Well-controlled ring-opening polymerization of cyclic esters catalyzed by aluminum amido complexes: Kinetics and mechanism

Article abstract of DOI:10.1002/pola.27359

A series of aluminum dimethyl complexes 1-6 bearing N-[2-(pyrrolidinyl)benzyl]anilido ligands were synthesized and well characterized. The molecular structure of complex 1 determined by an X-ray diffraction study indicates the bidentate chelating mode of

Full text of DOI:10.1002/pola.27359

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Well-Controlled Ring-Opening Polymerization of Cyclic Esters Catalyzed
by Aluminum Amido Complexes: Kinetics and Mechanism
Junpeng Liu, Haiyan Ma
Shanghai Key Laboratory of Functional Materials Chemistry and Laboratory of Organometallic Chemistry, East China University
of Science and Technology, 130 Meilong Road, Shanghai 200237, People’s Republic of China
Correspondence to: H. Ma (E-mail: haiyanma@ecust.edu.cn)
Received 2 June 2014; accepted 4 August 2014; published online 00 Month 2014
DOI: 10.1002/pola.27359
ABSTRACT: A series of aluminum dimethyl complexes 1–6 bear-
ing N-[2-(pyrrolidinyl)benzyl]anilido ligands were synthesized
and well characterized. The molecular structure of complex 1
determined by an X-ray diffraction study indicates the biden-
tate chelating mode of the pyrrolidinyl-anilido ligand. In the
absence of a coinitiator, these complexes exhibited excellent
control toward the polymerizations of e-caprolactone and rac-
lactide, affording polyesters with quite narrow molecular
weight distributions (Mw/Mn 5 1.04–1.26). The end group analy-
sis of e2CL oligomer via 1H NMR and ESI-TOF MS methods
gave strong support to the hypothesis that the polymerization
coordination-insertion mechanism involving a unique AlAN
(amido) bond initiation. Via 1H NMR scale oligomerization
studies, it is suggested that the insertion of the first lactide
monomer into AlAN bond of the complex is much easier than
the insertion of lactide monomer into the newly formed AlAO
(lactate) bond and might also be easier than the insertion of
C
V
the first e2CL monomer into AlAN bond. 2014 Wiley Periodi-
cals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2014, 00, 000–
000
KEYWORDS: aluminum complex; amido ligand; catalysis;
catalyzed by these aluminum complexes proceeds via
a
polyesters; ring-opening polymerization
methyl complexes were also moderately active for rac-LA
polymerization.28,29 However, for both systems, the polymer-
izations were poorly controlled, producing polymers with sig-
nificantly deviated molecular weights and rather broad
molecular weight distributions (Mw/Mn 5 1.66–3.74). Further
studies indicated that both AlAN and AlACH3 bonds in the
amidinate aluminum methyl complexes participated in the ini-
tiation. The strategy of generating in situ the desired alumi-
num alkoxide species capable of executing control on the
polymerization proved to be unsuccessful. The b-diketiminate
aluminum ethyl complexes were unreactive toward isopropa-
nol or benzyl alcohol even under violent reaction conditions,
whereas the treatment of the amindinate aluminum com-
plexes with isopropanol led to the release of free amindines.
Similar results were also reported by Peng’s group:22 for alu-
minum complexes bearing N-alkyl substituted b-diketiminate
ligands, the addition of benzyl alcohol led to the dissociation
of the ligand from metal center. Therefore, aluminum com-
plexes bearing multidentate ligands with only N-donors,
which are also capable of initiating the ROP of cyclic esters in
a well-controlled manner, are still less explored.
INTRODUCTION Structurally well-defined organometallic com-
plexes demonstrate prominent superiority in catalyzing the
ring-opening polymerization (ROP) of cyclic esters.1,2 Among
them, aluminum complexes have attracted significant research
focus, especially in the field of isoselective ROP of rac-lactide
(rac-LA).3–17 Although a variety of aluminum complexes have
been explored up to date, in principal, only those having at
least one alkoxy ligand could initiate the polymerization of
cyclic esters in a well-controlled manner.1–7 Due to the difficul-
ties encountered in synthesis and/or purification, structurally
well-defined aluminum alkoxide complexes are still limited. To
overcome this problem, many studies have involved the use of
coinitiators, such as alcohols, to convert the easily prepared
aluminum alkyl complexes to the corresponding aluminum alk-
oxide complexes in situ.18–27 Nevertheless, the addition of such
coinitiators often induced unpredictable reactions, for example,
the dissociation of ancillary ligand from the metal center,19
especially for aluminum complexes bearing multidentate
ligands with only N-donors.12,20–23
Previously, we reported series of aluminum alkyl complexes
bearing b-diketiminate and amidinate ligands.13,28,29 All of
them demonstrated sufficient catalytic activities toward the
ROP of e-caprolactone (e2CL), and the amidinate aluminum
Very recently, we reported a series of aluminum alkyl com-
plexes bearing bidentate N-[2-(1-piperidinyl)benzyl]anilino
Additional Supporting Information may be found in the online version of this article.
C
V
2014 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
or N-(2-morpholinobenzyl)anilino ligands.30 All complexes
proved to be capable of initiating the ROP of rac-lactide with
living features, providing narrowly distributed PLAs in the
absence of a coinitiator. To further expand the ligand system,
in this work several N-[2-(1-pyrrolidinyl)benzyl]anilines
were synthesized. The corresponding aluminum dimethyl
complexes were obtained and evaluated for the ROP of cyclic
esters. Detailed kinetic and mechanistic studies were also
performed to have a thorough understanding on these
systems.
solution with dichloromethane (15 mL 3 3). The combined
organic layer was dried over anhydrous MgSO4. Evaporation
of all the volatile under reduced pressure yielded the crude
compound, which was further purified by column chromato-
graph eluted by anhydrous EtOH to provide yellow oil, char-
acterized as N-[2-(1-pyrrolidinyl)benzyl]aniline (1.79 g, 95%
purity, which could not be further purified). 1H NMR (400
MHz, CDCl3, d): 7.35 (m, 1H, ArAH), 7.23 (m, 3H, ArAH),
7.01 (m, 1H, ArAH), 6.90 (m, 1H, ArAH), 6.75 (m, 1H,
ArAH), 6.67 (m, 2H, ArAH), 4.36 (s with shoulder, 3H,
ArACH2 and NH), 3.24 (br, 4H, ANCH2), 1.95 (br, 4H,
ACH2A). 13C NMR (100 MHz, CDCl3, d): 148.7 (ArAC), 148.6
(ArAC), 130.2 (ArAC), 129.2 (ArAC), 127.9 (ArAC), 120.6
(ArAC), 118.3 (ArAC), 117.2 (ArAC), 116.6 (ArAC), 115.0
(ArAC), 112.8 (ArAC), 51.2 (ArACH2), 46.7 (ANCH2), 25.0
(ACH2A).
EXPERIMENTAL
Materials and Methods
All manipulations were carried out under a dry argon atmos-
phere using standard Schlenk techniques or an MBraun
glove-box. Toluene and n-hexane were refluxed over sodium/
benzophenone prior to use. Dichloromethane and chloro-
form-d were dried over calcium hydride. Benzene-d6 was
refluxed over sodium and distilled under argon. AlMe3
(2.0 M solution in toluene, Aldrich) was used as received. 2-
(1-Pyrrolidinyl)benzaldehyde was prepared according to the
published procedure.31 e2CL (99%, Aldrich) was dried over
CaH2 for 24 h at 80 ꢀC, then vacuum distilled and stored
under argon. rac-Lactide and L-lactide (Aldrich) were recrys-
tallized with dry toluene and then sublimated twice under
vacuum at 80 ꢀC. All other chemicals were commercially
available and used after appropriate purification. Glassware
and vials used in the polymerization were dried in an oven
at 120 ꢀC overnight and exposed to vacuum-argon cycle
three times.
In the glove-box, trimethylaluminum (2 mL, 4.00 mmol, 2 M
solution in toluene) was slowly added to a solution of the
above yellow oil (1.00 g) in hexane (15 mL), and then the
solution was stirred for 24 h at r.t. A large amount of white
solids precipitated. The reaction mixture was filtered, and the
solid was recrystallized with a mixture of dichloromethane
ꢀ
and hexane at 220 C to afford complex 1 as colorless crys-
1
tals (0.380 g, 30.8%). H NMR (400 MHz, CDCl3, d): 7.35 (dd,
J 5 7.5, 1.5 Hz, 1H, ArAH), 7.28 (td, J 5 7.5, 1.5 Hz, 1H,
ArAH), 7.22 (dd, J 5 7.3, 1.1 Hz, 1H, ArAH), 7.20–7.13 (m, 3H,
ArAH), 6.71 (d, J 5 8.0 Hz, 2H, ArAH), 6.58 (t, J 5 8.0 Hz, 1H,
ArAH), 4.44 (s, 2H, ArACH2), 3.51 (m, 4H, ANCH2), 2.04 (m,
4H, ACH2A), 20.89 (s, 6H, AlACH3). 13C NMR (100 MHz,
CDCl3, d): 154.2 (ArAC), 144.9 (ArAC), 134.6 (ArAC), 132.1
(ArAC), 128.8 (ArAC), 127.9 (ArAC), 127.1 (ArAC), 119.0
(ArAC), 114.5 (ArAC), 114.4 (ArAC), 54.1 (ArACH2), 53.5
(NCH2), 23.7 (CH2), 210.4 (AlACH3). Calcd. for C19H25AlN2:
C, 74.00; H, 8.17; N, 9.08. Found: C, 73.80; H, 8.18; N, 9.08%.
NMR spectra were recorded on a Bruker AVANCE-400 spec-
trometer with CDCl3 or C6D6 as solvent (1H: 400 MHz; 13C:
1
100 MHz). Chemical shifts for H and 13C NMR spectra were
referenced internally using the residual solvent resonances
and reported relative to tetramethylsilane. Elemental analy-
ses were performed on an EA-1106 instrument. Gel permea-
tion chromatography (GPC) analyses were carried out on a
Waters instrument (M515 pump, Optilab Rex injector) in
THF at 40 ꢀC, at a flow rate of 1 mLꢁmin21, with narrowly
distributed linear polystyrene samples (103 < Mn < 2 3 106
g mol21) as calibration standards.
Synthesis of 2-Methyl-N-[2-(1-pyrrolidinyl) benzyl]anilido
aluminum dimethyl (2)
The procedure was similar to that of complex 1. o-Toluidine
(2.14 g, 20.0 mmol) was reacted with 2-(1-pyrrolidinyl)ben-
zaldehyde (3.50 g, 20.0 mmol) and sequentially reduced with
NaBH4 (11.4 g, 0.30 mol). After work-up, light yellow oil was
isolated, characterized mainly as 2-methyl-N-[2-(1-pyrrolidi-
nyl)benzyl]aniline (1.73 g, 90% purity, which could not be
further purified). 1H NMR (400 MHz, CDCl3, d): 7.37 (dd,
J 5 7.6, 1.6 Hz, 1H, ArAH), 7.27 (td, J 5 7.6, 1.6 Hz, 1H,
ArAH), 7.18 (t, J 5 7.8 Hz, 1H, ArAH), 7.10 (d, J 5 7.3 Hz, 1H,
ArAH), 7.06 (dd, J 5 8.1, 1.0 Hz, 1H, ArAH), 6.96 (td, J 5 7.3,
1.0 Hz, 1H, ArAH), 6.71 (m, 2H, ArAH), 4.40 (br s, 3H,
ArACH2 and NH), 3.25 (m, 4H, ANCH2), 2.18 (s, 3H, ArACH3),
1.96 (m, 4H, ACH2A). 13C NMR (100 MHz, CDCl3, d): 148.9
(ArAC), 146.6 (ArAC), 130.6 (ArAC), 130.0 (ArAC), 129.2
(ArAC), 128.1 (ArAC), 127.3 (ArAC), 122.1 (ArAC), 120.8
(ArAC), 116.9 (ArAC), 116.7 (ArAC), 109.8 (ArAC), 51.1
(ArACH2), 47.2 (ANCH2), 25.1 (ACH2A), 17.6 (ArACH3).
Syntheses of Aluminum Dimethyl Complexes
Synthesis of N-[2-(1-Pyrrolidinyl)benzyl]anilido aluminum
dimethyl (1)
To a solution of 2-(1-pyrrolidinyl) benzaldehyde (3.50 g,
20.0 mmol) in benzene (80 mL), freshly distilled aniline
(1.86 g, 20.0 mmol) and para-toluenesulfonic acid monohy-
drate (2 mg) were added at r.t. The reaction solution was
refluxed for 12 h and the water formed in the reaction was
separated by water segregator. The solution was then cooled
and filtered. The filtrate was concentrated to dryness under
vacuum and the crude product was dissolved in absolute
alcohol (200 mL), followed by an addition of sodium borohy-
dride (11.4 g, 0.3 mol). The mixture was heated to reflux for
36 h, and water was added, followed by extracting the
The above light yellow oil (1.064 g) was treated with trime-
thylaluminum (2 mL, 4.00 mmol, 2 M solution in toluene) to
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
afford the target product as colorless crystals (0.399 g,
31.0%). 1H NMR (400 MHz, CDCl3, d): 7.27 (d, J 5 7.2 Hz,
1H, ArAH), 7.21 (t, J 5 7.1 Hz, 1H, ArAH), 7.14 (t, J 5 7.1 Hz,
1H, ArAH), 7.11–7.05 (m, 2H, ArAH), 6.99 (d, J 5 7.1 Hz, 1H,
ArAH), 6.83 (d, J 5 8.0 Hz, 1H, ArAH), 6.56 (t, J 5 7.2 Hz,
1H, ArAH), 4.45 (s, 2H, ArACH2), 3.46 (m, 4H, ANCH2A),
2.26 (s, 3H, ArACH3), 1.96 (m, 4H, ACH2A), 21.10 (s, 6H,
AlACH3). 13C NMR (100 MHz, CDCl3, d): 153.8 (ArAC), 144.7
ArAC), 134.2 (ArAC), 130.9 (ArAC), 129.6 (ArAC), 126.9
(ArAC), 126.7 (ArAC), 126.0 (ArAC), 125.5 (ArAC), 117.7
(ArAC), 115.4(ArAC), 114.7 (ArAC), 54.4 (ArACH2), 53.2
(NCH2), 24.0 (CH2), 19.8 (ArACH3), 210.8 (AlACH3). Calcd.
for C20H27AlN2: C, 74.50; H, 8.44; N, 8.69. Found: C, 74.24; H,
8.46; N, 8.53%.
reduced with NaBH4 (11.4 g, 0.30 mol). After work-up, light
yellow solids were isolated mainly as 2-chloro-N-[2-(1-pyrro-
lidinyl)benzyl]aniline (1.92 g, 95% purity, which could not
be further purified). 1H NMR (400 MHz, CDCl3, d): 7.27 (d,
J 5 7.5 Hz, 1H, ArAH), 7.21 (dd, J 5 8.0, 1.3 Hz, 1H, ArAH),
7.18 (td, J 5 7.4, 1.3 Hz, 1H, ArAH), 7.08 (td, J 5 7.8, 1.3 Hz,
1H, ArAH), 6.98 (d, J 5 8.0 Hz, 1H, ArAH), 6.88 (t, J 5 7.3
Hz, 1H, ArAH), 6.62 (d, J 5 8.0 Hz, 1H, ArAH), 6.58 (td,
J 5 7.5, 1.0 Hz, 1H, ArAH), 4.33 (s, 2H, phenylACH2), 3.14
(m, 4H, ANCH2), 1.89 (m, 4H, ACH2A). 13C NMR (100 MHz,
CDCl3, d): 148.7 (ArAC), 144.1 (ArAC), 129.9 (ArAC), 128.9
(ArAC), 128.7 (ArAC), 127.9 (ArAC), 127.7 (ArAC), 120.8
(ArAC), 119.1 (ArAC), 116.9 (ArAC), 116.8 (ArAC), 111.2
(ArAC), 51.0 (ArACH2), 46.4 (NCH2), 24.9 (CH2).
The solution of above light yellow solids (1.14 g) in hexane
was treated with trimethylaluminum (2 mL, 4.00 mmol, 2 M
solution in toluene) to afford the target product as colorless
Synthesis of 4-Isopropyl-N-[2-(1-pyrrolidinyl)
benzy]anilido aluminum dimethyl (3)
The procedure was similar to that of complex 1.
4-Isopropsylaniline (2.70 g, 20.0 mmol) was reacted with 2-
(1-pyrrolidinyl)benzaldehyde (3.50 g, 20.0 mmol) and
sequentially reduced with NaBH4 (11.4 g, 0.30 mol). After
work-up, light yellow oil was isolated, characterized mainly
as 4-isopropyl-N-[2-(1-pyrrolidinyl) benzyl]aniline (1.79 g,
90% purity, which could not be further purified). 1H NMR
(400 MHz, CDCl3, d): 7.31 (dd, J 5 7.5, 1.5 Hz, 1H, ArAH),
7.20 (td, J 5 8.0, 1.5 Hz, 1H, ArAH), 7.05 (d, J 5 8.4 Hz, 2H,
ArAH), 6.96 (d, J 5 8.0 Hz, 1H, ArAH), 6.89 (td, J 5 7.5, 0.75
Hz, 1H, ArAH), 6.60 (d, J 5 8.4 Hz, 2H, ArAH), 4.30 (s, 2H,
ArACH2), 4.22 (br s, 1H, NH), 3.21 (m, 4H, ANCH2), 2.81
(sept, J 5 6.8 Hz, 1H, ACH(CH3)2), 1.92 (m, 4H, ACH2A),
1.21 (d, J 5 6.8 Hz, 6H, ACH(CH3)2). 13C NMR (100 MHz,
CDCl3, d): 148.4 (ArAC), 146.4 (ArAC), 137.3 (ArAC), 130.1
(ArAC), 129.0 (ArAC), 127.6 (ArAC), 126.8 (ArAC), 120.3
(ArAC), 116.2 (ArAC), 112.5 (ArAC), 50.9 (ArACH2), 46.7
(NCH2), 32.9 [CH(CH3)2], 24.7 (CH2), 24.1 [CH(CH3)2].
1
crystals (0.526 g, 38.5%). H NMR (400 MHz, CDCl3, d): 7.35
(d, J 5 7.4 Hz, 1H, ArAH), 7.30 (td, J 5 7.8, 1.5 Hz, 1H,
ArAH), 7.23 (m, 2H, ArAH), 7.17 (m, 2H, ArAH), 6.81 (d,
J 5 8.1 Hz, 1H, ArAH), 6.67 (td, J 5 7.5, 1.0 Hz, 1H, ArAH),
4.44 (s, 2H, ArACH2), 3.50 (br, 4H, ANCH2), 2.02 (m, 4H,
ACH2A), 20.94 (s, 6H, AlACH3). 13C NMR (100 MHz, CDCl3,
d): 151.9 (ArAC), 146.4 (ArAC), 133.8 (ArAC), 132.2
(ArAC), 129.0 (ArAC), 128.3 (ArAC), 127.7 (ArAC), 126.8
(ArAC), 122.3 (ArAC), 118.6 (ArAC), 115.7 (ArAC), 114.7
(ArAC), 54.6 (ArACH2), 54.1 (NCH2), 24.3 (CH2), 29.2
(AlACH3). Calcd. for C19H24AlClN2: C, 66.56; H, 7.06; N, 8.17.
Found: C, 66.48; H, 7.20; N, 7.85%.
Synthesis of 3-Chloro-N-[2-(1-pyrrolidinyl) benzyl]anilido
aluminum dimethyl (5)
The procedure was similar to that of complex 1.
3-Chloroaniline (2.54 g, 20.0 mmol) was reacted with 2-(1-
pyrrolidinyl)benzaldehyde (3.50 g, 20.0 mmol) and sequen-
tially reduced with NaBH4 (11.4 g, 0.30 mol). After work-up,
light yellow solids were isolated mainly as 3-chloro-N-[2-(1-
pyrrolidinyl)benzyl]aniline (1.83 g, 80% purity, which could
not be further purified). 1H NMR (400 MHz, CDCl3, d): 7.25
(dd, J 5 7.5, 1.5 Hz, 1H, ArAH), 7.20 (td, J 5 7.4, 1.5 Hz, 1H,
ArAH), 7.03 (t, J 5 8.0 Hz, 1H, ArAH), 6.96 (t, J 5 7.3 Hz, 1H,
ArAH), 6.88 (td, J 5 7.5, 1.0 Hz, 1H, ArAH), 6.63 (d, J 5 8.0
Hz, 1H, ArAH), 6.59 (t, J 5 2.0 Hz, 1H, ArAH), 6.45 (dd,
J 5 8.0, 2.0 Hz, 1H, ArAH), 4.45 (br, 1H, ANH), 4.26 (br s,
2H, ArACH2), 3.16 (m, 4H, ANCH2), 1.90 (m, 4H, ACH2A).
13C NMR (100 MHz, CDCl3, d): 149.5 (ArAC), 148.6 (ArAC),
134.9 (ArAC), 130.12 (ArAC), 130.16 (ArAC), 128.4 (ArAC),
128.0 (ArAC), 120.6 (ArAC), 116.9 (ArAC), 116.7 (ArAC),
112.2 (ArAC), 111.0 (ArAC), 51.1 (ArACH2), 46.5 (ANCH2),
24.9 (ACH2A).
The above light yellow oil (1.18 g) was treated with trime-
thylaluminum (2 mL, 4.00 mmol, 2 M solution in toluene) to
afford the target product as colorless crystals (0.468 g,
33.4%). 1H NMR (400 MHz, CDCl3, d): 7.33 (dd, J 5 7.4, 1.7
Hz, 1H, ArAH), 7.27 (td, J 5 7.9, 1.7 Hz, 1H, ArAH), 7.21 (td,
J 5 7.4, 1.0 Hz, 1H, ArAH), 7.15 (d, J 5 7.9, 1.0 Hz, 1H,
ArAH), 7.05 (d, J 5 8.0 Hz, 2H, ArAH), 6.66 (d, J 5 8.0 Hz,
2H, ArAH), 4.44 (s, 2H, ArACH2), 3.51 (m, 4H, ANCH2), 2.80
(sept, J 5 6.9 Hz, 1H, CH(CH3)2), 2.04 (m, 4H, ACH2A), 1.21
(d, J 5 6.9 Hz, 6H, CH(CH3)2), 20.93 (s, 6H, AlACH3). 13C
NMR (100 MHz, CDCl3, d): 152.2 (ArAC), 144.9 (ArAC),
134.74 (ArAC), 134.70 (ArAC), 132.0 (ArAC), 127.8 (ArAC),
127.0 (ArAC), 126.7 (ArAC), 119.0 (ArAC), 114.0 (ArAC),
54.0 (ArACH2), 53.8 (NCH2), 33.0 [CH(CH3)2], 24.3 (CH2),
23.7 [CH(CH3)2], 210.3 (AlACH3). Calcd. for C22H31AlN2: C,
75.39; H, 8.92; N, 7.99. Found: C, 74.92; H, 8.95; N, 8.13%.
The solution of above light yellow solids (1.14 g) in hexane
was treated with trimethylaluminum (2 mL, 4.00 mmol, 2 M
solution in toluene) to afford the target product as colorless
crystals (0.583 g, 42.6%). H NMR (400 MHz, CDCl3, d): 7.34
(d, J 5 7.3 Hz, 1H, ArAH), 7.29 (td, J 5 7.9, 1.7 Hz, 1H,
ArAH), 7.23 (td, J 5 7.3, 1.0 Hz, 1H, ArAH), 7.17 (d, J 5 7.9
Synthesis of 2-Chloro-N-[2-(1-pyrrolidinyl) benzyl]anilido
aluminum dimethyl (4)
The procedure was similar to that of complex 1. 2-
Chloroaniline (2.54 g, 20.0 mmol) was reacted with 2-(1-pyr-
rolidinyl)benzaldehyde (3.50 g, 20.0 mmol) and sequentially
1
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Hz, 1H, ArAH), 7.02 (t, J 5 7.9 Hz, 1H, ArAH), 6.64 (t, J 5 1.9
Hz, 1H, ArAH), 6.58 (dd, J 5 8.0, 1.9 Hz, 1H, ArAH), 6.54
(dd, J 5 8.0, 1.9 Hz, 1H, ArAH), 4.38 (s, 2H, ArACH2), 3.50
(m, 4H, ANCH2), 2.05 (br, 4H, ACH2A), 20.90 (s, 6H,
AlACH3). 13C NMR (100 MHz, CDCl3, d): 155.6 (ArAC), 144.8
(ArAC), 134.7 (ArAC), 134.0 (ArAC), 132.1 (ArAC), 129.3
(ArAC), 128.2 (ArAC), 127.3 (ArAC), 119.0 (ArAC), 114.3
(ArAC), 113.9 (ArAC), 113.1 (ArAC), 54.2 (ArACH2), 53.4
(NCH2), 23.8 (CH2), 210.5 (AlACH3). Calcd. for
C19H24AlClN2: C, 66.56; H, 7.06; N, 8.17. Found: C, 65.97; H,
7.04; N, 8.07%.
SCHEME 1 Synthesis of aluminum dimethyl complexes 1–6.
Synthesis of 4-Chloro-N-[2-(1-pyrrolidinyl) benzy]anilido
aluminum dimethyl (6)
anisotropically.33 Hydrogen atoms of 1 were treated using a
riding model and refined by the geometry method. Molecule
structure was generated using ORTEP III program.34
The procedure was similar to that of complex 1. 4-
Chloroaniline (2.54 g, 20.0 mmol) was reacted with 2-(1-pyr-
rolidinyl)benzaldehyde (3.50 g, 20.0 mmol) and sequentially
reduced with NaBH4 (11.4 g, 0.30 mol). After work-up, color-
less solids were isolated, characterized mainly as 4-chloro-N-
[2-(1-pyrrolidinyl)benzyl] aniline (1.97 g, 90% purity, which
Polymerization Procedure
In a glove-box, a proper amount of e2CL (in toluene) or rac-
LA was added to a Schlenk tube, and a certain amount of tol-
uene and the solution of aluminum complex in toluene were
injected sequentially ([e2CL] 5 4 mol/L, [e-CL]0/[Al]0 5 100;
or [rac-LA] 5 1 mol/L, [rac-LA]0/[Al]0 5 100). The tube was
taken out of the glove-box and immerged into an oil bath
stabilized at desired temperature. The mixture was stirred
for a specific time interval and then quenched with wet
petroleum ether. Monomer conversion was monitored by
integration of monomer vs. polymer methine (for rac-lactide)
1
could not be further purified). H NMR (400 MHz, CDCl3, d):
7.27 (dd, J 5 7.3, 1.3 Hz, 1H, ArAH), 7.18 (td, J 5 7.6, 1.3 Hz,
1H, ArAH), 7.10 (m, 2H, ArAH), 6.97 (d, J 5 8.8 Hz, 2H,
ArAH), 6.54 (d, J 5 8.8 Hz, 2H, ArAH), 4.39 (br, 1H, NH),
4.27 (s, 2H, ArACH2), 3.17 (m, 4H, ANCH2), 1.91 (m, 4H,
ACH2A). 13C NMR (100 MHz, CDCl3, d): 148.9 (ArAC), 147.2
(ArAC), 130.3 (ArAC), 129.2 (ArAC), 128.9 (ArAC), 128.3
(ArAC), 122.0 (ArAC), 120.9 (ArAC), 117.0 (ArAC), 114.0
(ArAC), 51.4 (ArACH2), 47.1 (NCH2), 25.2 (CH2).
1
or methylene (for e2CL) resonance in the H NMR spectrum
(CDCl3, 400 MHz) after removal of the volatiles. The purifica-
tion of the polymer was managed by dissolving the crude
sample in CH2Cl2 and precipitating the polymer solution
with methanol. The obtained polymer was further dried in a
vacuum oven at 60 ꢀC for 36 h and subjected to GPC
measurement.
The solution of above solids (1.14 g) in hexane was treated
with trimethylaluminum (2 mL, 4.00 mmol, 2 M solution in
toluene) to afford the target product as colorless crystals
(0.564 g, 41.0%). 1H NMR (400 MHz, CDCl3, d): 7.35 (dd,
J 5 7.2, 1.2 Hz, 1H, ArAH), 7.29 (td, J 5 7.8, 1.6 Hz, 1H,
ArAH), 7.22 (td, J 5 7.8, 1.2 Hz, 1H, ArAH), 7.17 (dd, J 5 7.2,
1.6 Hz, 1H, ArAH), 7.08 (d, J 5 8.9 Hz, 2H, ArAH), 6.60 (d,
J 5 8.9 Hz, 2H, ArAH), 4.38 (s, 2H, ArACH2), 3.50 (m, 4H,
ANCH2), 2.04 (m, 4H, ACH2A), 20.91 (s, 6H, AlACH3). 13C
NMR (100 MHz, CDCl3, d): 152.8 (ArAC), 144.8 (ArAC),
134.2 (ArAC), 132.2 (ArAC), 128.5 (ArAC), 128.1 (ArAC),
127.2 (ArAC), 119.05 (ArAC), 119.00 (ArAC), 115.3 (ArAC),
54.2 (ArACH2), 53.6 (NCH2), 23.7 (CH2), 210.5 (AlACH3).
Calcd. for C19H24AlClN2: C, 66.56; H, 7.06; N, 8.17 %. Found:
C, 66.83; H, 6.99; N, 8.22%.
RESULTS AND DISCUSSION
Synthesis and Characterization of Aluminum Dimethyl
Complexes 1–6
As shown in Scheme 1, the nucleophilic substitution reaction
between pyrrolidine and 2-fluorobenzaldehyde afforded 2-
(1-pyrrolidinyl) benzaldehyde; then it was condensed with
substituted anilines and sequentially reduced with sodium boro-
hydride to produce the target N-[2-(1-pyrrolidinyl)benzyl]ani-
line proligands L1–6H. Due to the strong hygroscopicity of these
compounds and their similar polarity to the starting anilines, all
of them could not be isolated in an analytically pure form.
X-Ray Crystallography
Single crystals of complex 1 suitable for X-ray diffraction
studies were obtained from
a
saturated hexane-
dichloromethane solution at 220 ꢀC. The crystallographic
data for complex 1 was collected on a Bruker SMART APEX
diffractometer with graphite-monochromated Mo-Ka radia-
Unlike the previously reported N-[2-(1-piperidinyl)benzyl]a-
niline system,30 the reaction of proligands L1–6H with one
equiv. of AlMe3 at r.t. readily afforded the target aluminum
dimethyl complexes 1–6 (Scheme 1). No production of the tri-
methylaluminum adduct with the neutral ligand was detected.
ꢀ
tion (k 5 0.71073 Å). All data were collected at 220 C using
omega-scan techniques. The structure of complex 1 was
solved by direct methods and subsequently refined on F2 by
full-matrix least-squares methods using the SHELXTL pro-
gram package.32 SADABS absorption corrections were
applied to the data. All non-hydrogen atoms were refined
1
In the H NMR spectra of complexes 1–6, the signal of Al-CH3
is observed at 20.89 ꢂ 21.10 ppm as a characteristic singlet.
The resonance of NCH2 (pyrrolidinyl) at around 3.50 ppm is
significantly downfield shifted in comparison with that of the
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
crystallizes in a monoclinic space group P2(1)/c, and the alu-
minum center is coordinated by the two nitrogen donors of
the ligand and two methyl groups in a distorted tetrahedral
geometry. The N1-Al1-N2 bond angle of 96.62(7)ꢀ, smaller
than the ideal tetrahedral bond angle of 109.5ꢀ, is close to
those in related known Al complexes.11
The Al1-N1 bond length of 1.8583(17) Å is significantly
shorter than the Al1-N2 bond length of 2.0130(19) Å, implying
the difference between the amido N1 atom and the amine N2
atom in bonding with the aluminum center. Interestingly, the
nitrogen atom N1 is roughly in a planar fashion surrounded by
C7, C8 and Al1 with the sum of the corresponding bond angles
being 359.7ꢀ, which might imply some double bonding charac-
ter of Al1-N1 bond. The N2 atom has a distorted tetrahedral
environment in which the C17-N2-Al1 and C14-N2-Al1 have
different bond angles of 115.61(14)ꢀ and 104.66(13)ꢀ, prob-
ably due to the steric requirements of the two methyl ligands.
FIGURE 1 ORTEP drawing of 1 (thermal ellipsoids drawn at the
˚
50% probability level). Selected bond lengths [A] and angles
[ꢀ]: Al1-N1 1.8583(17), Al1-N2 2.0130(19), Al1-C18 1.963(2), Al1-
C19 1.968(3). N1-Al1-C18 114.74(10), N1-Al1-C19 113.43(10),
C18-Al1-C19 118.61(11), N1-Al1-N2 96.62(7), C18-Al1-N2
103.87(9), C19-Al1-N2 105.99(10), C8-N1-C7 116.38(15), C8-N1-
Al1 121.99(13), C7-N1-Al1 121.37(13), C1-N2-C17 109.13(15), C1-
Moreover, the alcoholysis reaction of complexes 1–6 failed to
afford the desired N-[2-(1-pyrrolidinyl)benzyl]anilino alumi-
num alkoxide complexes. The reaction of representative com-
plex 6 with isopropanol led to the production of the free
ligand L6H and some undissolvable solids (see Supporting
Information), which hampered a further understanding on
the reaction.
N2-C14
114.75(16),
C17-N2-C14
102.42(17),
C1-N2-Al1
110.20(13), C17-N2-Al1 115.61(14), C14-N2-Al1 104.66(13).
proligand (3.14–3.24 ppm), indicative of the coordination of
pyrrolidinyl-N donor to aluminum center in these complexes.
ROP of e2CL
ꢀ
The molecular structure of complex 1 with selected bond
lengths and bond angles is shown in Figure 1. Complex 1
The ROP of e2CL was studied at 35 and 45 C in toluene by
using aluminum complexes 1–6 as single component
TABLE 1 The ROP of e2CL Initiated by Complexes 1–6a
Conv. (%)b
Mn, calcd (3104)
Mn, NMR (3104)
Mn (3104)
Mw/Mn
c
d
e
e
0
Run
Cat.
Temp. (ꢀC)
Time (min)
1
1 (H)
35
35
45
35
35
45
35
35
45
35
35
45
35
45
35
45
40
34
83
90
19
84
83
30
81
88
20
87
86
39
96
35
92
0.39
0.95
1.03
0.22
0.96
0.95
0.34
0.92
1.00
0.23
0.99
0.98
0.44
1.09
0.34
1.05
0.41
0.94
1.10
0.21
0.98
1.01
0.41
0.92
1.11
0.28
1.10
1.13
0.52
1.16
0.41
1.25
0.52
0.92
1.23
1.07
1.04
1.11
2
150
60
3
4
2 (o-Me)
3 (p-iPr)
4 (o-Cl)
40
5
210
90
0.82
0.73
0.55
1.15
1.06
1.04
6
7
40
8
150
60
9
1.09
1.09
10
11
12
13
14
15
16
40
210
90
0.90
1.05
0.51
1.37
1.11
1.14
1.04
1.07
5 (m-Cl)
6 (p-Cl)
40
60
40
60
0.99
1.04
a
d
[e2CL]0 5 4.0 M, [e2CL]0/[Al]0 5 100, in toluene.
Determined by the integration ratio of the methylene protons in mono-
Determined by the integration ratio of the methylene protons in poly-
mer main chain versus the characteristic resonance of the amido ligand.
The number average molecular weight (Mn) and molecular weight dis-
b
e
mer and polymer in CDCl3.
c
Mn,calcd 5 ([e2CL]0/[Al]0)
3
114.14
3
conv. (%). 1 mass of the free
tribution (Mw/Mn) were determined by a gel permeation chromatograph
in THF at 40 ꢀC, using narrowly distributed polystyrenes as standards,
Mn’ 5 0.56 3 Mn.
ligand.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
0
FIGURE 2 Plots of Mn versus monomer conversion for e2CL
polymerization and PDI valves using complex 1 as initiator (in
toluene, 35 ꢀC, [e2CL]0 5 4 mol/L, [e2CL]0/[Al]0 5 100. ᭿: GPC
FIGURE 3 Plots of ln([e2CL]0/[e2CL]t) versus time for e2CL
polymerization using complex 1 as initiator (in toluene, 35 ꢀC,
[e2CL]0 5 4 mol/L. ᭿: [Al]0 5 0.01 mol/L,
k
app 5 4.85
3
values; : Theoretic values).
1025 6 0.800 3 1026 s21
;
: [Al]0 5 0.02 mol/L, kapp 5 9.73 3
•
•
1025 6 6.67 3 1026 s21; ~: [Al]0 5 0.03 mol/L, kapp 5 1.47 3
1024 6 2.67 3 1026 s21; !: [Al]0 5 0.04 mol/L, kapp 5 1.97 3
1024 6 6.33 3 1026 s21).
initiators ([e2CL]0:[Al]0hx2009;5 100:1), and the results are
summarized in Table 1. All these aluminum complexes
proved to be active initiators toward the polymerization of
e2CL, and high monomer conversions up to 96% could be
reached within 1–4 h by adopting a monomer concentration
of 4.0 mol/L. It is found that the nature and position of the
substituent on the anilino moiety of the supporting ligand
have different effects on the catalytic activity of these com-
plexes. In all cases, the introduction of an ortho-substituent
to the anilino moiety leads to an obvious decrease of the cat-
alytic activity [2 (o-Me) and 4 (o-Cl) vs. 1 (H)], implying that
the steric hindrance around the metal center is unfavorable
for the polymerization. In contrast to the dominant steric
effect played by an ortho-chloro group in complex 4, the
electron-withdrawing effect of a chloro-group at the meta- or
para-position of the anilino moiety seems to be more crucial.
Both complexes 5 (m-Cl) and 6 (p-Cl) show higher catalytic
activities than the rest aluminum complexes, and complex 5
is more active than complex 6. The benefit of an electron-
withdrawing group at these positions is also witnessed by
the lower activity of complex 3 (p-iPr) in comparison with
that of complex 1. All these results are however similar to
those reported previously for aluminum complexes with
analogous piperidinyl-anilino ligands.30
e2CL polymerization with complex 1 as the initiator was
studied in detail. The semilogarithmic plots of ln([e 2 CL]0/
[e 2 CL]t) versus reaction time (t) at different initiator concen-
trations are shown in Figure 3. The plots exhibit perfect linear
relationship and no obvious induction period is observed in
each case, indicating the first order dependence of the poly-
merization on monomer concentration. Thus, the polymeriza-
tion of e2CL by complex 1 proceeds according to eq 1, where
k
app 5 kp[Al]x, kapp and kp are the apparent propagation and
propagation rate constants, respectively. The order in alumi-
num concentration (x) was determined by plotting lnkapp vs.
ln[Al]0, and a straight line was obtained with a slope of 1.09
(Figure 4). So the polymerization of e2CL follows an overall
kinetic law given by eq 2 with kp 5 0.052 Lꢁmol21ꢁs21. That
is, the rate determining step involves one metal center and
one monomer molecule stoichiometrically.
As shown in Table 1, PCLs with very narrow molecular
weight distributions (Mw/Mn 5 1.04–1.15) are provided by
these aluminum complexes. The corrected number average
0
molecular weights (Mn ) of the polymers (Mn0 5 0.56 3 Mn;
Mn, determined by GPC)35 approximately match the theoreti-
cal values calculated on the assumption that each active
metal center initiates one polymer chain. Furthermore, a lin-
0
ear relationship between Mn and the conversion of e2CL
monomer could be observed over the entire conversion
range with complex 1 as the initiator (Figure 2). All these
features imply some living characters of the polymerization.
FIGURE 4 Plot of lnkapp versus ln[Al]0 for e2CL polymerization
using complex 1 as initiator (in toluene, 35 ꢀC, [e2CL]0 5 4 mol/
L; slope 5 1.09).
To have a better understanding on the polymerization pro-
cess initiated by these aluminum complexes, the kinetics of
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
tive enthalpy and negative entropy imply the endothermic
nature of the polymerization.
To gain some insight into how these complexes, acting as
single component catalysts, initiated the ROP of e-CL and the
role played by the bidentate pyrrolidinyl-anilido ligands dur-
ing the polymerization, the NMR scale reaction of complex 6
with e-CL ([CL]0:[6]0 5 9) in C6D6 was carried out (see Sup-
porting Information). In the 1H NMR spectrum measured at
ꢀ
15 C, the resonances attributable to the pyrrolidinyl-anilido
ligand as well as Al-CH3 groups could still be identified
unambiguously, which however are slightly different from
those of complex 6. For instance, one multiplet assignable to
NCH2 (pyrrolidinyl) is displayed at 2.84 ppm, instead of two
multiple signals at 2.84 and 2.72 ppm for complex 6.
Besides, a tiny, broad signal at 5.11 ppm as well as one mul-
tiplet at 2.70 ppm in stoichiometric ratio could be observed.
Upon slightly warming the reaction mixture at 25 ꢀC for
about 30 min, the signals at 5.11 and 2.70 ppm became
apparent and the gradual oligomerization occurred. After
standing at 45 ꢀC overnight, the monomer was almost con-
sumed and the species represented by the signals at 5.11
and 2.70 ppm became the only identifiable species. The sig-
nals of aluminum methyl groups at the high field region
were still observable (see Supporting Information). Based on
these features, we suggest that at a relatively low tempera-
ture the treatment of complex 6 with excess e2CL leads to
the formation of monomer-coordinated species, where the
nitrogen atom of pyrrolidinyl may dissociate from the metal
center. The minor species formed upon slightly warming is
the monomer-inserted active species, which initiate the poly-
merization of e2CL. The significant downfield shift of ArCH2
proton signal from 4.15 ppm (complex 6) to 5.11 ppm dur-
ing this process is likely due to the insertion of the monomer
into the Al-N (amido) bond (see Supporting Information).
FIGURE 5 Plots of ln([e2CL]0/[e2CL]t) versus time for e2CL
polymerization using complex 1 as initiator at different temper-
atures (in toluene, [e2CL]0 5 4 mol/L, [e2CL]0/[Al]0 5 100. ᭿: T
535 ꢀC, kapp 5 1.97 3 1024 6 6.33 3 1026 s21
;
: T 5 45 ꢀC,
•
k
app 5 0.7553 1023 6 0.192 3 1024 s21; ~: T 5 55 ꢀC, kapp 5 1.60
3
1023 6 1.50
3
1024 s21
;
!: T 5 65 ꢀC,
kapp 5 3.40
3
1023 6 2.88 3 1024 s21).
2d½e2CLꢃt=dt5kapp½e2CLꢃt
2d½e2CLꢃt=dt5kp½Alꢃ½e2CLꢃt
(1)
(2)
Using complex 1 as the initiator, conversions of e2CL in tolu-
ene at various temperatures of 35, 45, 55, and 65 ꢀC with
fixed monomer and initiator concentrations were monitored
via 1H NMR spectroscopy. By plotting ln([e 2 CL]0/[e 2 CL]t)
versus time (t) (Figure 5), kapp values at different temperatures
could be obtained and therefore the kp values according to
kapp 5 kp[Al]0. Based on the Eyring equation, the thermody-
namics curve of ln(kp/T) vs. 1/T was plotted (Figure 6). From
the slope and intercept of the curve, the transition enthalpy
¼
¼
DH 5 78.11 6 7.43
KJ/mol
and
entropy
DS 5
End-group analysis of a typical e-CL oligomer obtained by
using complex 1 as the initiator was further performed via
225.57 6 2.53 J/(molꢁK) could be obtained, which then
enabled the calculation of the transition activation Gibbs free
1
1H NMR and ESI-TOF MS methods. In the H NMR spectrum
¼
¼
¼
energy, DG 5 DH 2 TDS 5 85.72 kJ/mol at 298 K. The posi-
of the oligomer, except for the typical signals of polymer
main chain, signals belonging to the pyrrolidinyl-anilido
ligand could be clearly identified (Figure 7), indicating that
the polymer may have the pyrrolidinyl-anilido ligand as one
of the end groups. This result is further confirmed by the
ESI-TOF MS (see Supporting Information), where a series of
peaks systematically ended with the same group of ligand L1
are displayed.
All the results suggest that the bidentate pyrrolidinyl-anilido
ligands in these aluminum dimethyl complexes do act as ini-
tiating groups to initiate the ROP of e2CL in a living manner.
It is well-known that amido groups are relatively poor in ini-
tiation when exposed to the catalytic ROP of cyclic esters.
Mu and coworker36 reported that, in the absence of alcohol,
dinuclear aluminum dimethyl complexes supported by
aldimine-anilido ligands were inactive toward the polymer-
ization of cyclic esters. Thibault and Fontaine37 found that,
without the addition of alcohol, the ROP of e-CL catalyzed by
FIGURE 6 Plot of In(kp/T) versus 1/T for e-CL polymerization
using complex 1 as initiator.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 7 The 1H NMR spectra (CDCl3, 400 MHz) of (a) the oligomer of e2CL obtained using complex 1 as initiator ([e2CL]/[1] 5 30,
in toluene, 35 ꢀC); (b) the proligand L1H (*, the signal of dichloromethane).
aluminum methyl complexes with phosphine/thiol-amido
ligands was badly controlled and no clear initiation mecha-
nism could be drawn. The aluminum complexes reported in
this work are rare examples that could initiate the ROP of
cyclic esters with living features in addition to our previous
report.30
ROP of rac-LA, while complexes 5 and 6 with a meta- or
para-chloro group are more active than the rest complexes.
Therefore, the same activity order of 5 (m-Cl) > 6 (p-Cl) > 1
(H) > 3 (p-iPr) > 4 (o-Cl) > 2 (o-Me) is observed for com-
plexes 1–6 toward rac-LA polymerization. In comparison to
the previously reported piperidinyl-anilino aluminum com-
plexes,30 complexes 1–6 are less active, probably due to the
stronger nucleophilicity of nitrogen atom of pyrrolidinyl ring.
The most active complex 5 could convert 500 equiv. of rac-
LA to high conversion of 97% within 42 h at low initiator
concentration ([5]0 5 0.005 mol/L, 85 ꢀC in toluene). How-
ever, due to the sharply increased viscosity of the reaction
mixture, the transesterification reactions became significant
which led to slight broadening of the molecular distribution
(Mn 5 6.31 3 104 g/mol, Mw/Mn 5 1.47). Moreover, polyl-
actides obtained by complexes 1–6 possess atactic
Based on the above results, a coordination-insertion mecha-
nism could be assumed for e-CL polymerization initiated by
these aluminum complexes (Scheme 2). e2CL monomer
coordinates to the aluminum center through the carbonyl
oxygen to form a four-coordinated aluminum intermediate,
where the pyrrolidinyl-anilido ligand is ligated monoden-
tately. Subsequently, the amido-nitrogen atom of the ligand
attacks to the carbonyl carbon of the coordinated monomer.
While the acyl-oxygen bond ruptures, the oxygen atom coor-
dinates to the aluminum metal center to produce active alu-
minum alkoxide species, which are in a dimeric form.38 The
ROP reaction is promoted by the continued insertion of
e2CL monomer into the active aluminum-alkoxo bonds until
being quenched by wet solvent.
ROP of rac-LA
All aluminum complexes 1–6 were also found to be efficient
initiators for the ROP of rac-LA. As shown in Table 2, high
monomer conversions could be obtained within 15–24 h
when the polymerizations were carried out in toluene at 65
ꢀC. The resultant polymers possess narrow molecular weight
distributions (Mw/Mn 5 1.09–1.26), which are slightly
broader than those obtained from e2CL polymerization. Sim-
ilar to the results of e2CL polymerization, complexes 2 and
4 with an ortho-substituent on the anilino moiety of the
ligand display relatively low catalytic activities toward the
SCHEME 2 Proposed polymerization mechanism of e2CL initi-
ated by aluminum amido complexes.
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 2 The ROP of rac-LA Initiated by Complexes 1–6a
Conv. (%)b
Mn, calcd. (3104)
Mn, NMR (3104)
Mn (3104)
Mw/Mn
Pm
c
d
e
e
f
0
Run
Cat.
1 (H)
Time (h)
1
2
3
4
5
6
15
24
15
24
15
15
82
88
78
92
88
84
1.18
1.27
1.12
1.32
1.27
1.21
1.30
1.29
1.20
1.27
1.21
1.30
1.27
1.18
1.09
1.56
1.23
1.14
1.11
1.15
1.09
1.10
1.26
1.10
0.51
2 (o-Me)
3 (p-iPr)
4 (o-Cl)
5 (m-Cl)
6 (p-Cl)
0.50
0.51
0.50
a
e
[rac-LA]0 5 1.0 M, [rac-LA]0/[Al]0 5 100, in toluene, 65 ꢀC.
Determined by the integration ratio of the methine protons in mono-
The number average molecular weight (Mn) and molecular weight dis-
tribution (Mw/Mn) were determined by a gel permeation chromatograph
in THF at 40 ꢀC, using narrowly distributed polystyrenes as standards,
Mn0 5 0.58 Mn.
b
mer and polymer in CDCl3.
c
M
n,calcd. 5 ([rac-LA]0/[Al]0) 3 144.14 3 conv. (%)1 mass of the free
f
ligand.
d
Pm is the probability of forming a new m-dyad, determined by homo-
Determined by the integration ratio of the methylene protons in poly-
nuclear decoupled 1H NMR spectroscopy.
mer main chain versus the characteristic resonance of the amido
ligand.
microstructures, as indicated by homonuclear decoupled 1H
NMR spectroscopic analyses.
became dominant. The two closely located doublets at 5.09
ppm could be assigned to ArCH2 protons of the ligand, which
were in a stoichiometric ratio with the four quartets at 4.78,
4.64, 4.61, and 4.57 ppm (4:1:1:1:1) assignable to the
methine protons of the inserted lactide monomers [Figure
8(a)]. Three singlets at 20.16, 20.25, and 20.40 ppm
(3:6:3) accounting for AlACH3 protons could also be
observed. Meanwhile, the signals of NCH2 (pyrrolidinyl) obvi-
ously up-field shifted to 2.70 and 2.55 ppm (see Supporting
Information), implying the dissociation of the neutral nitro-
gen donor from the metal center. All these features indicated
the insertion of lactide monomer into AlAN (amido) bond.
Two diastereomeric, lactide-inserted aluminum species in
dimeric form were generated, as reported previously by
Lewinski’s group.38 To further verify the dimeric nature of
the monomer-inserted aluminum species, the NMR reaction
of complex 6 with 2 equiv. of L-LA under the same
For a comparison purpose, the NMR scale reactions of com-
plex 6 with 3 equiv. of rac-LA in C6D6 at different tempera-
tures were also monitored (see Supporting Information). As
displayed in the 1H NMR spectrum measured at 5 ꢀC, the
resonances belonging to complex 6 were nearly unchanged
except that the two multiplets of NCH2 (pyrrolidinyl) at 2.84
and 2.72 ppm were somewhat coalescent. This implies that
the pyrrolidinyl arm might show some hemilabile coordina-
tion effect in the presence of LA monomer. After standing at
ꢀ
15 C for 30 min, the signals of NCH2 (pyrrolidinyl) became
broader, and some small signals represented by peaks at
5.09, 4.78, 4.64–4.57 ppm appeared. When the mixture was
kept at 15 ꢀC overnight, the signals attributable to complex
6 almost disappeared and the above mentioned signals
FIGURE 8 Part of 1H NMR spectra (C6D6, 400 MHz) of (a) the reaction mixture of complex 6 and rac-LA ([6]0/[rac-LA]0 5 1: 3) at 15
ꢀC; (b) the reaction mixture of complex 6 and L-LA ([6]0/[L-LA]0 5 1:2) at 15 ꢀC.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
9

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
CONCLUSIONS
In conclusion, a series of aluminum dimethyl complexes
bearing bidentate pyrrolidinyl-anilino ligands have been pre-
pared and fully characterized. In the absence of a coinitiator,
all the reported complexes were found to catalyze the ROPs
of e2CL and rac-LA in a well-controlled manner, producing
polyesters with quite narrow molecular weight distributions
(in most cases, Mw/Mn 5 1.04–1.15). The catalytic activities
of these aluminum complexes were affected greatly by the
steric effect of the ortho- substituent of the anilido moiety in
the ligand. The polymerization of e2CL showed a first-order
dependence on the concentrations of monomer and initiator.
Polymers end-capped by the whole fragment of the
pyrrolidinyl-anilino ligand were produced, which gave strong
supports that the polymerization was initiated by a mono-
mer insertion into the AlAN (amido) bond via a coordina-
tion–insertion mechanism. Based on the 1H NMR scale
oligomerizations, it is suggested that the insertion of the first
lactide monomer might be easier than that of e2CL mono-
mer. The formation of dimeric aluminum species in the for-
mer case where the last lactyl unit is bis-chelating to
aluminum center through alkoxy and carbonyl oxygen atoms
should be responsible for the lower reactivity of lactide
toward the catalytic polymerization.
SCHEME 3 The reaction of complex 6 with rac-LA at ambient
temperature (the structure of S,S-isomer is not shown).
conditions was monitored. As expected, only two closely
located quartets at 4.64, 4.61 ppm and one singlet at 20.25
ppm were observed in the related regions [Figure 8(b)]. It is
clear that these signals belong to the racemic R,R-/S,S-iso-
mers and the disappeared signals belong to the R,S-isomer
as shown in Scheme 3. The two diastereomers obtained
from the reaction of complex 6 with rac-LA was roughly in a
2:3 molar ratio (R,S- isomer vs. R,R-/S,S-isomers), implying
the preferable production of the R,R/S,S-isomers.
When up to about 6 equiv. of L-LA was added, except for the
downfield shift of methine resonance of lactide monomer
from 3.78 to 3.95 ppm, the other signals almost remained
the same (see Supporting Information), indicating no further
insertion of lactide monomer into the newly formed AlAO
(lactate) bonds of the dimeric aluminum species. The poly-
ACKNOWLEDGMENTS
ꢀ
merization was completed when warmed at 75 C overnight.
Financial supports from the National Natural Science Founda-
tion of China (NNSFC, 21074032 and 21274041) and the Fun-
damental Research Funds for the Central Universities
(WK1214048 and WD1113011) are gratefully acknowledged.
The signals accounting for the PLA main chain could be
observed at 5.0 and 1.3 ppm in the 1H NMR spectrum. The
resonances assignable to the pyrrolidinyl-anilino ligand as
well as AlACH3 groups although varied somewhat but still
could be identified unambiguously, demonstrating the basi-
cally constant nature of the active centers during the
polymerization.
REFERENCES AND NOTES
1 S. Dagorne, D. A. Atwood, Chem. Rev. 2008, 108, 4037–4071.
According to the literature reports,22,27,39,40 it is widely
accepted that in comparison with e2CL, lactide is in general
less active toward the catalytic ROP, that is, longer reaction
time or higher polymerization temperature is required for
equal amount of lactide to be polymerized. It is also the case
in this work. From above NMR studies, it was however found
that at a relatively low temperature the insertion of lactide
into the AlAN (amido) bond already occurred, while under
the same conditions e2CL monomer only coordinated to the
metal center without apparent ring-opening reaction. Based
2 N. E. Kamber, W. Jeong, R. M. Waymouth, R. C. Pratt, B. G.
G. Lohmeijer, J. L. Hedrick, Chem. Rev. 2007, 107, 5813–5840.
3 M. O. Tina, G. W. Coates, J. Am. Chem. Soc. 1999, 121,
4072–4073.
4 T. M. Ovitt, G. W. Coates, J. Polym. Sci. Part A: Polym.
Chem. 2000, 38, 4686–4692.
5 T. M. Ovitt, G. W. Coates, J. Am. Chem. Soc. 2002, 124,
1316–1326.
6 M. H. Chisholm, J. C. Gallucci, K. T. Quisenberry, Z. Zhou,
Inorg. Chem. 2008, 47, 2613–2624.
7 M. H. Chisholm, N. J. Patmore, Z. P. Zhou, Chem. Commun.
2005, 127–129.
on the structural characterization of
a lactide-inserted,
dimeric aluminum species reported by Lewinski’s group,38 it
is therefore suggested that, although the insertion of the first
lactide monomer might be easier than that of e2CL mono-
mer, the bis-chelating mode of the lactyl unit adjacent to alu-
minum center in the case of lactide insertion may stabilize
the active center to a certain degree, which raises the energy
barrier of further insertion step considerably. These results
may also explain well why a random copolymerization of
e2CL and lactide could not be realized by most of the metal-
lic initiators.
8 D. Chakraborty, E. Y.-X. Chen, Organometallics 2002, 21,
1438–1442.
9 N. Nomura, A. Akita, R. Ishii, M. Mizuno, J. Am. Chem. Soc.
2010, 132, 1750–1751.
10 C. T. Chen, H. J. Weng, M. T. Chen, C. A. Huang, K. F. Peng,
Eur. J. Inorg. Chem. 2009, 14, 2129–2135.
11 A. Gao, Y. Mu, J. Zhang, W. Yao, Eur. J. Inorg. Chem. 2009,
24, 3613–3621.
12 M. Shen, W. Zhang, K. Nomura, W. H. Sun, Dalton Trans.
2009, 41, 9000–9009.
10
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
13 S. Gong, H. Ma, Dalton Trans. 2008, 25, 3345–3357.
27 D. Ilaria, L. Marina, M. Mina, M. Stefano, P. Claudio, J.
Polym. Sci. Part A: Polym. Chem. 2014, 52, 49–60.
14 A. Otero, A. Lara, J. Fernandez, C. Alonso, J. A. Castro, I.
Marquez, L. F. Sanchez, A. M. Rodriguez, J. C. Garcia, Organo-
metallics 2011, 30, 1507–1522.
28 F. Qian, K. Liu, H. Ma, Dalton Trans. 2010, 39, 8071–8083.
29 F. Qian, K. Liu, H. Ma, Chin. J. Catal. 2011, 32, 189–1195.
30 J. Liu., H. Ma, Dalton Trans. 2014, 43, 9098–9110.
15 E. Kober, Z. Janas, T. Nerkowski, L. B. Jerzykiewicz, Dalton
Trans. 2013, 42, 10847–10854.
31 R. L. Bentley, H. Suschitzky, J. Chem. Soc. Perkin Trans. 1
1976, 16, 1725–1734.
16 C. Li, D. Liu, B. Ko, Dalton Trans. 2013, 42, 11488–11496.
17 S. L. Hancock, M. F. Mahon, M. D. Jones, Dalton Trans.
2013, 42, 9279–9285.
32 G. M. Sheldrick, Acta Crystallogr. Sect.
1122122.
A 2008, 64,
18 H. Han, Y. Liu, J. Liu, K. Nomura, Y. Li, Dalton Trans. 2013,
42, 12346–12353.
33 SADABS, Bruker Nonius Area Detector Scaling and Absorp-
tion Correction-V2.05, Bruker AXS Inc.: Madison, WI, 1996.
19 D. Li, Y. Peng, C. Geng, K. Liu, D. Kong, Dalton Trans. 2013,
42, 11295–11303.
34 C. K. Johnson, ORTEP-III: A FORTRAN Thermal Ellipsoid
Plot Program for Crystal Structure Illustrations, Report
ORNL5138, Oak Ridge National Laboratory, Oak Ridge, TN,
USA, 1976.
20 M. H. Chisholm, C. Lin, J. C. Gallucci, B. Ko, Dalton Trans.
2003, 406–412.
21 S. Qiao, W.-A. Ma, Z.-X. Wang, J. Organometal. Chem.
2011, 696, 2746–2753.
35 I. Barakat, P. Dubois, R. Jerome, P. Teyssie, J. Polym. Sci.
A: Polym. Chem. 1993, 31, 505–514.
22 D. Li, Y. Peng, C. Geng, K. Liu, D. Kong, Dalton Trans. 2013,
42, 11295–11303.
36 W. Yao, Y. Mu, A. Gao, W. Gao, L. Ye, Dalton Trans. 2008,
3199–3206.
23 F. Hild, L. Brelot, S. Dagorne, Organometallics 2011, 30,
37 M. H. Thibault, F. G. Fontaine, Dalton Trans. 2010, 39, 5688–
5457–5462.
5697.
24 C. Bakewell, R. H. Platel, S. K. Cary, S. M. Hubbard, J. M.
Roaf, A. C. Levine, A. J. P. White, N. J. Long, M. Haaf, C. K.
Williams, Organometallics 2012, 31, 4729–4736.
38 J. Lewinski, P. Horeglad, K. Wojcik, I. Justyniak, Organome-
tallics 2005, 24, 4588–4593.
39 S. Dagorne, F. L. Bideau, R. Welter, S. Bellemin-Laponnaz,
A. Maisse-Franc¸ois, Chem. Eur. J. 2007, 13, 3202–3217.
25 F. Hild, L. Brelot, S. Dagorne, Organometallics 2011, 30,
5457–5462.
40 F. Hild, N. Neehaul, F. Bier, M. Wirsum, C. Gourlaouen, S.
Dagorne, Organometallics 2013, 32, 587–598.
26 X.-F. Yua, Z.-X. Wang, Dalton Trans. 2013, 42, 3860–3868.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 00, 000–000
11