Synthesis of novel hyperbranched poly(ester-amide)s based on neutral α-amino acids via "aD + CBB′" couple-monomer approach

Article abstract of DOI:10.1002/pola.24340

A series of novel hyperbranched poly(ester-amide)s (HBPEAs) based on neutral α-amino acids have been synthesized via the "AD + CBB′" couple-monomer approach. The ABB′ intermediates were stoichiometrically formed through thio-Michael addition reaction because of reactivity differences between functional groups. Without any purification, in situ self-polycondensations of the intermediates at elevated temperature in the presence of a catalyst afforded HBPEAs with multihydroxyl end groups. The degrees of branching (DBs) of the HBPEAs were estimated to be 0.40-0.58 and 0.24-0.54 by quantitative ¹³C NMR with two different calculation methods, respectively, depending on polymerization conditions and structure of monomers. The influences of catalyst, temperature, and intermediate structure on the polymerization process and molecular weights as well as properties of the resultant polymers were investigated. FTIR, NMR, and DEPT-135 NMR analyses revealed the branched structure of the resultant polymers. The HBPEAs possess moderately high molecular weights with broad distributions, glass transition temperatures in the range of -25.5 to 36.5 °C, and decomposition temperatures at 10% weight loss under nitrogen and air in the regions of 243.4-289.1 °C and 231.4-265.6 °C, respectively. Among them, those derived from D,L-phenylalanine display the lowest degree of branching, whereas the highest glass transition temperature and the best thermal stability.

Full text of DOI:10.1002/pola.24340

Synthesis of Novel Hyperbranched Poly(ester-amide)s Based on
Neutral a-Amino Acids via ‘‘AD 1 CBB0’’ Couple-Monomer Approach
YOU-MEI BAO,1,2 XIAO-HUI LIU,1 XIU-LAN TANG,3 YUE-SHENG LI1,3
1State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry,
Chinese Academy of Sciences, Changchun 130022, China
2Graduate School of the Chinese Academy of Sciences, Changchun Branch, Changchun 130022, China
3College of Environment and Chemical Engineering, Yanshan University, Qinhuangdao 066004, China
Received 20 July 2010; accepted 24 August 2010
DOI: 10.1002/pola.24340
Published online 5 October 2010 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: A series of novel hyperbranched poly(ester-amide)s
(HBPEAs) based on neutral a-amino acids have been synthe-
sized via the ‘‘AD þ CBB0’’ couple-monomer approach. The
ABB0 intermediates were stoichiometrically formed through
thio-Michael addition reaction because of reactivity differences
between functional groups. Without any purification, in situ
self-polycondensations of the intermediates at elevated tem-
perature in the presence of a catalyst afforded HBPEAs with
multihydroxyl end groups. The degrees of branching (DBs) of
the HBPEAs were estimated to be 0.40–0.58 and 0.24–0.54 by
quantitative 13C NMR with two different calculation methods,
respectively, depending on polymerization conditions and
structure of monomers. The influences of catalyst, tempera-
ture, and intermediate structure on the polymerization process
and molecular weights as well as properties of the resultant
polymers were investigated. FTIR, NMR, and DEPT-135 NMR
analyses revealed the branched structure of the resultant poly-
mers. The HBPEAs possess moderately high molecular weights
with broad distributions, glass transition temperatures in the
range of ꢀ25.5 to 36.5 ꢁC, and decomposition temperatures at
10% weight loss under nitrogen and air in the regions of
243.4–289.1 ꢁC and 231.4–265.6 ꢁC, respectively. Among them,
those derived from D,L-phenylalanine display the lowest degree
of branching, whereas the highest glass transition temperature
C
V
and the best thermal stability.
2010 Wiley Periodicals, Inc.
J Polym Sci Part A: Polym Chem 48: 5364–5374, 2010
KEYWORDS: ‘‘AD 1 CBB0’’; amino acid; biocompatibility; hyper-
branched; poly(ester-amide)s; polyesters; step-growth polymer-
ization; strategy; thiol-ene chemistry
INTRODUCTION Amino acids play a significant role in serv-
ing as the building blocks of proteins and intermediates in
metabolism of all living things. These compounds are also
low cost and efficient feedstocks especially in chiral organic
synthesis and biochemistry. Importantly, amino acids have
been increasingly developed as the components of a range of
biodegradable and biocompatible polymers to modify their
physical properties and reactivities. Generally, the resultant
polymers possess linear,1–7 grafted,8–11 branched, or star-
shaped12–15 architecture. In recent years, significant efforts
have focused on the synthesis and application of a new class
of modern polymer called hyperbranched polymer (HBP)
with randomly branched topology16,17 as they exhibit many
special merits,18 such as three-dimensional globular architec-
ture, low viscosity, high solubility, abundance of functional
end groups, and internal cavities in the molecule.19,20
Harada and Fox investigated the thermal homo- and co-poly-
merization of L-lysine and reported the formation of water-
soluble polymer materials.21,22 Although it was speculated
that the polymers exhibited high degrees of branching and
three-dimensional structure, no detailed characterization
data were provided. Scholl et al. investigated the thermal po-
lymerization of L-lysine monohydrochloride and character-
ized the structure of the resultant hyperbranched polymers
by means of 1H NMR spectroscopy.29(a) The DBs of the
hyperbranched polylysines obtained were typically 0.35–
0.45, which are smaller than the predicted value of 0.5.29(a)
They further discussed the feasibility of three approaches to
control polymer architecture during the thermal hyper-
branched polymerization of L-lysine hydrochloride.29(b) Menz
and Chapman explored the synthesis of hyperbranched poly-
lysine using the N-hydroxysuccinimide ester of L-lysine dihy-
drochloride as the AB2 building block. The DBs of these poly-
mers ranged from 0.32 to 0.64.30 Rohlfing reported the
thermal copolymꢁerization of aspartic acid and glutamic acid
at less than 100 C. Although the preparations have not been
Amino acid-based HBPs, mainly hyperbranched polypeptides
date back to the 1950s, when a number of authors investi-
gated the thermal polymerization of AB2-type amino acids
such as L-aspartic acid, L-glutamic acid, and L-lysine.21–28
Additional Supporting Information may be found in the online version of this article. Correspondence to: Y.-S. Li (E-mail: ysli@ciac.jl.cn)
C
V
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 48, 5364–5374 (2010) 2010 Wiley Periodicals, Inc.
5364
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
characterized in detail, one demonstrated property of the
polymer is the formation of microsphere.27 Recently, Pan and
coworkers have successfully prepared well-defined hyper-
branched polystyrenes by polymerization of AB2 macromo-
nomer derived from L-aspartic acid.31 However, most of the
amino acid-based HBPs were prepared from acidic and (or)
basic amino acids, typically glutamic acid, aspartic acid, and
lysine, whereas those based on neutral amino acids are rare,
except some examples copolymerized with acidic or basic
amino acids.22,24,25 Recently, a series of novel hyperbranched
poly(ester-amide)s (HBPEAs) from neutral amino acid and
gallic acid were reported by our group.32 The hydrolytic and
enzymatic degradation studies indicated that the HBPEAs
were degradable hydrolytically as well as enzymatically, and
the rate of hydrolytic degradation increases with the pH
value of the solution.32 The attractive properties of the
degradable polymers promote us to further synthesize other
hyperbranched polymers from neutral amino acids.
lyzer. The sample was heated from 40.0 to 900.0 ꢁC with a
heating rate of 10 ꢁC/min under nitrogen and air. HPLC
were obtained on a Waters 1525 binary HPLC pump and
2487 dual k absorbance detector with a Symmetry C18 5 lm
4.6 ꢂ 150 mm column as immobile phase and methanol as
mobile phase. ESI-MS was measured by LCQ ion trap instru-
ment (Finnigan MAT, San Jose, CA) with an electrospray
source in positive or negative ion mode. Electrospray ꢁvoltage
was 5.0 kV, and capillary temperature was set as 260 C. Ele-
mental analysis was determined with an Elementar Analy-
sensysteme GmbH VarioEL (Germany) elemental analyzer.
Size exclusion chromatography (SEC) was performed with a
Waters 1525 fitted with two columns (Styragel HT3 and
HT4 THF 7.8 ꢂ 300 mm column) connected in series and
2414 refractive index detector with TEDIA dimethylforma-
mide (DMF) containing 0.05 M LiBr as the mobile phase. The
SEC measurements were calibrated against narrow-dispersity
polystyrene standards.
In this article, we reported the synthesis of another class of
HBPEAs derived from neutral a-amino acids via a technique
termed as ‘‘AD þ CBB0,’’ which is based on nonequal reactiv-
ities of different functional groups. The rapid reaction
between C and D groups resulted in the predominant forma-
tion of an ABB0 intermediate, which was subjected to in situ
self-polycondensation in the presence of a catalyst to obtain
hyperbranched macromolecules. Influences of catalyst, tem-
perature, and steric effect of the intermediate on the poly-
merization and properties of resultant polymers were inves-
tigated. The hyperbranched structure of the resulting
Syntheses of Vinyl Monomers AD (2a–d) Containing
Amino Acid Residues
First, the various amino acids were transformed to their
hydrochloride salts of amino acid methyl esters (1a–d) by
reacting with 15 equiv methanol and 2 equiv thionyl chlo-
ride at room temperature according to the literature proce-
dure.33 Subsequently, the acylation processes of compounds
1a–d were accomplished by treating with acryloyl chloride
in the presence of excess Et3N to give the AD monomers
(2a–d) according to the ref. 34.
Syntheses of ABB0 Intermediates (3a–d) by
Thio-Michael Addition Reaction
1
polymers was verified by FTIR in combination with H NMR,
13C NMR, and DEPT-135 NMR techniques.
In a flask, compound 2b (1.57 g, 10 mmol), 2 mL of CH2Cl2,
and 0.14 mL of Et3N (10 mol % with respect to the com-
pound 2b) under nitrogen atmosphere were placed. The mix-
ture was stirred for a few minutes until compound 2b was
dissolved. Then 1-thioglycerol (0.96 mL, 10 mmol) was
added slowly into the flask immersed in ice-water bath, and
then the solution was kept at room temperature for 4 h.
Under reduced pressure, CH2Cl2 was removed from the reac-
tion system to yield green-yellow viscous liquid. Intermedi-
ates 3a, 3c, and 3d were prepared analogously.
EXPERIMENTAL
General Procedures and Materials
Glycine, D,L-alanine, D,L-2-aminobutyric acid, and D,L-phenylal-
anine were purchased from ACROS and used as received. 1-
Thioglycerol (90 wt % aqueous solution) was purchased
from Aladdin and used without any purification. Acryloyl
chloride was used as received. CH2Cl2 was purified by a sol-
vent purification system (MBraun, Germany). Triethylamine
(Et3N) was distilled over CaH2 after refluxing for 12 h. Other
organic reagents and solvents were analytically pure and
used without purification.
3b
Purity (HPLC): 94.8%. 1H NMR (300 MHz, D2O, ppm): 1.25
(d, ACHCH3, 3H), 2.41–2.70 (m, ASCH2CH(OH)A,
ASCH2CH2A, 6H), 3.42 (dd, J1 ¼ 6.3 Hz, J2 ¼ 11.4 Hz,
ACHH(OH)A, 1H), 3.51 (dd, J1 ¼ 3.9 Hz, J2 ¼ 11.4 Hz,
ACHH(OH)A, 1H), 3.58 (s, AOCH3, 3H), 3.67 (m,
ASCH2CHA, 1H), 4.27 (q, ANHCHA, 1H). 13C NMR (100
MHz, D2O, ppm): 14.5, 26.0, 32.9, 33.8, 47.2, 51.4, 62.7, 69.1,
172.7, 173.6. ESI-MS (m/z): calcd. for C10H19NO5S, 265.1;
found, 190.0 ([Fragment þ H]þ); 247.8 ([M ꢀ H2O þ H]þ);
265.9 ([M þ H]þ); 288.1 ([M þ Na]þ). Anal. Calcd. for
1H NMR spectra were recorded on a Bruker AV 300 MHz
spectrometer, 13C NMR spectra were recorded using a Varian
Unity 400 spectrometer operating at 100.0 MHz with D2O or
DMSO-d6 as the solvent, the residual 1H solvent peak as ref-
erence and the solvent carbon signal as the standard, respec-
tively. FTIR spectra were recorded on a Bio-Rad FTS-135
spectrophotometer. Glass transition temperature (Tg) was
measured by differential scanning calorimetry (DSC) (TA
Instruments DSC Q20) under nitrogen atmosphere with the
heating rate of 10 ꢁC/min from ꢀ50 to 100 ꢁC, The speci-
mens were crimp sealed in aluminum crucibles. The Tg was
taken as the midpoint of the inflection tangent, upon the sec-
ond heating scan. Thermogravimetric analysis (TGA) was
conducted on a Perkin-Elmer Pyris 1 thermogravimetric ana-
C10H19NO5S: C, 45.27; H, 7.17; N, 5.02; S, 12.07. Found: C,
45.39; H, 7.21; N, 5.03; S, 12.02.
3a
Purity (HPLC): 95.9%. 1H NMR (300 MHz, D2O, ppm): 2.44–
2.74 (m, ASCH2CH(OH)A, ASCH2CH2A, 6H), 3.44 (dd, J1
¼
SYNTHESIS OF HYPERBRANCHED POLY(ESTER-AMIDE)s, BAO ET AL.
5365

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
TABLE 1 In Situ Polymerizations of Intermediate 3b Derived from D,L-Alaninea
Tb Timec
(ꢁC) (h)
Mn
(kg/mol) (kg/mol) PDId (%)
Mw
Yield
Tg
[L1]/[L2]h (ꢁC)
Td10%j
(ꢁC)
T0
d
d
d
i
10%k
DPe DB1
DB2
(ꢁC)
f
g
No. Cat.
Pb1
Pb2
Pb3
Pb4
Pb5
Pb6
Pb7
a
Ti(OBu)4
155
160
165
170
5
15.9
30.2
38.8
18.0
89.8
96.8
1.13
2.97
2.49
83.4
85.5
88.3
19
45
60
0.40
0.24
0.43
0.54
1:0.23
1:0.39
1:0.46
ꢀ11.52 282.6
1.33 281.4
5.35 286.7
223.1
254.2
246.4
Ti(OBu)4
Ti(OBu)4
Ti(OBu)4
9
0.48
0.58
5
0.5
Gel
SnO(Bu)2 165
11
30.3
30.2
18.9
75.3
51.2
42.6
2.48
1.69
2.20
87.8
86.5
73.0
h
52
55
23
0.43
0.57
0.46
0.39
0.52
0.41
1:0.24
1:0.45
1:0.33
ꢀ25.52 243.2
4.75 271.8
230.2
260.9
255.7
Sb2O3
165
165
9
Zn(OAc)2
24
ꢀ0.29 272.6
Amount of catalyst was 0.5 g per mole of ABB0 intermediate and the
initial polymerization temperatures were 60 ꢁC for 1 h, 120 ꢁC for 2 h.
The ratios of L1 units to L2 units ([L1]/[L2]) in polymers were calculated
by quantitative 13C NMR.
b
i
The temperature in the last stage of polymerization.
The time of the last stage of polymerization.
The glass transition temperatures were determined by DSC measure-
c
ments with a heating and cooling rate of 10 ꢁC/min in nitrogen ꢀ50 to
d
The molecular weights and PDI values were measured by SEC cali-
100 ꢁC.
j
brated against narrow-dispersity polystyrene standards.
The temperatures of 10% mass loss in nitrogen were determined by
e
DP values determined by quantitative 13C NMR.
TGA measurements with a heating rate of 10 ꢁC/min from 40.0 to 900.0
ꢁC.
f
DB1 were determined by formula 1 on the basis of quantitative 13C NMR.
g
k
DB2 were determined by formula 2 on the basis of quantitative 13C
The temperatures of 10% mass loss in air were determined by TGA
NMR.
measurements with a heating rate of 10 ꢁC/min from 40.0 to 900.0 ꢁC.
6.3, J2 ¼ 11.4 Hz, ACHH(OH)A, 1H), 3.53 (dd, J1 ¼ 3.9 Hz, J2
¼ 11.4 Hz, ACHH(OH)A, 1H), 3.61 (s, AOCH3, 3H), 3.70 (m,
ASCH2CH(OH)A, 1H), 3.87 (s, ANHCH2A, 2H). 13C NMR
(100 MHz, D2O, ppm): 30.1, 37.3, 38.0, 43.2, 54.7, 66.7, 73.4,
173.3, 175.3. ESI-MS (m/z): calcd. for C9H17NO5S, 251.1;
found, 234.0 ([M ꢀ H2O þ H]þ); 252.0 ([M þ H]þ). Anal.
Calcd. for C9H17NO5S: C, 43.01; H, 6.77; N, 5.58; S, 12.74.
Found: C, 42.89; H, 6.73; N, 5.58; S, 12.79.
In Situ Polymerization
The typical polymerization of intermediate 3b (Pb3 in Ta-
ble 1) is as follows. The preparation of intermediate 3b was
presented above. The CH2Cl2 and Et3N was removed com-
pletely under vacuum and without further purification and
isolation, 0.005 g of Ti(OBu)4 was added into the flask con-
taining the intermediate (ca. 10 mmol). Under vigorous me-
chanical stirringꢁ and reduced pressure, the reaction mixture
ꢁ
ꢁ
was kept at 60 C for 1 h, 120 C for 2 h, then 165 C until
the viscosity of the system increased suddenly. Finally, the
crude product was dissolved in DMF, and then poured into
200 mL of acetone. The precipitate was collected and puri-
fied by reprecipitation from DMF solution by adding excess
acetone thrice. The catalyst was removed from the polymers
via reprecipitation process. The resultant polymer was dried
at 40 ꢁC under vacuum for 24 h. IR (KBr, cmꢀ1): 3408
(OAH), 1739 (OC¼¼O), 1652 (NC¼¼O), 1545 (HANCO). 13C
NMR (100 MHz, DMSO-d6, ppm): 15.4 (CH3CHA), 26.2
(ASCH2CH2CONHA in four different units), 29.3–29.9 (ACH2
CONHA in four different units), 33.2–33.9 (ACH(OH)CH2SA
in four different units), 46.0 (CH3CHA), 51.0 (CH3OA), 59.7,
62.2, 62.9, 65.4 (ACH2OH in four different units), 66.4, 69.7,
72.4, 73.0 (ACH(OH)A in four different units), 168.9–171.5
(ACONH and ACOOA).
3c
Purity (HPLC): 95.5%. 1H NMR (300 MHz, D2O, ppm): 0.86
(t, J ¼ 6.9 Hz, ACH2CH3, 3H), 1.58–1.68 (m, ACHHCH3 1H),
1.72–1.81 (m, ACHHCH3, 1H), 2.48–2.77 (m, ASCH2
CH(OH)A, ASCH2CH2A, 6H), 3.48 (dd, J1 ¼ 6.3 Hz, J2 ¼ 11.7
Hz, ACHH(OH)A, 1H), 3.57 (dd, J1 ¼ 3.3 Hz, J2 ¼ 11.7 Hz,
ACHH(OH)A, 1H), 3.65 (s, AOCH3, 3H), 3.71–3.77 (m,
ASCH2CH(OH)A, 1H), 4.19–4.23 (m, ACHCH2CH3, 1H). 13C
NMR (100 MHz, D2O, ppm): 9.6, 24.1, 27.7, 34.5, 35.4, 52.8,
54.4, 64.3, 70.7, 173.6, 174.4, ESI-MS (m/z): calcd. for
C
11H21NO5S, 279.1; found, 204.0 ([Fragment þ H]þ); 261.6
([M ꢀ H2O þ H]þ); 279.9 ([M þ H]þ). Anal. Calcd. for
C11H21NO5S: C, 47.29; H, 7.52; N, 5.02; S, 11.46. Found: C,
47.16; H, 7.56; N, 5.04; S, 11.41.
3d
Pa
Purity (HPLC): 96.4%. 1H NMR (300 MHz, D2O, ppm): 2.41–
2.76 (m, SCH2CH(OH), SCH2CH2, 6H), 3.03 (m, ACHHC6H5,
1H), 3.12 (m, ACHHC6H5, 1H), 3.48–3.74 (m, ASCH2
CH(OH)A, AOCH3, ACH2(OH)A, 6H), 4.76–4.83 (m, ACHCH2
C6H5, 1H), 7.06–7.25 (m, AC6H5, 5H). 13C NMR (100 MHz,
D2O, ppm): 28.3, 35.7, 36.7, 38.1, 52.7, 53.8, 65.6, 71.5,
127.4, 128.9, 129.6, 136.3, 171.8, 172.6, ESI-MS (m/z): calcd.
for C16H23NO5S, 341.1; found, 266.0 ([Fragment þ H]þ);
323.7 ([M ꢀ H2O þ H]þ); 341.9 ([M þ H]þ). Anal. Calcd. for
The polymerization procedure was similar to that of Pb. 13C
NMR (100 MHz, DMSO-d6, ppm): 26.5–27.6 (ASCH2CH2
CONHA in four different units), 31.0–31.2 (ACH2CONHA in
four different units), 34.6–35.4 (ACH(OH)CH2SA in four dif-
ferent units), 51.6 (CH3OA), 61.0, 63.6, 64.4, 67.0 (ACH2OH
in four different units), 67.9, 70.8, 71.2, 73.6 (ACH(OH)A in
four different units), 171.0–169.3 (ACONH and ACOOA).
Pc
C16H23NO5S: C, 56.28; H, 6.74; N, 4.10; S, 9.38. Found: C,
The polymerization procedure was similar to that of Pb. 13C
NMR (100 MHz, DMSO-d6, ppm): 10.3 (CH3CH2A), 24.4
56.16; H, 6.74; N, 4.15; S, 9.39.
5366
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
SCHEME 1 Possible reaction between AD monomer and diethanolamine CB2 monomer.
(CH3CH2A), 27.6–27.9 (ASCH2CH2CONHA in four different
units), 31.4–31.7 (ACH2CONHA in four different units),
34.9–35.6 (ACH(OH)CH2SA in four different units), 51.8
(CH3OA), 53.4 (CH3CH2CHA), 61.3, 63.8, 64.6, 67.0 (ACH2OH
in four different units), 68.1, 71.4, 74.0, 74.4 (ACH(OH)A in
four different units), 171.0–172.6 (ACONH and ACOOA).
hydroxyl toward the unsaturated double bond of acrylamide
derivative.36 However, detailed experiments (see Supporting
Information Figs. S1 and S2) showed that mixtures of compli-
cated products were obtained instead of dominant formation
of the desired AB2 intermediate under various reaction condi-
tions, because of existence of competitive ester aminolysis
side reaction producing the DB2 species (Scheme 1), which
happened under the similar conditions to the prominent reac-
tion of aza-Michael addition yielding the AB2 intermediate.
Furthermore, the inefficient reactivity of the nitrogen-donor
may be another reason for failure. Considering that (i) mer-
capto group is more nucleophilic than amino group and there-
fore more reactive, followed by hydroxyl (SH > NH (original)
> NH2 > NH (formed) > OH);37 (ii) mercapto group will not
react with ester group under mild conditions, accordingly the
side reaction occurred at the ester group can be suppressed.
Therefore, the desired AB2 intermediate can be afforded
almost exclusively at the first stage. The reactions of sulfur
containing compounds with alkenes were termed as thiol-ene
chemistry, which are used by most authors today to describe
the reactions of thiols with a wide variety of unsaturated
functional groups, in addition to unactivated CAC double
bond. The thiol-ene reaction can take place either by classical
radical addition mechanism or Michael-type nucleophilic
addition, even by a mixed mechanism. The merits, mostly
resulting from the high efficiency and orthogonality of the
reaction, mild reaction conditions and the compatibility of the
process with sensitive functional groups and biological proc-
esses, have led to thiol-ene chemistry being used increasingly
in polymer functionalization and macromolecular synthesis,
as well as more traditional applications ranging from cross-
linked networks to functionalized biomaterials.38–42 In fact,
many of the design features of click chemistry as proposed by
Sharpless and coworkers,43 which is widely applicable in the
syntheses of dendritic and star polymers,31,44 are also pre-
sented with the thiol-ene reaction.
Pd
The polymerization procedure was similar to that of Pb. 13C
NMR (100 MHz, DMSO-d6, ppm): 27.3–27.7 (ASCH2CH2
CONHA in four different units), 30.7–31.4 (ACH2CONHA in
four different units), 34.8–35.6 (ACH(OH)CH2SA in four dif-
ferent units), 36.8 (PhACH2A), 51.8 (CH3OA), 53.5 (PhCH2
CHA), 61.1, 64.0, 64.5, 67.1 (ACH2OH in four different
units), 67.9, 71.3, 74.0, 74.3 (ACH(OH)A in four different
units), 126.5, 128.2, 129.0, 137.1 (Ph), 170.5–172.0 (ACONH
and ACOOA).
RESULTS AND DISCUSSION
Molecular Design of the Approach to HBPEAs
Based on Amino Acids
On the basis of the principle of ‘‘AD þ CB2’’ methodology, an
AB2-type intermediate will be generated due to reactivity
differences between different functional groups in selected
monomer pair, further in situ reactions among the AB2 species
result in hyperbranched macromolecules. In our study, the
AB2 intermediate was expected to be created through Michael
addition reaction, which is benefited from high conversion,
low probability of side reactions, mild reaction conditions,
and favorable reaction rate.35 The vinyl monomer AD contain-
ing one activated C¼¼C double bond and one ester group,
which was prepared from amino acid and acryloyl chloride,
performed as the Michael acceptor. In the preliminary optimi-
zation of the Michael donor, namely, the CB2 monomer, widely
investigated dienthanolamine and diisopropanolamine were
chosen due to higher reactivity of the amino group than the
SYNTHESIS OF HYPERBRANCHED POLY(ESTER-AMIDE)s, BAO ET AL.
5367

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 2 ‘‘AD þ CBB0’’ strategy
for novel hyperbranched poly
(ester-amide)s based on neutral
a-amino acids.
Consequently, a commercially available monomer 1-thiolgly-
cerol was finally chosen to serve as the CB2 monomer
(actually a CBB0 monomer because of different reactivities of
the primary and the secondary hydroxyl groups) in our sys-
tem. Scheme 2 shows the design idea. The vinyl monomer
AD, an easily prepared amino acid-based compound bearing
one ester group and one activated C¼¼C double bond, with a
commercially available CBB0 monomer 1-thioglycrol, thiol
with one primary and one secondary hydroxyls, creates a
suitable monomer pair. The amide-derivative is dominantly
formed as an ABB0 intermediate with one ester and two dif-
ferent hydroxyl groups via the thio-Michael addition. Without
further purification and isolation, the intermediate is sub-
jected to transesterification polymerization in the presence
of a catalyst to produce HBPEAs bearing numerous hydroxyl
end groups.
The formation of the intermediate 3b was evidenced by 1H
NMR at the initial stage of the reaction. Figure 1 displays the
evolution of 1H NMR spectra for the reaction mixture of
compound 2b and equimolar 1-thiolglycerol from 10 min to
24 h at ambient temperature in the presence of catalytic
amount of Et3N. The intermediate 3b was formed as soon as
the mixing of the two reactants and catalyst. It is clear that
the peaks at 5.5–6.5 ppm ascribed to the protons of the dou-
ble bond in compound 2b gradually decreased with the pro-
longing of the reaction time. The new peaks emerged at
higher field of 2.4–2.7 ppm which are partly overlapping, are
Synthesis of ABB0 Intermediate Originated
from D,L-Alanine
The use of weak base catalysts, such as Et3N, is sufficient to
catalyze the thio-Michael addition process due to readily ac-
cessible pKa of thiol (9.6645 for 1-thiolglycerol vs. 14.4046 for
glycerol).35(a),47 Reaction of thiol with Et3N results in depro-
tonation of the thiol to the corresponding thiolate anion and
formation of the triethylammonium cation.47,48 The thiolate,
a powerful nucleophile, adds to the activated C¼¼C bond at
the electrophilic b-carbon forming a carbon-centered anion
(or enolate) intermediate, which is a very strong base. Then,
this anion abstracts a proton (either from the thiol or from
the ammonium cation) producing the thio-Michael addition
product following regioselective anti-Markovnikov’s rule.48 In
this process, a relatively weak base Et3N is used to generate
a much stronger base (the carbanion or enolate) in the cata-
lytic cycle, yielding the expected results.
FIGURE 1 Evolution of the 1H NMR spectra for the reaction sys-
tem between 2b and equimolar 1-thioglycerol over time.
5368
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
because of ether formation and cyclic side reactions, while
HBPEA with high molecular weight cannot be obtained at
150 ꢁC even after 30 h in the presence of Ti(OBu)4. In the
ꢁ
region of 155–165 C, the soluble polymers with high molec-
ular weights were obtained successfully. The optimum poly-
ꢁ
merization temperature for intermediate 3b is 165 C.
The effect of catalysts on the polymerization was also inves-
tigated. The polymerization catalyzed by Ti(OBu)4 (Pb3) was
faster, and higher molecular weight polymer was more easily
obtained than the other catalysts used, such as SnO(Bu)2
(Pb5), Sb2O3 (Pb6), and Zn(OAc)2 (Pb7). Part of the polymers
was carbonized or oxidized (evidenced by dark color of the
resultant polymers) when the later three catalysts were used
due to a long time staying at high temperature to obtain
polymers with high molecular weights. Therefore, Ti(OBu)4
is the most efficient catalyst among several catalysts used
here for the transesterification polymerization of such an
intermediate.
FIGURE 2 13C NMR spectra of 2b in DMSO and intermediate 3b
in D2O (*: Peaks originated from the protons of Et3N).
attributed to the protons of the two methylenes formed cor-
respondingly, indicating that the Michael addition was pro-
moted at the b-carbon but not the a-carbon. The desired
reaction was fast within the initial 1 h, and about 90% of
the monomers reacted with each other. Reaction between
the remaining monomers needed another 3 h, and almost no
changes could be observed between the spectra recorded at
4 h and that at 24 h, which indicated that the reaction fin-
ished in 4 h, obtaining the desired molecule near quantita-
tively. Moreover, in this process, Michael addition of the
hydroxyls to the double bond would not occur under the
same conditions (see Supporting Information Fig. S3). The
formation of the desired molecule was further verified by
13C NMR and ESI-MS analyses of the product. As expected,
13C NMR spectra (Fig. 2) proved the successful formation of
the desired amide-derived intermediate. Obviously, the peaks
originated from the carbons of C¼¼C at 131.8 and 126.7 ppm
exhibit a full high-yield shift to 69.1 and 62.7 ppm. ESI-MS
analyses (Fig. 3) also confirmed that the Michael addition
reaction was carried out in a high yield because no peak at-
tributable to any byproducts or residuals is observed.
Characterization of HBPEAs Based on D,L-Alanine
The soluble hyperbranched polymers derived from D,L-ala-
nine were obtained under the optimal conditions, and the
structure of the corresponding polymers was characterized
by FTIR and NMR techniques. The FTIR spectra (see Sup-
porting Information Fig. S4) provide evidence for the chemi-
cal structure of the polymer (Pb3), showing that the charac-
teristic absorptions of ester carbonyl and amide carbonyl
groups are at 1739 and 1652 cmꢀ1, respectively, and NAH
bending mode for amide group at 1545 cmꢀ1. When com-
pared with that of the intermediate, the FTIR spectra of the
polymer broaden due to the complicated repeat units and
branched architecture, which are the characteristic features
of HBP.29
Generally, there are three different units, branched units (B),
terminal units (T), and linear units (L) in HBP prepared
from a symmetric AB2 monomer.49 However, the intermedi-
ate formed here is actually an asymmetric AB2 monomer,
namely an ABB0 monomer, due to higher reactivity of the pri-
mary hydroxyl than the secondary hydroxyl group. Therefore,
four different units exist in the resultant polymer, that is,
In Situ Self-Polycondensation to Prepare
HBPEAs Derived from D,L-Alanine
The ABB0 intermediate obtained at the first stage of the reac-
tion was directly subjected to self-polycondensation to pre-
pare HBPEA after removal of the solvent. It is extremely im-
portant to search suitable polymerization conditions to
obtain soluble HBPEA with high molecular weight. The typi-
cal results are summarized in Table 1. High vacuum was
applied during the whole polymerization process till the vis-
cosity of the polymer increased suddenly. The methanol gen-
erated was removed from the system by vacuum to break
the balance to reach a high conversion of the intermediate.
The results show that the temperature in the last stage of
the polymerization significantly influences the molecular
weights of the resultant polymers. The data in Table 1 indi-
cate that the molecular weight rapidly increased with the
increase of reaction temperature. However, crosslinking was
observed at the temperature above 170 ꢁC (Pb4) probably
FIGURE 3 ESI mass spectrum of intermediate 3b.
SYNTHESIS OF HYPERBRANCHED POLY(ESTER-AMIDE)s, BAO ET AL.
5369

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
confirm different units in the polymer. Obviously, the positive
resonances for methyl and methine carbons are readily dis-
tinguished from the negative resonances of methylene car-
bons, whereas the quaternary carbons are not found in
DEPT-135 spectra. By comparing 13C NMR with DEPT-135
spectra of the polymer, the four negative peaks at 65.4–59.7
ppm are correspondingly assigned to the carbon j (Cj in Fig.
5) of the four different units. Commonly, the exact assign-
ments of these peaks should be based on low molecular
weight model compounds, which possess structure respec-
tively similar to the branched, terminal, and linear units in
the hyperbranched polymers, to determine accurate chemical
shifts of B, T, L1, and L2. However, the attempts to synthesize
the corresponding low molecular weight compounds failed.
Fortunately, Ohta and Hikino have reported that the a-carbon
of the hydroxyl group will show a downfield shift, whereas
b-carbon show a upfield shift when the hydroxyl is con-
verted to acetyl functional group.50 They further confirmed
this issue in the determination of lyoniatoxin structure.51
This theory can also be applied to our study.
FIGURE 4 1H NMR spectra of intermediate 3b and polymer Pb3
in D2O. The polymerization was conducted in situ at 165 ꢁC for
5 h with Ti(OBu)4 as the catalyst (*: Peaks originated from the
protons of Et3N).
branched units (B), terminal units (T), linear units 1 (L1,
only the primary hydroxyl group reacts), and linear units 2
(L2, only the secondary hydroxyl group reacts). The complex
structure can be confirmed by NMR spectra. The 1H NMR
spectra (Fig. 4) of the polymer resemble that of its monomer
as their structure is alike except that the peaks in the spec-
tra of the polymer broaden and a few new peaks in the
range of 4.0–4.5 ppm are found, which indicate the complex
microenvironment resulted from various architectures in the
polymer. However, it is difficult to precisely assign the peaks
in 1H NMR spectra of the polymer because of serious over-
lapping of different protons. Thus, 13C NMR analysis along
with distortionless enhancement by polarization transfer 135
(DEPT-135) experiment (Fig. 5) was performed to further
First, given that the chemical shift of Cj in terminal unit is
dT,j, when the primary hydroxyl group is acylated to obtain
linear unit 1 and the secondary hydroxyl group to linear unit
2, respectively, Cj will show downfield shift to dL (dL
>
1;j
1;j
dT,j) and upfield shift to dL (dL < dT,j) according to the
2;j
2;j
theory mentioned above. However, it is difficult to assign Cj
in branched unit (dB,j) on the basis of this theory because of
conflicting influences of the primary and secondary hydroxyl
acylation on the chemical shift of Cj. Fortunately, the chemi-
cal shift of Cj in terminal unit should be very similar to that
of the corresponding carbon in the intermediate due to their
similar structure. On the other hand, typically for hyper-
branched polymers, the signals of terminal units are
FIGURE 5 13C NMR spectra of Pb3
in DMSO: Pb3 (A); DEPT-135 spec-
trum of Pb3 (B) where the CH and
CH3 resonances are above the
baseline and the CH2 resonances
are directed below the baseline.
5370
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 2 In Situ Polymerizations of Intermediates 3a, 3c, and 3d Originated from Glycine, D,L-2-Aminobutyric Acid
and D,L-Phenylalanine, Respectivelya
Tb
(ꢁC)
Timec
(h)
Mn
Mw
Yield
(%)
Tg
(ꢁC)
Td10%j
(ꢁC)
T0
d
d
i
10%k
d
(ꢁC)
PDId
DPe
DB1
DB2
[L1]/[L2]h
f
g
No.
(kg/mol)
(kg/mol)
Pb3
Pa1
Pa2
Pc1
Pc2
Pd1
Pd2
Pd3
a
165
160
165
165
175
165
175
180
5
7.5
2
38.8
20.2
17.1
40.7
40.0
84.8
53.7
117.3
96.8
53.0
2.49
2.60
3.24
1.98
2.42
7.65
4.68
2.23
88.3
85.4
88.9
79.4
80.7
85.7
89.5
83.4
60
43
0.58
0.56
0.54
0.53
1:0.46
1:0.53
u.d.
5.35
286.7
275.9
246.4
247.2
25.21
55.4
19
8.5
26
15
12
80.8
74
75
0.53
0.41
0.50
0.33
1:0.45
u.d.
16.21
243.4
289.1
231.4
265.6
97.0
649.5
252.0
263.0
u.d.
1:0.21
u.d.
36.49
h
The catalyst of the polymerization was Ti(OBu)4 with amount of 0.5 g
per mole of ABB0 intermediate and the initial polymerization tempera-
tures were 60 ꢁC for 1 h, and then 120 ꢁC for 2 h.
The ratios of L1 units to L2 units ([L1]/[L2]) in polymers were calculated
by quantitative 13C NMR.
i
The glass transition temperatures were determined by DSC measure-
b
The temperature in the last stage of polymerization.
ments with a heating and cooling rate of 10 ꢁC/min in nitrogen ꢀ50 to
c
The time of the last stage of polymerization.
100 ꢁC.
j
d
The molecular weight was measured by SEC calibrated against nar-
The temperatures of 10% mass loss in nitrogen were determined
row-dispersity polystyrene standards.
by TGA measurements with a heating rate of 10 ꢁC/min from 40.0 to
900.0 ꢁC.
e
DP values determined by quantitative 13C NMR.
f
k
DB1 were determined by formula 1 on the basis of quantitative 13C
The temperatures of 10% mass loss in air were determined by TGA
measurements with a heating rate of 10 ꢁC/min from 40.0 to 900.0 ꢁC.
NMR.
g
DB2 were determined by formula 2 on the basis of quantitative 13C
NMR.
narrower than those of branched units because of their
higher mobility. As a result, the peak at 62.9 ppm in the 13C
NMR spectra of the polymer should be assigned to Cj in ter-
minal unit (T) by comparison with that of the intermediate
at 62.7 ppm. Moreover, the peaks at 59.7, 62.2, and 65.4
ppm will be assigned to Cj in L2, B, and L1, respectively. Simi-
larly, the peaks at 73.0, 72.4, 69.7, and 66.4 ppm originate
from Ci of the four subunits L2, B, T, and L1, respectively.
molecular weights and then slowly increases with increasing
molecular weight, until it finally reaches its limiting value.
Subsequently, the DB2 increases slightly and also approaches
the same plateau value as DB1 at higher molecular weights.54
In our study, because of severe overlapping of the polymer
protons in the 1H NMR spectra, DB values were calculated
on the basis of quantitative 13C NMR using the two formulas
mentioned above, respectively. The DB values of the poly-
mers prepared from intermediate 3b were determined to be
0.43–0.58 (by Formula 1) and 0.39–0.54 (by Formula 2)
excepꢁt the lower one produced at relatively low temperature
(155 C) in the presence of Ti(OBu)4 (Pb1 in Table 1). It indi-
cated that the molecular weights of the resultant polymers
were not large enough to reach the plateau of the degree of
branching referring to the results of lower values calculated
by Formula 2 than Formula 1. In addition, the Tables 1 and
2 give the content ratios of linear unit 1 to linear unit 2
([L1]/[L2]) in the polymers calculated by quantitative 13C
NMR. Much more contents of L1 units in the polymers sup-
port the fact that the primary hydroxyl group exhibits
greater reactivity than that of the secondary hydroxyl group.
The [L1]/[L2] decreases with increasing polymerization tem-
perature under the same catalyst (Pb1–Pb3 in Table 1), and
the lower is the [L1]/[L2], the higher is the DB value.
The clear assignments of the four different units in the NMR
spectra of the polymer are important for the determination
of degree of branching (DB),52 which is one of the most im-
portant molecular parameters of HBP, characterizing the dif-
ference in molecular structure from their linear analogue.
The DBs are calculated on the basis of the terminal (T),
branched (B), and linear units (L). The DB according to the
definition of Hawker and Fre´chet is given by52
:
B þ T
DB1 ¼
(1)
B þ T þ L
whereas a more general definition for the DB in AB2 system
is given by53
:
2B
DB2 ¼
(2)
2B þ L
Because of the existence of numerous hydroxyl and amide
groups, the HBPEAs obtained are highly soluble in extremely
polar solvents such as methanol, DMF, and DMSO,55 as show
poor solubility in THF and CHCl3. The molecular weights
(Mws) and the polydispersity indices (PDIs) of the HBPEAs
were determined by SEC equipped with refractive index de-
tector using narrow-dispersity polystyrene as standards with
DMF as solvent and eluent. However, this method has only
DB1 and DB2 differ considerably at low molecular weights.
The DB1 possesses a value of 1 at low molecular weights
and decreases with increasing molecular weight, and finally
reaches a minimum value, which is the most likely explana-
tion for DB values exceeding the predicted value of 0.5 for a
random AB2 polymerization in some literatures according to
formula 1. However, the DB2 assumes a value of 0 for low
SYNTHESIS OF HYPERBRANCHED POLY(ESTER-AMIDE)s, BAO ET AL.
5371

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
limited suitability for HBPs. On one hand, underestimation
will be caused because of their smaller hydrodynamic radii
compared with their linear analogues with the same molar
mass.56 On the other hand, however, the strong interactions
between the numerous polar end groups of the polymer and
the solvent make the measurement results overestimate the
true values. The measured data listed in Table 1 can only be
used for relative comparisons. The Mn of the resultant
HBPEAs ranges from 15.9 to 38.8 kg/mol and shows a
strong dependence on polymerization conditions, such as dif-
ferent temperatures and catalysts. The PDIs estimated from
SEC measurement are in the region of 1.13–2.97, which are
smaller than prediction due to removal of the polymers with
low molecular weights by the multistep purification process.
The similar results have been reported by previous literatur-
es.18(a),32,36,57 Actually, the molecular weights of the poly-
mers can be estimated on the basis of quantitative 13C NMR
by comparing intensity of signal (Ia) for the methylester focal
group at ꢃ51 ppm with that (Ic) of the CH signals (a-C of
the amino acid residue) at ꢃ46 ppm. According to the
assignment, the DP values of the polymers were calculated
by dividing the average integration values of peak c by
peak a:
same results were obtained as those based on D,L-alanine.
The predicted ABB0 intermediates were firstly formed domi-
nantly under the same mild conditions, and then in situ self-
polycondensations were carried out at higher temperatures
in the presence of Ti(OBu)4. The polymerization conditions
and results were illustrated in Table 2.
When compared with the polymerization of intermediate 3b
(Pb3 in Table 2), crosslinking occurred more easily for inter-
mediate 3a, indicating a faster polymerization rate for the
later, possibly because of its smaller steric hindrance. For in-
termediate 3a, lower polymerization temperature should be
adopted (Pa1). In contrast, the molecular weights of the poly-
mers Pc and Pd prepared from the intermediates 3c and 3d,
respectively, at the same conditions to Pb3, were too low to
be measured by SEC due to larger steric effects of ethylene
and benzyl groups in the intermediates. For the intermediate
3c, the polymers with molecular weight of 40.7 (Pc1) and
40.0 kg/mol (Pc2) were obtained, respectively, when the
ꢁ
reaction time was prolonged to 19 h at 165 C and the poly-
merization temperature was raised to 175 ꢁC for 8.5 h. The
steric effect on the polymerization was more obvious for in-
termediate 3d possessing larger substituent. Polymer with
molecular weight of 84.8 kg/mol was obtained when the
reaction was proceeded for 26 h at 165 ꢁC (Pd1), and even
117.3 kg/mol when the temperature was increased to
180 ꢁC for 12 h (Pd3). However, temperatures higher than
180 ꢁC should not be considered because serious side reac-
tions and carbonization of the products occurred undesir-
ably. The DP values evaluated by quantitative 13C NMR spec-
tra are showed in Table 2, indicating that the data
determined by SEC measurements seriously deviate from
that by quantitative 13C NMR spectra which may be more
credible. The DB values of the polymers based on glycine,
D,L-2-aminobutyric acid, and D,L-phenylalanine were deter-
mined to be 0.56, 0.53, and 0.41 (Formula 1) and 0.53, 0.50,
and 0.33 (Formula 2) on the basis of quantitative 13C NMR,
respectively. The PDIs of the polymers, in accordance with
those of HBPEAs based on D,L-alanine, were close to 2.0
because of repeating purification process.
Ic
DP ¼ þ 1
(3)
Ia
It should be noted that the calculation is based on the ab-
sence of intramolecular cyclization, thus each molecule bears
a single unreacted A focal group.58 However, slight intramo-
lecular cyclization to ether formation may be unavoidable
under such a high-polymerization temperature. The number
of focal point groups, relative to repeat units, would be
lower due to intramolecular cyclization, resulting in overesti-
mation of the molecular weight evaluated with formula 3.59
The DP values determined by quantitative 13C NMR are illus-
trated in Table 1, which indicate that the Mn values of the
polymers based on quantitative 13C NMR spectra were much
lower than those determined by SEC. That is to say the over-
estimation effect of SEC measurements calibrated by polysty-
rene is probably more serious than underestimation for
hyperbranched poly(ester-amide)s with numerous hydroxyl
end groups.
The thermal properties of the hyperbranched polymers
obtained are also listed in Tables 1 and 2. The Tg of Pb3 cata-
lyzed by Ti(OBu)4 is the highest in comparison with that of
the polymers catalyzed by other three catalyst (Pb5–Pb7).
This can be ascribed to the difference of their end group
densities as a result of the different DBs of the polymers.
The DB of Pb3 is higher than those of Pb5–Pb7, which pro-
duces greater terminal group density and result in stronger
intermolecular interaction of the H-bond. The results suggest
that Tg slightly increases with the increase of DB for the
hyperbranched poly(ester-amide)s with hydroxyl end groups
although hardly influenced by the molecular weight of the
polymer.60 On the other hand, the higher Tg of the polymer
derived from D,L-phenylalanine comparing those of polymers
based on glycine, D,L-alanine, and D,L-2-aminobutyric acid
may be attributed to hindered segmental rotation and
greater rigidity as a result of the existence of the phenyl
group, whereas the latter three contain more flexible alkyl
Synthesis and Characterization of HBPEAs
Based on Other Neutral a-Amino Acids
The previous results show that the ‘‘AD þ CBB0’’ technique
was successfully used to the monomer pair composed of 1-
thioglycerol and D,L-alanine-based monomer to produce high
level molecular weight HBPEAs with Ti(OBu)4 as the efficient
polymerization catalyst. Therefore, we tried extending this
strategy to other neutral a-amino acids with different sub-
stituents at the a-position, such as glycine, D,L-2-aminobutyric
acid, and D,L-phenylalanine, to investigate the steric effect of
the intermediate on the polymerization and polymer proper-
1
ties. The H and 13C NMR, ESI-MS, HPLC, and elemental anal-
yses were carried out at the first stage to ensure the forma-
tion of expected ABB0 intermediates with good yields. The
5372
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
backbone. The temperatures of 10% mass loss (Td10%) of the
HBPEAs measured in air medium are commonly lower than
those in nitrogen medium (see Supporting Information Fig.
S5). This is possibly because a complicated set of chemical
transformations was caused by exposure to high tempera-
tures, which are a function of the environment. Thermooxi-
dative processes take place in oxidizing medium (in air),
thermal degradation predominant in neutral medium (in
nitrogen or a vacuum), whereas pyrolysis takes place as high
temperatures in neutral medium with formation of coke resi-
due.61 Among the various polymers, those based on D,L-phe-
nylalanine display the highest thermal stability due to special
structure. The Td10%s are 289.1 ꢁC in nitrogen and 265.6 ꢁC
2 van Dijk, M.; Nollet, M. L.; Weijers, P.; Dechesne, A. C.; van
Nostrum, C. F.; Hennink, W. E.; Rijkers, D. T. S.; Liskamp, R. M.
J. Biomacromolecules 2008, 9, 2834–2843.
3 Fan, X. H.; Zhou, J. L.; Chen, X. F.; Wan, X. H.; Zhou, Q. F.
Chin J Polym Sci 2004, 22, 497–500.
4 Montane, J.; Armelin, E.; Asin, L.; Rodriguez-Galan, A.; Puig-
gali, J. J Appl Polym Sci 2002, 85, 1815–1824.
5 Okada, M.; Yamada, M.; Yokoe, M.; Aoi, K. J Appl Polym Sci
2001, 81, 2721–2734.
6 Asin, L.; Armelin, E.; Montane, J.; Rodriguez-Galan, A.; Puig-
gali, J. J Polym Sci Part A: Polym Chem 2001, 39, 4283–4293.
7 Armelin, E.; Paracuellos, N.; Rodriguez-Galan, A.; Puiggali, J.
in air for Pd2
.
Polymer 2001, 42, 7923–7932.
8 Andrianov, A. K.; Marin, A. Biomacromolecules 2006, 7,
CONCLUSIONS
1581–1586.
The ‘‘AD þ CBB0’’ methodology have been successfully devel-
oped to prepare amino acid-based hyperbranched poly(ester-
amide)s with numerous hydroxyl end groups in this study.
Benefiting from merits of Michael addition reaction and
higher nucleophilicity of mercapto than hydroxyl group, ABB0
intermediates were dominantly formed by thio-Michael addi-
tion of commercially available monomer 1-thioglycerol to the
compounds prepared from neutral a-amino acids, which
were demonstrated by 1H and 13C NMR, ESI-MS, HPLC, and
elemental analyses. In situ self-condensations of these inter-
mediates started at elevated temperature in the presence of
a catalyst to produce HBPEAs with molecular weights range
of 11.8–117.7 kg/mol with broad distribution determined by
SEC measurements in contrast to 5.0–25.6 kg/mol estimated
on the basis of quantitative 13C NMR spectra. The influence
factors of polymerization were investigated. Ti(OBu)4 was
found to be the most efficient catalyst compared with
SnO(Bu)2, Sb2O3, and Zn(OAc)2. The propositional polymer-
ization temperature was in the range of 160–180 ꢁC. Struc-
ture of the intermediates significantly affected the molecular
weights and properties of the corresponding HBPEAs, which
originated from different steric hindrances on the a-carbon
of the amino acids. For the HBPEAs obtained, the Td10%s are
above 220 ꢁC, showing lower values in air medium than
those in nitrogen, and glass trꢁansition temperatures (Tg) are
in the range of ꢀ25.5 to 36.5 C, depending on the structure
of the monomers and slightly influenced by the DBs of the
polymers bearing hydroxyl end groups. Among them, those
derived from D,L-phenylalanine display the lowest degree of
branching, the highest glass transition temperature and the
best thermal stability. The large number of peripheral
hydroxyl groups of these potential biodegradable and biode-
gradable HBPEAs will expand their range of applications.
9 Singh, A.; Krogman, N. R.; Sethuraman, S.; Nair, L. S.; Stur-
geon, J. L.; Brown, P. W.; Laurencin, C. T.; Allcock, H. R. Bio-
macromolecules 2006, 7, 914–918.
10 Chen, R.; Khormaee, S.; Eccleston, M. E.; Slater, N. K. H.
Biomaterials 2009, 30, 1954–1961.
11 Chen, R. J.; Eccleston, M. E.; Yue, Z. L.; Slater, N. K. H.
J Mater Chem 2009, 19, 4217–4224.
12 Qu, J. Q.; Sanda, F.; Masuda, T. Eur Polym J 2009, 45,
448–454.
13 Abraham, S.; Ha, C. S.; Kim, I. J Polym Sci Part A: Polym
Chem 2006, 44, 2774–2783.
14 Bag, D. S.; Dutta, D.; Shami, T. C.; Rao, K. U. B. J Polym Sci
Part A: Polym Chem 2009, 47, 2228–2242.
15 Hopkins, T. E.; Wagener, K. B. Macromolecules 2003, 36,
2206–2214.
16 Saha, A.; Ramakrishnan, S. J Polym Sci Part A: Polym
Chem 2009, 47, 80–91.
17 Khan, A.; Malkoch, M.; Montague, M. F.; Hawker, C. J.
J Polym Sci Part A: Polym Chem 2008, 46, 6238–6254.
18 (a) Kricheldorf, H. R.; Stukenbrock, T. Polymer 1997, 38,
3373–3383; (b) Huber, T.; Voit, B.; Wolf, D. Macromol Symp
1999, 142, 133–143; (c) Yan, D. Y.; Gao, C. Macromolecules
2000, 33, 7693–7699; (d) Gao, C.; Yan, D. Y. Macromolecules
2001, 34, 156–161; (e) Wu, D. C.; Liu, Y.; Jiang, X.; He, C. B.;
Goh, S. H.; Leong, K. W. Biomacromolecules 2006, 7, 1879–
1883.
19 Voit, B. J Polym Sci Part A: Polym Chem 2000, 38, 2505–
2525.
20 Constantinos, M. P.; Dimitris, T.; Zili, S. Mol Pharm 2007, 4,
169–188.
We thank for subsidy by the National Natural Science Founda-
tion of China (No. 20923003).
21 Harada, K. Bull Chem Soc Jpn 1959, 32, 1007–1008.
22 Harada, K.; Fox, S. W. Arch Biochem Biophys 1965, 109,
49–56.
REFERENCES AND NOTES
23 Heinrich, M. R.; Rohlfing, D. L.; Bugna, E. Arch Biochem Bio-
phys 1969, 130, 441–448.
1 Couffin-Hoarau, A. C.; Aubertin, A. M.; Boustta, M.; Schmidt,
S.; Fehrentz, J. A.; Martinez, J.; Vert, M. Biomacromolecules
2009, 10, 865–876.
24 Hennon, G.; Plaquet, R.; Biserte, G. Biochimie 1975, 57,
1395–1396.
SYNTHESIS OF HYPERBRANCHED POLY(ESTER-AMIDE)s, BAO ET AL.
5373

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
25 Fox, S. W.; Suzuki, F. Biosystems 1976, 8, 40–44.
26 Fouche, C. E.; Rohlfing, D. L. Biosystems 1976, 8, 57–65.
27 Rohlfing, D. L. Science 1976, 193, 68–70.
42 Becer, C. R.; Hoogenboom, R.; Schubert, U. S. Angew Chem
Int Ed Engl 2009, 48, 4900–4908.
43 Hawker, C. J.; Fokin, V. V.; Finn, M. G.; Sharpless, K. B. Aust
J Chem 2007, 60, 381–383.
28 Rohlfing, D. L.; McAlhaney, W. W. Biosystems 1976, 8,
44 Altintas, O.; Demirel, A. L.; Hizal, G. J Polym Sci Part A:
139–145.
Polym Chem 2008, 46, 5916–5928.
29 (a) Scholl, M.; Nguyen, T. Q.; Bruchmann, B.; Klok, H. A.
J Polym Sci Part A: Polym Chem 2007, 45, 5494–5508; (b)
Scholl, M.; Nguyen, T. Q.; Bruchmann, B.; Klok, H. A. Macro-
molecules 2007, 40, 5726–5734.
45 Kreevoy, M. M.; Harper, E. T.; Duvall, R. E.; Wilgus, H. S.;
Ditsch, L. T. J Am Chem Soc 1960, 82, 4899–4902.
46 Ballinger, P.; Long, F. A. J Am Chem Soc 1960, 82, 795–
798.
30 Menz, T. L.; Chapman, T. M. Polym Prepr (Am Chem Soc
47 Lowe, A. B. Polym Chem 2010, 1, 17–36.
Div Polym Chem) 2003, 44, 842–843.
48 Gao, C.; Yan, D. Chem Commun 2001, 1, 107–108.
49 Gao, C.; Yan, D. Prog Polym Sci 2004, 29, 183–275.
50 Ohta, T.; Hikino, H. Org Magn Reson 1979, 12, 445–449.
51 Ohta, T.; Hikino, H. Chem Pharm Bull 1981, 29, 280–282.
31 Kong, L. Z.; Su, M.; Qiao, H. M.; Pan, C. Y. J Polym Sci Part
A: Polym Chem 2010, 48, 454–462.
32 Li, X. R.; Su, Y. L.; Chen, Q. Y.; Lin, Y.; Tong, Y. J.; Li, Y. S.
Biomacromolecules 2005, 6, 3181–3188.
33 Ronald, G. W.; Malcolm, W. H.; Charles, H. S. J Org Chem
52 Hawker, C. J.; Lee, R.; Fre´chet, M. J. J Am Chem Soc 1991,
1969, 34, 576–580.
113, 4583–4588.
34 Jan, S.; Martin, N.; Kilian, M. Chem Eur
J 2006, 12,
53 Holter, D.; Burgath, A.; Frey, H. Acta Polym 1997, 48, 30–35.
¨
4367–4371.
54 Hanselmann, R.; Holter, D.; Frey, H. Macromolecules 1998,
¨
35 (a) Mather, B. D.; Viswanathan, K.; Miller, K. M.; Long, T. E.
Prog Polym Sci 2006, 31, 487–531; (b) Vernon, B.; Tirelli, N.;
Bachi, T.; Haldimann, D.; Hubbell, J. J Biomed Mater Res A
2003, 64, 447–456.
31, 3790–3801.
55 Strongly reduced solubility of the resultant HBPEAs were of-
ten observed after drying at high temperature which might be
the result of the formation of a strong hydrogen bond network.
36 (a) Yan, D. Y. Biomacromolecules 2003, 4, 704–712; (b)
Wang, G. J.; Wang, X. G. Polym Bull 2002, 49, 1–8.
56 Kim, Y. H.; Webster, O. W. J Am Chem Soc 1992, 114,
4947–4948.
37 Hong, C. Y.; You, Y. Z.; Wu, D. C.; Liu, Y.; Pan, C. Y. J Am
Chem Soc 2007, 129, 5354–5355.
57 (a) Gao, C.; Yan, D. Y.; Tang, W. Chem Phys 2001, 202,
2623–2629; (b) Gao, C.; Tang, W.; Yan, D. Y. J Polym Sci Part
A: Polym Chem 2002, 40, 2339–2340.
38 Ma, X. P.; Tang, J. B.; Shen, Y. Q.; Fan, M. H.; Tang, H. D.;
Radosz, M. J Am Chem Soc 2009, 131, 14795–14803.
58 Flory, P. J. J Am Chem Soc 1952, 74, 2718–2723.
39 Kang, T.; Amir, R. J.; Khan, A.; Ohshimizu, K.; Hunt, J. N.;
Sivanandan, K.; Montan˜ ez, M. I.; Malkoch, M.; Ueda, M.;
Hawker, C. J. Chem Commun 2010, 46, 1556–1558.
59 Wang, L.; He, X. H. J Polym Sci Part A: Polym Chem 2009,
47, 523–533.
40 Montan˜ ez, M. I.; Campos, L. M.; Antoni, P.; Hed, Y.; Walter,
M. V.; Krull, B. T.; Khan, A.; Hult, A.; Hawker, C. J.; Malkoch, M.
Macromolecules 2010, 43, 6004–6013.
60 Li, X. R.; Zhan, J.; Li, Y. S. Macromolecules 2004, 37,
7584–7594.
61 Perepelkin, K. E.; Maan´ ina, O. B.; Basok, M. O.; Makarova,
41 Dondoni, A. Angew Chem Int Ed Engl 2008, 47, 8995–8997.
R. A.; Oprits, Z. G. Fibre Chem 2005, 37, 196–198.
5374
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA