Molecularly defined caprolactone oligomers and polymers: Synthesis and characterization

Article abstract of DOI:10.1021/ja077149w

The synthesis of molecularly defined ε-caprolactone oligomers and polymers up to the 64-mer, via an exponential growth strategy, is described. By careful selection of orthogonal protecting groups, t-butyldimethylsilyl (TBDMS) ether for the hydroxyl group and benzyl (Bn) ester for the carboxylic acid group, a highly efficient synthetic strategy was developed with yields for both deprotection steps being essentially quantitative and for the coupling reactions using 1,3-dicyclohexylcarbodiimide (DCC), yields of 80-95% were obtained even at high molecular weights. This allows monodisperse dimers, tetramers, octamers, 16-mers, 32-mers and 64-mers to be prepared in gram quantities and fully characterized using mass spectroscopy, size exclusion chromatography (SEC), and IR and NMR spectroscopy. Thermal and physical properties were measured using thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), atomic force microscopy (AFM), and small-angle X-ray scattering (SAXS). These results conclusively show a distinct structure/property relationship with a close correlation between the number of repeat units and physical properties. In addition, a number of marked differences were observed on comparison with the parent poly(caprolactone) polymer.

Full text of DOI:10.1021/ja077149w

Published on Web 01/12/2008
Molecularly Defined Caprolactone Oligomers and Polymers:
Synthesis and Characterization
Kenichi Takizawa,†,‡ Chuanbing Tang,† and Craig J. Hawker*,†
Materials Research Laboratory and Departments of Materials, Chemistry and Biochemistry,
UniVersity of California, Santa Barbara, California 93106, and Mitsubishi Chemical Group
Science and Technology Research Center, Inc., 1000 Kamoshida-cho, Aoba-ku,
Yokohama, 227-8502, Japan
Received September 14, 2007; E-mail: hawker@mrl.ucsb.edu
Abstract: The synthesis of molecularly defined ꢀ-caprolactone oligomers and polymers up to the 64-mer,
via an exponential growth strategy, is described. By careful selection of orthogonal protecting groups,
t-butyldimethylsilyl (TBDMS) ether for the hydroxyl group and benzyl (Bn) ester for the carboxylic acid
group, a highly efficient synthetic strategy was developed with yields for both deprotection steps being
essentially quantitative and for the coupling reactions using 1,3-dicyclohexylcarbodiimide (DCC), yields of
80-95% were obtained even at high molecular weights. This allows monodisperse dimers, tetramers,
octamers, 16-mers, 32-mers and 64-mers to be prepared in gram quantities and fully characterized using
mass spectroscopy, size exclusion chromatography (SEC), and IR and NMR spectroscopy. Thermal and
physical properties were measured using thermogravimetric analysis (TGA), differential scanning calorimetry
(DSC), atomic force microscopy (AFM), and small-angle X-ray scattering (SAXS). These results conclusively
show a distinct structure/property relationship with a close correlation between the number of repeat units
and physical properties. In addition, a number of marked differences were observed on comparison with
the parent poly(caprolactone) polymer.
Introduction
Traditionally, poly(caprolactone) has been prepared via the
ring opening polymerization of ꢀ-caprolactone using a catalyst
such as stannous octanoate with significant effort in recent years
being devoted to living processes,3 complex macromolecular
architectures,4 and metal-free systems.5 The resulting polymer
is semicrystalline with a Tm of 58-60 °C and is widely used as
a reactive block in the preparation of copolymers as well as in
blends with other commodity polymers. Although there is a
growing academic and industrial interest in poly(caprolactone)
A wide range of biocompatible, biodegradable, and resorbable
synthetic polymers has been developed for bioengineering and
biomedical applications with poly(caprolactone) (PCl) being one
of the most widely studied synthetic materials.1 The leading
role of PCl as a biomaterial is further evidenced by its approval
by the Food and Drug Administration (FDA) for use in the
human body as a drug delivery device, suture (sold under the
brand name Monocryl), adhesion barrier, and is being investi-
gated as a scaffold for tissue engineering. Over time, the ester
groups in the backbone of PCL undergo hydrolysis to the
constituent hydroxy acid which is eliminated by general
metabolic pathways.2 Unfortunately, most implanted materials
are seen by the body as foreign objects, which leads to a range
of undesirable immune responses. The recent negative publicity
regarding drug-covered stents is a perfect example of the
significant unmet medical need for both the rational design of
functionalized synthetic biomaterials as well as a complete
understanding of structure/property relationships.
(3) (a) Wisse, E.; Renken, R. A. E.; Roosma, J. R.; Palmans, A. R. A.; Meijer,
E. W. Biomacromolecules 2007, 8, 2739-2745. (b) Parrish, B.; Breiten-
kamp, R. B.; Emrick, T. J. Am. Chem. Soc. 2005, 127, 7404-7410. (c)
O’Keefe, B. J.; Breyfogle, L. E.; Hillmyer, M. A.; Tolman, W. B. J. Am.
Chem. Soc. 2002, 124, 4384-4393. (d) Trollsas, M.; Atthoff, B.; Claesson,
H.; Hedrick, J. L. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 1174-
1188. (e) Hedrick, J. L.; Magbitang, T.; Connor, E. F.; Glauser, T.; Volksen,
W.; Hawker, C. J.; Lee, V. Y.; Miller, R. D. Chem.-Eur. J. 2002, 8, 3308-
3319. (f) Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 2002, 124, 1316-
1326. (g) Van, Horn, B. A.; Wooley, K. L. Soft Matter 2007, 3, 1032-
1040.
(4) (a) Barrera, D. A.; Zylstra, E.; Lansbury, P. T.; Langer, R. J. J. Am. Chem.
Soc. 1993, 115, 11010. (b) Tian, D.; Dubois, Ph.; Grandfils, C.; Je´roˆme, J.
Macromolecules 1997, 30, 406. (c) Trollsås, M.; Lee, V. Y.; Mecerreyes,
D.; Lo¨wenhielm, P.; Mo¨ller, M.; Miller, R. D.; Hedrick, J. L. Macromol-
ecules 2000, 33, 4619. (d) Lenoir, S.; Riva, R.; Detrembleur, C.; Je´roˆme,
R.; Lecomte, P. Macromolecules 2004, 37, 4055. (e) Zhang, Q.; Remsen,
E. E.; Wooley, K. L. J. Am. Chem. Soc. 2000, 122, 3642-3651. (f) Moller,
M.; Nederberg, F.; Lim, L. S.; Kange, R.; Hawker, C. J.; Hedrick, J. L.;
Gu, Y. D.; Shah, R.; Abbott, N. L. J. Polym. Sci., Part A: Polym. Chem.
2001, 39, 3529-3538.
† University of California.
‡ Mitsubishi.
(1) (a) Ha, C.-S.; Gardella, J. A., Jr. Chem. ReV. 2005, 105, 4205-4232. (b)
Grayson, S. M.; Frechet, J. M. J. Chem. ReV. 2001, 101, 3819-3868. (c)
Alteheld, A.; Feng, Y.; Kelch, S.; Lendlein, A. Angew. Chem., Int. Ed.
2005, 44, 1188-1192. (d) Kopecek, K.; Ulbrich, K. Prog. Polym. Sci. 1993,
9, 1.
(5) (a) Coulembier, O.; Kiesewetter, M. K.; Mason, A.; Dubois, P.; Hedrick,
J. L.; Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 2007, 46, 4719-
4721. (b) Coulembier, O.; Degee, P.; Hedrick, J. L. Prog. Polym. Sci. 2006,
31, 723-747. (c) Cordova, A.; Hult, A.; Hult, K.; Ihre, H.; Iversen, T.;
Malmstrom, E. J. Am. Chem. Soc. 1998, 120, 13521-13522. (d) Chen, B.;
Miller, M. E.; Gross, R. A. Langmuir 2007, 23, 6467-6474. (e) Marcilla,
R.; de Geus, M.; Mecerreyes, D.; Duxbury, C. J.; Koning, C. E.; Heise, A.
Eur. Polym. J. 2006, 42, 1215-1221.
(2) (a) Echeverria, E.; Jimenez, J. Surgery 1970, 131, 1. (b) Alexander, H.;
Parsons, J. R.; Straucher, I. D.; Corcoran, S. F.; Gona, O.; Mayott, C.;
Weiss, A. B. Orthop. ReV. 1981, 10, 41. (c) Sawhney, A. S.; Pathak, C. P.;
Hubbell, J. A. Macromolecules 1993, 26, 581.
9
1718
J. AM. CHEM. SOC. 2008, 130, 1718-1726
10.1021/ja077149w CCC: $40.75 © 2008 American Chemical Society

Caprolactone Oligomers and Polymers
A R T I C L E S
Matrix assisted laser desorption/ionization (MALDI-TOF-MS) was
carried out at room temperature on a Dynamo Thermo machine using
dithranol in THF as the matrix and sodium trifluoroacetate (NaOTf) in
THF as the cation agent. Size exclusion chromatography (SEC) was
carried out at room temperature on a Waters Alliance HPLC System
(Waters 2695 Separation Module) connected to Waters Styragel HR
columns (HR 0.5, 2, and 4) using THF as eluent (flow rate: 1 mL/
min). A Waters 2414 differential refractometer and a 2996 photodiode
array detector were employed. Preparative GPC was carried out at room
temperature on a Waters 1525 Binary HPLC connected to a Waters
2414 differential refractometer and Waters Styragel columns (Ul-
trastyragel 100, 10-3, and 10-4 Å) using THF as eluent (flow rate: 6
mL/min). The molecular weights of the polymers were calculated
relative to linear polystyrene standards. Thermogravimetric analysis
was conducted using Mettler TGA/DTA 851e under N2 atmosphere.
Differential scanning calorimetry (DSC) measurements were performed
with a TA Instruments DCS 2920 and a ramp rate of 5 degrees per
minute with data collected during third cycle in the selected temperature
ranges. Calibrations were made using indium as a standard for both
temperature transitions and the heats of fusion. Melting transition
temperatures (Tm) were determined as the peak maxima of the transition.
Small-angle X-ray scattering (Ultra-SAXS (X-ray Source; Fine focus
(0.2 mm) Rigaku rotationg anode generator, Wavelength; 1.54 Å,
Sample to Detector Distance; 172.5 cm, Interface; Bruker SAXS
software and SPEC) and Intermediate-SAXS (X-ray Source; 18kW
Rigaku rotationg anode generator, Wavelength; 1.54 Å, Sample to
Detector Dintance; 75.8 cm, Interface; SPEC)) were carried out using
quartz capillary cell. Tapping mode AFM experiments were carried
out using a Multimode Nanoscope III system equipped with a J-type
vertical engage scanner (Digital Instruments, Santa Barbara, CA). The
measurements were performed under ambient atmosphere using com-
mercial Si cantilevers with a nominal spring constant and resonance
frequency respectively equal to 48 N/m and 190 kHz (ACL, Applied
Nanostructures, Santa Clara, CA).
and biodegradable polyesters in general, there have been no
studies on the synthesis of well-defined oligomers based on
poly(caprolactone), poly(lactide), etc. This is unfortunate bec-
uase the availability of precisely defined oligomers6 would
enable a wide range of structure property studies to fully
understand, predict, and tune the degradation rate, crystal
structure, self-assembly, and performance of these materials in
a variety of applications.
The significant benefit afforded material design by the
development of synthetic approaches to well-defined oligomers
arises from a fundamental understanding of the parent polymer
through the availability and detailed study of dimers, tetramers,
octamers, etc. Previous work has demonstrated the power of
this approach with molecular-defined oligomers being produced
in both cellular systems7 as well as in stepwise synthetic
strategies.8 Of particular note is the work of Brooke,8b,c who
has prepared a series of oligomers related to commercially
important polymers such as Nylon 6,6 and poly(ethylene
terephthalate). In addition, Moore9 has demonstrated the change
in conformation and associated physical/chemical properties of
poly(phenyleneethynylenes) with oligomer length. Similarly,
Meijer10 has shown that well-defined π-conjugated oligomers,
such as sexithiophene can play an important role in the field of
organic electronics due to their precise chemical structure and
conjugation length which gives rise to well-defined electronic
properties and tunable intermolecular solid state organization.
These and other studies demonstrating the syntheses of molec-
ular rods of precise length11 have illustrated the dramatic insights
that can be gained from the study of well-defined oligomers
and polymeric derivatives of both scientifically and industrially
important macromolecules. Herein, we report the development
of a synthetic strategy for the synthesis of well-defined polyester
oligomers and demonstrate the preparation of a series of poly-
(caprolactone) derivatives up to the 64-mer. The physical and
structural properties of these essentially single molecule species
allow a fundamental insight into the physical and structural
properties of the widely studied parent polymer.
General Procedure of Coupling Reaction: Synthesis of Dimer
4. A mixture of the protected acid 2 (20.50 g, 83.19 mmol), protected
alcohol 3 (17.74 g, 79.81 mmol), 1,3-dicyclohexylcarbodiimide (DCC)
(19.29 g, 93.49 mmol), and 4-(dimethylamino)pyridine (DMAP) (11.17
g, 91.43 mmol) was dissolved in CH2Cl2 (200 mL) and stirred overnight
at room temperature. The resulting mixture was filtered and the filtrate
poured into a separatory funnel and washed with 200 mL of sat. CuSO4
and 200 mL of H2O. The organic layer was then dried over MgSO4,
filtered, and concentrated under reduced pressure. The crude product
was purified via flash chromatography using 3:1 hexanes/ethyl acetate
as eluent to give the doubly protected dimer, 4, as a clear, colorless oil
(29.60 g, 81% yield); 1H NMR (CDCl3): δ 7.33 (m, Ar, 5H), 5.10 (s,
CO2CH2Ph, 2H), 4.03 (t, J ) 6.5 Hz, CO2CH2CH2, 2H), 3.58 (t, J )
6.3 Hz, CH2OSi, 2H), 2.31 (t, t, J ) 7.6 Hz, 7.6 Hz, CH2CH2CO2,
CH2CH2CO2, 4H), 1.50 (m, SiOCH2[CH2]3CH2CO2CH2[CH2]3CH2CO2-
CH2Ph, 12H), 0.88 (s, (CH3)3CSi, 9H), 0.04 (s, (CH3)2Si, 6H). 13C NMR
(CDCl3): δ 173.82 (CO, 1C), 173.35 (CO, 1C), 136.15 (Ar, CH2-C,
1C) 128.64 (Ar-meta, 2C), 128.29 (Ar-ortho, para, 3C), 66.22
(CO2CH2Ph, 1C), 64.10 (CO2CH2CH2, 1C), 63.04 (SiOCH2, 1C), 34.40
(CH2CH2CO2, 1C), 34.20 (CH2CH2CO2, 1C), 32.56 (SiOCH2CH2CH2,
1C), 28.43 (CO2CH2CH2CH2, 1C), 26.06 ((CH3)3CSi, 3C), 25.61
(CH2CH2CH2, 1C), 25.55 (CH2CH2CH2, 1C), 24.90 (CH2CH2CH2, 1C),
24.66 (CH2CH2CH2, 1C), 18.43 ((CH3)3CSi, 1C), -5.18 ((CH3)2Si, 2C).
Mass Spec for C25H42O5Si+Na Calculated: 473.2699; Found
(M+Na)+: 473.2697.
Experimental Section
Materials and General Procedures. 4-(Dimethylamino)pyridinium
p-toluenesulfonate (DPTS) was synthesized according to a previously
reported procedure.12 All of the other chemicals and solvents were
purchased from Aldrich, of reagent grade, and used without further
purification. Analytical TLC was performed on commercial Merck
Plates coated with silica gel GF254 (0.24 mm thick). Silica gel for
flash column chromatography was Merck Kieselgel 60 (230-400 mesh,
ASTM). 1H NMR (200 MHz) and 13C NMR (100 MHz) measurements
were performed on a Bruker AC 200 spectrometer at room temperature.
(6) (a) Zhang, J.; Moore, J. S.; Xu, Z.; Aguirre, R. A. J. Am. Chem. Soc. 1992,
114, 2273. (b) Zhou, X. Z.; Shea, K. J. J. Am. Chem. Soc. 2000, 122,
11515-11516.
(7) Krejchi, M. T.; Atkins, E. D. T.; Waddon, A. J.; Fournier, M. J.; Mason,
T. L.; Tirrell, D. A. Science 1994, 265, 1427-1432.
(8) (a) Hawker, C. J.; Malmstrom, E. E.; Frank, C. W.; Kampf, J. P. J. Am.
Chem. Soc. 1997, 119, 9903. (b) Brooke, G. M.; MacBride, J. A. H.;
Mohammed, S.; Whiting, M. C. Polymer 2000, 41, 6457. (c) Brooke, G.
M.; Cameron, N. R.; MacBride, J. A. H.; Whiting, M. C. Polymer 2002,
43, 1139.
(9) Stone, M. T.; Heemstra, J. M.; Moore, J. S. Acc. Chem. Res. 2006, 39,
11-20.
General Procedure of Deprotection of Benzyl Ester (Hydrogena-
tion): Synthesis of the Acid-Functionalized Dimer, 5. Palladium on
activated carbon (10 wt %, 0.87 g) was added to a solution of 4 (8.28
g, 18.37 mmol) in ethyl acetate (100 mL), and the reaction mixture
was stirred overnight at room temperature under hydrogen. The resulting
mixture was filtered through celite, the celite washed with 100 mL of
(10) Leclere, P.; Surin, M.; Viville, P.; Lazzaroni, R.; Kilbinger, A. F. M.; Henze,
O.; Feast, W. J.; Cavallini, M.; Biscarini, F.; Schenning, A. P. H. J.; Meijer,
E. W. Chem. Mater. 2004, 16, 4452-4466.
(11) Gothard, C. M.; Rao, N. A.; Nowick, J. S. J. Am. Chem. Soc. 2007, 129,
7272-7273.
(12) (a) Moore, J. S.; Stupp, S. I. Macromolecules 1990, 23, 65. (b) Hawker,
C. J.; Fre´chet, J. M. J. J. Chem. Soc., Perkin Trans 1 1992, 2459-2469.
9
J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008 1719

A R T I C L E S
Scheme 1.
Takizawa et al.
Scheme 2.
hot MeOH, and the combined filtrate was concentrated under reduced
pressure. Drying under high vacuum then gave the monodeprotected
dimer, 5, as a clear, colorless oil (yield 13.12 g, 100%); 1H NMR
(CDCl3): δ 4.05 (t, J ) 6.5 Hz, CO2CH2CH2, 2H), 3.59 (t, J ) 6.3
Hz, CH2OSi, 2H), 2.32 (t, t, J ) 7.2 Hz, 7.2 Hz, CH2CH2CO2, CH2CH2-
CO2, 4H), 1.50 (m, SiOCH2[CH2]3CH2CO2CH2[CH2]3CH2CO2H, 13H)
0.87 (s, (CH3)3CSi, 9H), 0.02 (s, (CH3)2Si, 6H). 13C NMR (CDCl3): δ
179.68 (CO2H, 1C), 174.10 (CO2CH2, 1C), 64.21 (CO2CH2CH2, 1C),
63.20 (SiOCH2, 1C), 34.51 (CH2CH2CO2, 1C), 34.05 (CH2CH2CO2,
1C), 32.61 (SiOCH2CH2CH2, 1C), 28.48 (CO2CH2CH2CH2, 1C), 26.14
((CH3)3CSi, 3C), 25.61 (CH2CH2CH2, 2C), 24.98 (CH2CH2CH2, 1C),
24.46 (CH2CH2CH2, 1C), 18.53 ((CH3)3CSi, 1C), -5.11 ((CH3)2Si, 2C);
HRMS-EI (m/z): [M]+ calcd for C18H36O5Si, 360.2332; found,
360.2333.
1C), 32.12 (HOCH2CH2CH2, 1C), 28.13 (CO2CH2CH2CH2, 1C), 25.33
(CH2CH2CH2, 1C), 25.18 (CH2CH2CH2, 1C), 24.56 (CH2CH2CH2, 1C),
24.37 (CH2CH2CH2, 1C); HRMS-EI (m/z): [M]+ calcd for C18H28O5,
324.1936; found, 324.1935.
Results and Discussion
The starting monomer, 6-hydroxycaproic acid, 1, was ob-
tained in essentially quantitative yield by continuous extraction
of the base-catalyzed ring opening of commercially available
ꢀ-caprolactone. In designing the appropriate growth strategy for
preparation of well-defined oligomers from 1, the selection of
orthogonal protecting groups for the hydroxyl and carboxylic
acid groups was critical. After extensive comparison of various
protecting groups pairs under growth and deprotection condi-
tions, the most efficient and selective protecting groups were
found to be a t-butyldimethylsilyl (TBDMS) ether for the
hydroxyl group and a benzyl (Bn) ester for the carboxylic acid.
Therefore, the hydroxyl group of 1 was protected with t-
butyldimethylsilyl (TBDMS) ether,13,14 yielding 2, and the
carboxylic acid group was protected with a benzyl (Bn) ester,
yielding the monohydroxy derivative, 3 (Scheme 1).
Having identified orthogonal protecting groups, the next
challenge in this exponential growth strategy was to elucidate
a reliable connection strategy that would permit large oligomers
to be coupled in essentially quantitative yields without cleavage
of the hydroxyl or carboxylic acid protecting groups. Initial
studies using acyl halide or active ester derivatives of 2 proved
to be unsuccessful with either low yields or minor amounts of
deprotection of the TBDMS group being observed. In contrast,
General Procedure of Deprotection of TBDMS Protected Alcohol
(Desilylation): Synthesis of Hydroxyl-Terminated Dimer, 6. A
mixture of glacial acetic acid (4.41 g, 73.41 mmol) and tetrabutylam-
monium fluoride (TBAF) (66.29 g, 73.41 mmol, 1.0 M solution in THF)
was added to a solution of 4 (16.55 g, 36.72 mmol) in THF (75 mL).
The reaction mixture was stirred overnight at 50 °C and then poured
into a separatory funnel containing 300 mL of CH2Cl2 and 300 mL of
H2O. The organic layer was then washed with sat. NaHCO3 (2 × 200
mL), 5 wt % Citric acid (2 × 200 mL), and H2O (1 × 200 mL), dried
over MgSO4, filtered, and concentrated under reduced pressure. The
crude product was purified via flash chromatography using 1:1 hexanes/
ethyl acetate as eluent to give 6 as a clear, colorless oil (yield 11.45 g,
1
93% yield); H NMR (CDCl3): δ 7.34 (s, Ar, 5H), 5.11 (s, CO2CH2-
Ph, 2H), 4.05 (t, J ) 6.4 Hz, CO2CH2CH2, 2H), 3.60 (t, J ) 6.4 Hz,
CH2OH, 2H), 2,99 (b, OH, 1H), 2.33 (t, t, J ) 7.3 Hz, 7.3 Hz, CH2CH2-
CO2, CH2CH2CO2, 4H), 1.50 (m, HOCH2[CH2]3CH2CO2CH2[CH2]3-
CH2CO2CH2Ph, 12H). 13C NMR (CDCl3): δ 173.69 (CO, 1C), 173.22
(CO, 1C), 135.86 (Ar, CH2-C, 1C) 128.38 (Ar-meta, 2C), 128.00 (Ar-
ortho, para, 3C), 65.97 (CO2CH2Ph, 1C), 63.94 (CO2CH2CH2, 1C),
62.05 (HOCH2, 1C), 34.06 (CH2CH2CO2, 1C), 33.92 (CH2CH2CO2,
(13) Corey, E. J.; Venkateswarlu, A. J. Am. Chem. Soc. 1972, 94, 6190.
(14) Smith, A. B., III.; Ott, R. G. J. Am. Chem. Soc. 1996, 118, 13095.
9
1720 J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008

Caprolactone Oligomers and Polymers
A R T I C L E S
Figure 1. Molecular Structure of ꢀ-Caprolactone 64-mer, 19.
Figure 2. 1H NMR spectrum of doubly protected ꢀ-Caprolactone hexadecamer, 13.
examination of carbodiimide coupling proved to be successful
for low molecular weight oligomers using DCC and DMAP up
to the tetramer. For larger oligomers, yields were observed to
decrease slightly for the octamer and to decrease significantly
for larger molecules. As previously reported,12 a frequent side
reaction of carbodiimide esterifications is the formation of
unreactive N-acylureas (DCC-Adducts), which was shown to
be the cause of decreasing yields for larger oligomers. Replace-
Continuation of this strategy to the hexadecamer, 13, 32-mer,
16, and 64-mer, 19 was facilitated by the high yields obtained
for the deprotection steps, greater than 90% for the desilylation
and essentially quantitative for the debenzylation and especially
by the efficient coupling step which was 65% even for the
coupling of two 32-mers to give the 64-mer, 19. The structure
of the doubly protected 64-mer is shown in Figure 1 with the
molecular structure corresponding to a molecular formulas of
ment of DMAP with 4-(dimethylamino)pyridinium p-toluene-
sulfonate (DPTS) was found to suppress this side reaction, and
even at high molecular weights, only minor amounts (<2%)
were observed.
C
397H662O129Si and a molecular weight of 7522.
Given the precision of the synthetic approach for preparing
well-defined caprolactone oligomers and the wide applicability
of these materials in fundamental structure/property studies in
both biomaterial engineering and polymer physics, it was critical
to fully characterize and determine the purity of these materials.
A range of analytical tools were employed to evaluate the
purity and molecular weight of the oligomeric/polymeric
derivatives including 1H, 13C NMR, mass spectroscopy (ESI+/
TOF) combined with MALDI and SEC. As has been found
The repetitive chain growth strategy is illustrated in Scheme
2 and involves initial coupling of the carboxylic acid, 2, with
the hydroxyl derivative, 3, to give the protected dimer, 4. The
tetramer sample is then split and the benzyl group is removed
by hydrogenation, affording the desired carboxylic acid, 5,
whereas treatment with tetra-n-butyl ammonium fluoride (TBAF)
in the presence of acetic acid gives the monohydroxy derivative,
6. In both cases, the optimized reaction procedures resulted in
essentially quantitative yields of the monofunctional dimers.
Repetition of this coupling strategy gave the tetramer, 7,
deprotection of which results in the two corresponding mono-
functional tetramers, 8 and 9, and subsequently coupling gives
the octamer, 10 (Scheme 2).
1
previously with dendrimer syntheses, H NMR spectroscopy
proved to be very diagnostic in following the growth of the
polymer chain due to unique resonances for both chain ends
and the singular CH2 group of the terminal hydroxyl-caprolac-
tone unit. Figure 2 shows the 300 MHz H NMR spectrum of
the doubly protected hexadecamer, 13, and the resonances for
1
the single benzyl ester chain end can be easily identified at 5.1
9
J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008 1721

A R T I C L E S
Takizawa et al.
Figure 3. Molecular ions from the ESI-TOF mass spectral analysis of the hexadecamer, 13.
Figure 4. MALDI mass spectra for the caprolactone oligomers from the dimer, 4, to the 64-mer, 19.
was shown to be sixteen, as expected from the synthesis. On
deprotection, the absence of resonances for the respective chain
ends could be easily observed due to the disappearance of peaks
above 5.0 and below 1.0 for the backbone, which greatly aided
identification of unreacted starting materials during the depro-
tection reactions. Additionally, the small triplet at 3.65 ppm,
due to the unique CH2 group next to the TBDMSO-group, also
functions as a reference peak to monitor the length of the oligo-
(caprolactone) backbone and on deprotection to give the HO-
group, a shift to 3.60 ppm is observed. Similar trends were
observed in the 13C NMR spectra and further aided structural
identification of these materials.
The molecular weights and associated purity for all of the
caprolactone oligomers was investigated by both electrospray
as well as MALDI mass spectrometry. In all cases, molecular
ions for the desired oligomeric species were observed with little
or no contamination from failure sequences or chain end
impurities. For example, the ESI-TOF spectrum for the doubly
protected hexadecamer, 13, shows only molecular ions corre-
sponding to the hexadecamer as well the corresponding water,
Na+ and K+ adducts (Figure 3). High-resolution mass spectral
Figure 5. SEC traces for the caprolactone oligomers from the dimer, 4, to
the 64-mer, 19.
and 7.3 ppm, and the single TBDMS group at the other chain
end is similarly clearly distinguishable at 0.0 and 0.85 ppm.
Integration of these resonances and comparison with the
integration values for the caprolactone backbone allowed the
number of repeat units to be determined, which in this case
9
1722 J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008

Caprolactone Oligomers and Polymers
A R T I C L E S
Figure 6. DSC traces for the caprolactone oligomers from the dimer, 4, to the 64-mer, 19.
analysis of the M+Na+ ion was also found to correspond with
the anticipated molecular formulas. The high degree of purity
for all of the oligomers is also represented in the overlay of
MALDI mass spectra for the dimer, 4, to the 64-mer, 19,
showing a prominent molecular ion for all of the oligomers
(Figure 4).
behavior with the dimer, 4, being totally amorphous. In direct
contrast, the tetramer, 7, and octamer, 10, show sharp melting
transitions at increasing values of -20 and 32 °C. Significantly,
the behavior for the higher molecular weight oligomers and
polymers is dramatically different with the hexadecamer, 13,
having two distinct melting transitions at 38 and 49 °C,
indicating the presence of either two crystallite sizes, a crystal
to crystal transition, or two coexisting crystalline structures. This
recrystallization of smaller crystallites into larger structures has
previously been observed in semicrystalline aliphatic polycar-
bonates.17 Doubling the size of the oligomer to the 32-mer, 16,
resulting in merging of these two transitions and a single peak
is observed at 51 °C with a small shoulder at 53 °C, whereas
two distinct peaks are again observed for the 64-mer, 19, with
increased melting transitions of 54 and 58 °C.
To gain a more quantitative understanding of the purity of
these materials and the presence of lower molecular weight
impurities, size exclusion chromatography analysis of the
complete series of oligomers was performed and the results are
shown in Figure 5. As expected, the SEC traces are very narrow
and monomodal with a complete absence of higher molecular
weight impurities for all samples. For the dimer, 4, to octamer,
10, derivatives, no low molecular weight impurities are ob-
served; however, a small lower molecular weight peak is
observed for the hexadecamer, 13, 32-mer, 16, and 64-mer 19.
This lower molecular weight peak corresponded closely with
the pervious oligomer, that is, 32-mer impurity for the 64-mer
and integration showed that the level of impurity was 2% for
13 increasing to 4% for the 64-mer. In combination with the
mass and NMR spectroscopy results, it can be concluded that
all of the oligomers have a very high level of purity (>95%)
and that this predominate structure correlates with that expected
from the synthetic strategy.
The effect of the chain ends on the thermal transitions of the
caprolactone oligomers was then investigated by comparing the
DSC traces for the 3 possible deprotected derivatives: the acid-
terminated 32-mer with a TBDMS protected chain end, 17, the
hydroxy-terminated 32-mer with a benzyl ester protected chain
end, 18, and the doubly deprotected 32-mer with a single
hydroxyl and carboxylic chain end, 21. The latter was obtained
in quantitative yield by hydrogenation of 18. Comparison of
these derivatives with the protected 32-mer, 16 (see Figure 6),
shows very little difference between all four samples and similar
melting/crystallization temperatures, peak shapes in each case.
In contrast, comparison of the oligomers with monodisperse
poly(caprolactone), Mw ) 14 000, PDI ) 1.1, shows signifi-
cantly broader peaks and altered peak shapes for the commercial
PCl sample (Figure 7).
The availability of molecularly discrete oligomers of poly-
(ꢀ-caprolactone) allows the systematic study of the influence
of molecular weight and chain end structure on the physical
properties of this scientifically and commercially important
semicrystalline polymer. In comparison, discrete oligomers and
polymers of polyethylene have also been prepared and the
structure and formation of noninteger and interger folded-chain
crystals studied in detail by Ungar.16 To investigate these unique
series of materials, DSC measurements were initially performed
on doubly protected derivatives from the dimer, 4, to the 64-
mer, 19. A composite of DSC traces shown in Figure 6 displays
an intriguing progression of crystallinity and melting point
Analysis of the decomposition of the caprolactone oligomers
with increasing molecular weight also proved to be very
instructive. As shown in Figure 8, there is a well-defined
increase in thermal stability from the dimer, 4, which begins to
decompose at 190 °C through the tetramer, 7, (275 °C) and
finally the 64-mer, 19, where decomposition is initiated at
365 °C. This dramatic increase in thermal stability is most pro-
(15) (a) Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.;
Voit, B.; Pyun, J.; Fre´chet, J. M. J.; Sharpless, K. B.; Fokin, V. V. Angew.
Chem., Int. Ed. 2004, 43, 3928-3932. (b) Hawker, C. J.; Fre´chet, J. M. J.
J. Am. Chem. Soc. 1990, 112, 7638-7647. (c) Ihre, H.; Hult, A.; Soderlind,
E. J. Am. Chem. Soc. 1996, 118, 6388-6395. (d) Percec, V.; Chu, P. W.;
Kawasumi, M. Macromolecules 1994, 27, 4441-4453.
(17) (a) Bass, C.; Battesti, P.; Albe´rola, N. D. J. Apple. Polym. Sci. 1994, 53,
1745. (b) Kricheldorf, H. R.; Dunsing, R.; Serra, I.; Albert, A. Makromol.
Chem. 1987, 188, 2453. (c) Sarel, S.; Pohoryles, L. A. J. Am. Chem. Soc.
1958, 80, 4596. (d) Sidoti, G.; Capelli, S.; Meille, S. V.; Alfonso, G. C.
Polymer 1997, 39, 165.
(16) Ungar, G.; Zeng, X.; Brooke, G. M.; Mohammed, S. Macromolecules 1998,
31, 1875.
9
J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008 1723

A R T I C L E S
Takizawa et al.
Figure 7. Comparison of DSC traces for 32-mers with different chain end
substitution, 17, 18, and 21, with commercially available poly(caprolactone).
Figure 10. SAXS data for oligo(caprolactone) derivatives from the tetramer,
7, to the 64-mer, 19.
Figure 8. Comparison of TGA traces for caprolactone oligomers from the
dimer, 4, to the 64-mer, 19. In all cases, the chain ends are TBDMSO and
benzyl ester.
Figure 11. Graphical representation of the evolution of crystal structure
for caprolactone oligomers from the octamer, 10, to the 64-mer, 19.
of chain end functionality and a direct comparison with
commercially available, low polydispersity poly(caprolactone).
Of particular note was the observation that all of the 32-mer
samples, irrespective of the end groups, were significantly more
stable that either commercial poly(caprolactone) (Mw ) 14 000,
PDI ) 1.1) or a synthetic version of low polydispersity poly-
(caprolactone) (Mw ) 15 000, PDI ) 1.08) that had well-defined
ethoxy and carboxylic acid end group. In each case, the onset
of decomposition was lower for the polydisperse samples with
the commercial sample (poorly defined end groups) being ca.
100 °C and the synthetic derivative ca. 50 °C less stable than
the monodisperse derivatives. Interestingly, the 32-mer with the
lowest thermal stability was the mono-hydroxy, benzyl ester
terminated derivative, 18, followed by both the mono-carboxylic
acid, TBDMSO-terminated oligomer, 17, and the doubly depro-
tected, hydroxyl-carboxylic acid 32-mer, 21 (Figure 9). From
these results, it is apparent that the presence of a hydroxyl chain
end leads to decreased thermal stability, presumably due to the
occurrence of chain backbiting and associated depolymerization.
The presence of a terminal carboxylic acid group is also
destabilizing but to a decreased degree when compared to a
hydroxyl group whereas the polydisperse nature and possible
impurities in the commercial PCl dramatically decrease thermal
stability.
Figure 9. Comparison of TGA traces for 4 different chain end function-
alized 32-mers, 16, 17, 18, and 21, and comparison with commercially
available poly(caprolactone), Mw ) 14 000 Da, PDI ) 1.1.
nounced on going from 4 to the octamer, 10, with the increase
being significantly less for the octamer, 10, to the 64-mer, 19,
even though the length of the polymer chain is increased 8-fold.
In all cases, the chain ends were the same, a single TBDMS
protected hydroxyl group and a single benzyl ester protected
carboxylic acid group.
The strong correlation between oligomer length and thermal
decomposition profile prompted an examination of the effect
9
1724 J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008

Caprolactone Oligomers and Polymers
A R T I C L E S
Figure 12. Sequence of TM-AFM images for crystallized thin films of the 32-mer, 16, 64-mer, 19, and a commercial PCl sample of Mw ) 14 000, scan
size is 50 microns × 150 microns.
The availability of precisely defined oligomers of a semi-
crystalline polymers also represents an unprecedented op-
portunity to study the effect of molecular size on crystal
formation in polyester systems. Preliminary SAXS measure-
ments were therefore conducted at room temperature for the
caprolactone derivatives from the tetramer, 7, to the 64-mer,
19 (Figure 10). As expected, the tetramer did not show any
crystalline structure at room temperature, however all other
samples displayed a lamellae morphology. For the octamer, 10,
and the 16-mer, 13, the thickness D of the lamellae structure (7
and 11 nm, respectively) was very close to the theoretical
molecular lengths (7 and 12 nm, respectively), indicating that
the chains are fully extended in the lamellae structure.
In direct contrast, the 32-mer, 16, and 64-mer, 19, were shown
to have lamellae thicknesses significantly smaller than the
calculated molecular lengths. SAXS analysis showed a thickness
of 12 nm for 16 and 13 nm for 19, which is approximately 50
and 30% of their respective theoretical extended chain lengths
of 23 and 46 nm. This suggests that the 32-mer, 16, is folded
once and the 64-mer, 19, is folded multiple times. Significantly,
the nature of the chain end groups showed no detectable effect
on the SAXS profiles for the 16-mer, 32-mer and 64-mer series
of derivatives, which suggests that the crystal structure is
predominately determined by the length of the oligomer and
not the chain end functional groups. As a result, a preliminary,
graphical picture can be drawn for the evolution of crystal
structure with molecular weight for poly(caprolactone) with
increasing lamellae thickness up to the 16-mer resulting in a
maximum in crystal thickness and the occurrence of chain
folding at higher chain lengths (Figure 11). A similar crystal
thickness was also observed for the low polydispersity poly-
(caprolactone) sample with a molecular weight of 14 000,
indicating that the lamellae thickness was again approximately
16 repeat units, though the peak was broader. A more detailed
study of the structure and morphology of these semicrystalline
oligomers is currently under way.
A similar strong dependence of crystal shape on molecular
weight and polydispersity of the sample was observed by atomic
force microscopy. Analysis of the oligomers showed a system-
atic increase in crystallite size with monomer length for thin
films under both normal spin casting conditions and after
identical thermal annealing profiles. As can be seen in Figure
12, the size of the crystallites increases from ca. 30 microns
for the 32-mer, 16, to over 100 microns for the 64-mer, 19.
Interestingly, increasing the molecular weight from 7318 for
the 64-mer to a Mw of 14 000 for the commercial PCl sample
gave significantly different results with a multitude of small,
ca. 10 micron, crystallites being formed. This study again
demonstrates the significant difference in physical properties
and behavior between the well-defined oligomers and low
polydispersity, commercial PCl materials of comparable mo-
lecular weight and a systematic, in-depth examinination of the
thin film properties of these materials is in progress.
Conclusions
A synthetic strategy for the preparation of well-defined
ꢀ-caprolactone oligomers up to the 64-mer was developed using
an iterative divergent-convergent approach. By the selection of
9
J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008 1725

A R T I C L E S
Takizawa et al.
orthogonal protecting groups, t-butyldimethylsilyl ether and
benzyl ester, high yields of the oligomers were obtained and
the effect of both oligomer length and chain end functionality
examined. Characterization of each oligomer by a combination
of spectroscopic and chromatographic techniques revealed
>95% purity for each oligomer and a direct correlation between
oligomer length and thermal properties was observed. Increasing
molecular length led to novel semicrystalline behavior being
observed for each of the oligomers above the tetramer. A similar
relationship between oligomer length and crystal structure was
observed by SAXS measurements which showed formation of
lamellae crystals that increased in thickness with oligomer length
up to the 16-mer, after which the chains undergo folding with
a unit length of approximately 16 caprolactone units. The
availability of well-defined oligomers of poly(ꢀ-caprolactone)
allows new insight into the fundamental physical and biomaterial
properties of this widely studied and commercially important
polymer.
Acknowledgment. Financial support from the NSF, MRSEC
Program DMR-0520415 (MRL-UCSB), Mitsubishi Chemical
Center for Advanced Materials (MC-CAM), and Mitsubishi
Chemical Group Science and Technology Research Center, Inc.
is gratefully acknowledged.
Supporting Information Available: Instrumentation and
analytical techniques; synthetic procedures for oligomer and
intermediates. This;material is available free of charge via the
Internet at http://pubs.acs.org.
JA077149W
9
1726 J. AM. CHEM. SOC. VOL. 130, NO. 5, 2008