isolated by filtration,
dried in vacuum for 12 h, and analyzed by DSC and SEM; the
results are presented in Figure 4. The crystallized PG− was
37
found to melt at 188−198 °C (two melting peaks), and
particles were ∼1 μm in diameter exhibiting a complex rose bud
Crystallization in Solution. Similar to PIPOX,14 the
possibility to produce crystalline particles by simple annealing
of dilute aqueous polymer solutions at above the cloud point
temperature was explored. Exemplarily, 0.5 wt % solutions of
type morphology. The PG= particles, which melted at 157−
44
165 °C (three melting peaks, possibly indicating the existence
of different polymorphs), were considerably larger, measuring
585
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
Figure 4. DSC first heating curves (1 °C min−1) (left) and scanning electron micrographs (scale bar = 3 μm, inset: 1 μm) (right) of the crystalline
poly(N-C3 glycine)s PG− and PG=44, obtained by the annealing of 0.5 wt % aqueous polymer solutions at 75 °C (60 h) or 70 °C (24 h),
37
respectively.
about 10 μm, and less uniform in shape and size as compared to
were found to increase in the order PG−n (TCP ∼ 15−25 °C) <
PG=n (27−54 °C) < PG⊥n (47−58 °C). Notably, the backbone
the PG− particles. Seemingly, a slight variation at the side
37
was most efficiently shielded for PG≡ , rendering these samples
chain, saturated versus unsaturated, had severe impact on the
phase separation (see above) and also on the morphology of
the crystalline particles. Further details on the structures and
crystallization kinetics/mechanism are not available yet.
n
insoluble in water, which might be attributable to intrachain
interactions between the amide groups and the acetylene units
(hydrogen bonding and/or n → π*-interactions). The sharpest
phase transition and the least pronounced hysteresis were
observed for PG= , which might be explained by a better
CONCLUSIONS
n
■
hydration of unsaturated as compared to saturated side chains.
A series of polypeptoids based on N-C3 glycines (C3 = n-
propyl, allyl, propargyl, and isopropyl) with molecular weights
of 1.8−6.6 kg mol−1 was prepared by the ring-opening
polymerization of the corresponding amino acid N-carboxy-
Similar to PIPOX, the PG− and PG= underwent
37
44
crystallization in 0.5 wt % aqueous solution upon long-term
annealing at ∼70−75 °C, i.e. above the cloud point
temperature. The formed precipitates were found to be
crystalline microparticles with melting points of 188−198 and
157−165 °C, respectively, and rose bud type morphology.
These fundamental investigations provided significant insight
into the thermo-responsive behavior and crystallization
phenomena of polypeptoids with well-defined structures.
However, further studies are required to address questions
toward the influence that structural features and electronic
properties appear to have on the solution behavior and
crystallinity. A key to the better understanding of the system
appears to be the conformation of the chains, which shall be
analyzed in detail by NMR, FT-IR, and Raman spectroscopy.
Also the kinetics and the mechanism of the polypeptoid
crystallization in solution deserves further attention (X-ray
scattering, calorimetry, and microscopy); results will be
reported in due time.
anhydrides. The poly(N-isopropyl glycine) (PG⊥ ), previously
n
unreported, was obtained by the bulk polymerization in the
melt. Initial attempts to dissolve the PG− (n-propyl), PG=
n
n
(allyl), and PG≡ (propargyl) in water were not successful, as
n
supported by preliminary investigations completed by
Luxenhofer et al.,18 although the PG⊥ demonstrated solubility
n
up to 40 g L−1. The insolubility of the PG− and PG= series
n
n
was caused by the semicrystallinity of the samples, which could
be removed by treatment with methanol. Afterward the samples
could be dissolved in water up to 20−40 g L−1. Overall, three
out of the four poly(N-C3 glycine)s were soluble and
demonstrated thermo-responsive behavior in water. The
cloud point temperatures (reflecting the hydration of polymer
chains) and the hysteresis (reflecting the interchain association
and dissociation process) were delicately affected by the
electronic and rotational flexibility of the side chains, less by
the degree of branching. The C3 chains contribute to the
shielding of the polyglycine backbone from the water phase,
thus affecting the hydration of the polymer chain and the cloud
point temperature. Accordingly, the cloud point temperatures
Future studies will be devoted to use this polypeptoid
platform for the preparation of smart (bio)functional micelles/
vesicles, microparticles, and hydrogels for possible biomedical
applications.
586
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587

Macromolecules
Article
(28) Kricheldorf, H. R.; Lossow, C. V.; Lomadze, N.; Schwarz, G. J.
Polym. Sci., Part A: Polym. Chem. 2008, 46, 4012−4020.
(29) Rosales, A. M.; Murnen, H. K.; Zuckermann, R. N.; Segalman,
R. A. Macromolecules 2010, 43, 5627−5636.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: Helmut.Schlaad@mpikg.mpg.de (HS).
(30) Litt, M.; Rahl, F.; Roldan, L. G. J. Polym. Sci: Part A-2 1969, 7,
463−473.
Author Contributions
‡These authors contributed equally.
(31) Diehl, C.; Schlaad, H. Chem.Eur. J. 2009, 15, 11469−11472.
(32) Roeser, J.; Moingeon, F.; Heinrich, B.; Masson, P.; Arnaud-Neu,
F.; Rawiso, M.; Mery, S. Macromolecules 2011, 44, 8925−8935.
(33) Miasnikova, A.; Laschewsky, A. J. Polym. Sci., Polym. Chem. 2012,
50, 3313−3323.
(34) Gorske, B. C.; Stringer, J. R.; Bastian, B. L.; Fowler, S. A.;
Blackwell, H. E. J. Am. Chem. Soc. 2009, 131, 16555−16567.
(35) Caumes, C.; Roy, O.; Faure, S.; Taillefumier, C. J. Am. Chem.
Soc. 2012, 134, 9553−9556.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Nora Fiedler, Jessica Brandt, Olaf Niemeyer, Marlies Grawert,
̈
Rona Pitschke, and Heike Runge are thanked for their
contributions to this work. Financial support was given by
the Max Planck Society.
REFERENCES
■
(1) Cohen Stuart, M. A.; Huck, W. T. S.; Genzer, J.; Muller, M.;
Ober, C.; Stamm, M.; Sukhorukov, G. B.; Szleifer, I.; Tsukruk, V. V.;
Urban, M.; Winnik, F.; Zauscher, S.; Luzinov, I.; Minko, S. Nat. Mater.
2010, 9, 101−113.
(2) Gil, E. S.; Hudson, S. M. Prog. Polym. Sci. 2004, 29, 1173−1222.
(3) Dimitrov, I.; Trzebicka, B.; Muller, A. H. E.; Dworak, A.;
Tsvetanov, C. B. Prog. Polym. Sci. 2007, 32, 1275−1343.
(4) Aseyev, V.; Tenhu, H.; Winnik, F. Adv. Polym. Sci. 2011, 242, 29−
89.
̈
(5) Hoogenboom, R. Angew. Chem., Int. Ed. 2009, 48, 7978−7994.
(6) Schlaad, H.; Diehl, C.; Gress, A.; Meyer, M.; Demirel, A. L.; Nur,
Y.; Bertin, A. Macromol. Rapid Commun. 2010, 31, 511−525.
(7) Hoogenboom, R.; Schlaad, H. Polymers 2011, 3, 467−488.
(8) Weber, C.; Hoogenboom, R.; Schubert, U. S. Prog. Polym. Sci.
2012, 37, 686−714.
(9) Luxenhofer, R.; Han, Y.; Schulz, A.; Tong, J.; He, Z.; Kabanov, A.
V.; Jordan, R. Macromol. Rapid Commun. 2012, 33, 1613−1631.
(10) Kelly, A. M.; Wiesbrock, F. Macromol. Rapid Commun. 2012, 33,
1632−1647.
(11) Park, J.-S.; Kataoka, K. Macromolecules 2006, 39, 6622−6630.
(12) Huber, S.; Hutter, N.; Jordan, R. Colloid Polym. Sci. 2008, 286,
1653−1661.
(13) Demirel, A. L.; Meyer, M.; Schlaad, H. Angew. Chem., Int. Ed.
2007, 46, 8622−8624.
(14) Diehl, C.; Cernoch, P.; Zenke, I.; Runge, H.; Pitschke, R.;
Hartmann, J.; Tiersch, B.; Schlaad, H. Soft Matter 2010, 6, 3784−3788.
(15) Katsumoto, Y.; Tsuchiizu, A.; Qiu, X. P.; Winnik, F. M.
Macromolecules 2012, 45, 3531−3541.
(16) Diehl, C.; Dambowsky, I.; Hoogenboom, R.; Schlaad, H.
Macromol. Rapid Commun. 2011, 32, 1753−1758.
(17) Bellomo, E. G.; Wyrsta, M. D.; Pakstis, L.; Pochan, D. J.;
Deming, T. J. Nat. Mater. 2004, 3, 244−247.
(18) Fetsch, C.; Grossmann, A.; Holz, L.; Nawroth, J. F.; Luxenhofer,
R. Macromolecules 2011, 44, 6746−6758.
(19) Hanson, B.; Bedrov, D.; Magda, J. J.; Smith, G. D. Eur. Polym. J.
2010, 46, 2310−2320.
(20) Kirshenbaum, K.; Barron, A. E.; Goldsmith, R. A.; Armand, P.;
Bradley, E. K.; Truong, K. T. V.; Dill, K. A.; Cohen, F. E.;
Zuckermann, R. N. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 4303−4308.
(21) Zhang, D.; Lahasky, S. H.; Guo, L.; Lee, C.-U.; Lavan, M.
Macromolecules 2012, 45, 5833−5841.
(22) Gallot, B. ACS Symp. Ser. 1991, 448, 103−113.
(23) Lahasky, S. H.; Hu, X.; Zhang, D. ACS Macro Lett. 2012, 1,
580−584.
(24) Robinson, J. W.; Schlaad, H. Chem. Commun. 2012, 48, 7835−
7837.
(25) Habraken, G. J. M.; Wilsens, K.; Koning, C. E.; Heise, A. Polym.
Chem. 2011, 2, 1322−1330.
(26) Ballard, D. G. H.; Bamford, C. H. J. Chem. Soc. 1958, 355−360.
(27) Fetsch, C.; Luxenhofer, R. Macromol. Rapid Commun. 2012, 33,
1708−1713.
587
dx.doi.org/10.1021/ma302412v | Macromolecules 2013, 46, 580−587