POLY(ESTER-ANHYDRIDES) BASED ON POLYLACTONE...
54
TKK Dissertations 104 Espoo 2008 POLY(ESTER-ANHYDRIDES) BASED ON POLYLACTONE PRECURSORS Doctoral Dissertation Helsinki University of Technology Faculty of Chemistry and Materials Sciences Department of Biotechnology and Chemical Technology Harri Korhonen
-
Upload
hoangtuyen -
Category
Documents
-
view
225 -
download
0
Transcript of POLY(ESTER-ANHYDRIDES) BASED ON POLYLACTONE...
TKK Dissertations 104Espoo 2008
POLY(ESTER-ANHYDRIDES) BASED ONPOLYLACTONE PRECURSORSDoctoral Dissertation
Helsinki University of TechnologyFaculty of Chemistry and Materials SciencesDepartment of Biotechnology and Chemical Technology
Harri Korhonen
TKK Dissertations 104Espoo 2008
POLY(ESTER-ANHYDRIDES) BASED ONPOLYLACTONE PRECURSORSDoctoral Dissertation
Harri Korhonen
Dissertation for the degree of Doctor of Science in Technology to be presented with due permission of the Faculty of Chemistry and Materials Sciences for public examination and debate in Auditorium KE2 (Komppa Auditorium) at Helsinki University of Technology (Espoo, Finland) on the 7th of March, 2008, at 12 noon.
Helsinki University of TechnologyFaculty of Chemistry and Materials SciencesDepartment of Biotechnology and Chemical Technology
Teknillinen korkeakouluKemian ja materiaalitieteiden tiedekuntaBiotekniikan ja kemian tekniikan laitos
Distribution:Helsinki University of TechnologyFaculty of Chemistry and Materials SciencesDepartment of Biotechnology and Chemical TechnologyP.O. Box 6100FI - 02015 TKKFINLANDURL: http://polymeeri.tkk.fi/Tel. +358-9-451 2616Fax +358-9-451 2622E-mail: [email protected]
© 2008 Harri Korhonen
ISBN 978-951-22-9219-6ISBN 978-951-22-9220-2 (PDF)ISSN 1795-2239ISSN 1795-4584 (PDF) URL: http://lib.tkk.fi/Diss/2008/isbn9789512292202/
TKK-DISS-2428
Multiprint OyEspoo 2008
AB
ABSTRACT OF DOCTORAL DISSERTATION HELSINKI UNIVERSITY OF TECHNOLOGY P.O. BOX 1000, FI-02015 TKK http://www.tkk.fi
Author Harri Korhonen
Name of the dissertation Poly(ester-anhydrides) based on polylactone precursors
Manuscript submitted 27.11.2007 Manuscript revised
Date of the defence 7.3.2008
Monograph Article dissertation (summary + original articles)
Faculty Faculty of Chemistry and Materials Sciences
Department Department of Biotechnology and Chemical Technology
Field of research Polymer Technology
Opponent(s) Professor Ari Rosling
Supervisor Professor Jukka Seppälä
Abstract Thermoplastic and crosslinked poly(ester-anhydrides) were prepared from biodegradable polyester precursors. The poly(ester-anhydrides) possess properties of individual polyesters and polyanhydrides and can therefore provide extended advantages compared to either polymer alone. The polymerization procedure consisted of ring-opening polymerization and subsequent coupling of polyester precursors to higher molecular weight poly(ester-anhydrides). Alternatively, carboxylic acid-functional prepolymers were allowed to react with methacrylic anhydride to form precursors for crosslinked poly(ester-anhydrides). Polyester precursors were prepared from L-lactide, DL-lactide and ε-caprolactone. In addition to different monomers used, the structure of the prepolymers was modified, using ricinoleic acid and alkenylsuccinic anhydrides with different chain lengths as hydrophobic components in the syntheses of prepolymers. In dissolution, the poly(ester-anhydrides) showed a hydrolysis of anhydride linkages within a few days. After hydrolysis of anhydride bonds, mass loss of poly(ester-anhydrides) depended most importantly on hydrophobicity and thermal properties of the polyester precursors. For poly(ester-anhydrides) prepared from prepolymers with thermal transitions below 37 °C, hydrolysis of anhydride linkages was accompanied by rapid mass loss caused by fast dissolution of the degradation products. When thermal transitions of prepolymers were above the hydrolysis temperature, the poly(ester-anhydrides) showed a clear two-stage degradation: a rapid hydrolysis of anhydride linkages was followed by slower hydrolysis and mass loss of the remaining polyester oligomer. Overall, the results demonstrated the potential to prepare biodegradable polymers with greatly modified degradation profiles through incorporation of anhydride linkages into the polyester backbone. These materials are expected to find applications in the field of drug release and tissue engineering.
Keywords poly(ester-anhydride), polylactone, hydrolysis
ISBN (printed) 978-951-22-9219-6 ISSN (printed) 1795-2239
ISBN (pdf) 978-951-22-9220-2 ISSN (pdf) 1795-4584
Language English Number of pages 52 p.+app. 56 p.
Publisher Helsinki University of Technology, Department of Biotechnology and Chemical Technology
Print distribution Helsinki University of Technology, Department of Biotechnology and Chemical Technology
The dissertation can be read at http://lib.tkk.fi/Diss/2008/isbn9789512292202/
VÄITÖSKIRJAN TIIVISTELMÄ TEKNILLINEN KORKEAKOULU PL 1000, 02015 TKK http://www.tkk.fi
Tekijä Harri Korhonen
Väitöskirjan nimi Polylaktoneihin perustuvat polyesterianhydridit
Käsikirjoituksen päivämäärä 27.11.2007 Korjatun käsikirjoituksen päivämäärä
Väitöstilaisuuden ajankohta 7.3.2008
Monografia Yhdistelmäväitöskirja (yhteenveto + erillisartikkelit)
Tiedekunta Kemian ja materiaalitieteiden tiedekunta
Laitos Biotekniikan ja kemian tekniikan laitos
Tutkimusala Polymeeriteknologia
Vastaväittäjä(t) Professori Ari Rosling
Työn valvoja Professori Jukka Seppälä
Tiivistelmä Biohajoavista polyestereistä valmistettiin termoplastisia ja ristisilloitettuja polyesterianhydridejä. Polyesterianhydridien etuna polyestereihin ja polyanhydrideihin nähden on se, että yhdistämällä kaksi erilaista polymeerityyppiä voidaan saada aikaan ominaisuusyhdistelmiä, joita kumpikaan polymeeri ei voi yksinään tarjota. Polymerointi tehtiin kaksivaiheisena siten, että laktonien renkaanavaavalla polymeroinnilla valmistettiin matalamoolimassaisia esipolymeereja, jotka toisessa vaiheessa kytkettiin korkeamoolimassaisiksi termoplastisiksi polyesterianhydrideiksi. Vaihtoehtoisesti esipolymeerit funtionalisoitiin metakryylihappoanhydridillä ja silloitettiin verkkorakenteisiksi polyesterianhydrideiksi. Esipolymeerit valmistettiin käyttäen lähtöaineena L-laktidia, DL-laktidia ja ε-kaprolaktonia. Erilaisten monomeerien lisäksi esipolymeerien valmistuksessa käytettiin hydrofobisina lähtöaineina joko risiiniöljyhappoa tai eripituisia alkenyyliketjuja sisältäviä meripihkahappoanhydridejä. Hydrolyysikokeissa polyesterianhydridien anhydridi-sidosten havaittiin hydrolysoituvan muutamassa päivässä. Anhydridisidosten hydrolysoitumisen jälkeen polymeerikappaleiden massan pienenemiseen eniten vaikuttavat tekijät olivat esipolymeerien hydrofobisuus ja termiset ominaisuudet. Kun esipolymeerien siirtymälämpötilat olivat alle 37 °C, polymeerikappaleiden massa laski nopeasti johtuen hajoamistuotteiden nopeasta liukenemisesta. Esipolymeerien termisten siirtymien ollessa korkeammalla kuin hydrolyysilämpötila, polyesterianhydridit hajosivat kaksivaiheisesti: nopeaa anhydridisidosten hydrolysoitumista seurasi jäljelle jääneiden polyesteri-oligomeerien hitaampi hydrolysoituminen ja sitä seuraava massan pieneneminen. Tulokset osoittavat, että sijoittamalla anhydridi-sidoksia polyestereiden pääketjuun voidaan muokata polymeerien hajoamiskäyttäytymistä huomattavasti. Tutkituille polymeereille uskotaan löytyvän sovellutuksia lääkeaineen vapautuksen ja kudosteknologian alalta.
Asiasanat polyesterianhydridi, polylaktoni, hydrolyysi
ISBN (painettu) 978-951-22-9219-6 ISSN (painettu) 1795-2239
ISBN (pdf) 978-951-22-9220-2 ISSN (pdf) 1795-4584
Kieli englanti Sivumäärä 52 s.+ liitteet 58 s.
Julkaisija Teknillinen korkeakoulu, Biotekniikan ja kemian tekniikan laitos
Painetun väitöskirjan jakelu Teknillinen korkeakoulu, Biotekniikan ja kemian tekniikan laitos
Luettavissa verkossa osoitteessa http://lib.tkk.fi/Diss/2008/isbn9789512292202/
AB
7
PREFACE This work was carried out in the Laboratory of Polymer Technology at Helsinki University of
Technology. The research was started in the National Technology Agency (TEKES) targeted
research projects and was then continued in the Bio- and Nanopolymers Research Group of the
Center of Excellence program funded by the Academy of Finland. The financial support from
TEKES and Academy of Finland is gratefully acknowledged.
I would like to thank Professor Jukka Seppälä for his guidance and interest in my work, and for
the opportunity to work in his laboratory. I gratefully acknowledge my co-authors Antti Helminen
and Risto Hakala for their contribution, invaluable comments, and collective hard work. Warm
thanks are extended to other former and present members of the biopolymers group, especially
Jaana Rich, Minna Malin, Jukka Tuominen, Thomas Gädda, and Janne Kylmä are thanked for
their help, thoughtful comments and discussions in many areas.
All colleagues and the personnel of the laboratory are thanked for creating a pleasant working
atmosphere and for all the help during the research. In particular, Jorma Hakala is thanked for his
help in literature research. Arto Mäkinen and Eija Ahonen are acknowledged for their technical
assistance.
I would like to thank my parents and brother for their support and help over the years. Finally, my
warmest thanks to my son Mauno.
Espoo, January 2008
Harri Korhonen
8
CONTENTS
LIST OF PUBLICATIONS........................................................................................................9 OTHER RELEVANT PUBLICATIONS AND PATENTS ....................................................10 NOMENCLATURE.................................................................................................................11 1 INTRODUCTION.................................................................................................................13
1.1 Background ....................................................................................................................13 1.2 Polyesters .......................................................................................................................14 1.3 Polyanhydrides...............................................................................................................16 1.4 Poly(ester-anhydrides) ...................................................................................................18 1.5 Scope of the study..........................................................................................................21
2 SYNTHESIS .........................................................................................................................22
2.1 Ring-opening polymerization ........................................................................................23
2.1.1 Co-initiators with different numbers of hydroxyl groups .......................................23 2.1.1.1 Initiation activity..............................................................................................23 2.1.1.2 High molecular weight polymers .....................................................................26
2.1.2 Preparation of polyester precursors for poly(ester-anhydrides) ..............................27 2.2 COOH-terminated prepolymers.....................................................................................28 2.3 Preparation of poly(ester-anhydrides)............................................................................30
2.3.1 Coupling of polyester precursors to linear poly(ester-anhydrides) .........................30 2.3.2 Preparation of crosslinked poly(ester-anhydrides)..................................................32
3 DEGRADATION..................................................................................................................33
3.1 Hydrolysis of anhydride bonds ......................................................................................34 3.2 Poly(ester-anhydrides) from different monomers ..........................................................34 3.3 Ricinoleic acid initiated poly(ester-anhydrides) ............................................................36 3.4 Alkenylsuccinic anhydride functionalized poly(ester-anhydrides) ................................38 3.5 Crosslinked poly(ester-anhydrides)................................................................................40 3.6 On-going research: a case study for the use of crosslinked poly(ester-anhydrides) as porogen materials.................................................................................................................41
4 CONCLUSIONS...................................................................................................................44 REFERENCES ........................................................................................................................47
9
LIST OF PUBLICATIONS This thesis is based on the following five publications (Appendices I-V), which are, throughout the summary, referred to by their Roman numerals. I Korhonen, H., Helminen, A., and Seppälä, J. V., Synthesis of polylactides in the presence
of co-initiators with different numbers of hydroxyl groups. Polymer 42 (2001) 7541-7549. II Korhonen, H. and Seppälä, J. V., Synthesis of poly(ester-anhydride)s based on poly(ε-
caprolactone) prepolymer. J. Appl. Polym. Sci. 81 (2001) 176-185. III Korhonen, H., Helminen, A. O, and Seppälä, J. V., Synthesis of poly(ester-anhydrides)
based on different polyester precursors. Macromol. Chem. Phys. 205 (2004) 937-945. IV Korhonen, H., Hakala, R. A., Helminen, A. O, and Seppälä, J. V., Synthesis and hydrolysis
behaviour of poly(ester-anhydrides) from polyester precursors containing alkenyl moieties. Macromol. Biosci. 6 (2006) 496-505.
V Helminen, A. O, Korhonen, H., Seppälä, J. V., Crosslinked poly(ester-anhydrides) based
on poly(ε-caprolactone) and polylactide oligomers. J. Polym. Sci. Part A: Polym. Chem. 41 (2003) 3788-3797.
The author’s contribution in the appended publications Publications I and III: Harri Korhonen and Antti Helminen are jointly responsible for the research plan, experimental work, interpretation of the results, and the preparation of the manuscript Publication II: Harri Korhonen is responsible for the research plan, experimental work, interpretation of the results, and the preparation of the manuscript Publication IV: Harri Korhonen and Risto Hakala planned the experiments, carried out polymerizations and the polymer characterizations and prepared the manuscript with Antti Helminen Publication V: Harri Korhonen and Antti Helminen are jointly responsible for the research plan, interpretation of the results, and the preparation of the manuscript. Antti Helminen carried out the experimental work.
10
OTHER RELEVANT PUBLICATIONS AND PATENTS Turunen, M. P. K., Korhonen, H., Tuominen, J., and Seppälä, J. V., Synthesis, characterization and crosslinking of functional star-shaped poly(ε-caprolactone), Polym. Int. 51 (2001) 92-100. Helminen, A., Korhonen, H., and Seppälä, J. V., Biodegradable crosslinked polymers based on triethoxysilane terminated polylactide oligomers, Polymer 42 (2001) 3345-3353. Helminen, A., Korhonen, H., and Seppälä, J. V., Structure modification and crosslinking of methacrylated polylactide oligomers, J. Appl. Polym. Sci. 86 (2002) 3616-3624. Helminen, A., Korhonen, H., and Seppälä, J. V., Crosslinked poly(ε-caprolactone/D,L-lactide) copolymers with elastic properties, Macromol. Chem. Phys. 203 (2002) 2630-2639. Seppälä, J. V., Korhonen, H., Kylmä, J., and Tuominen, J., General methodology for chemical synthesis of polyester, in Biopolymers Vol 3b Polyesters II – Properties and chemical synthesis, Doi, Y. and Steinbüchel, A. (Eds.), Wiley-VCH, Germany 2002, pp. 327-369. Seppälä, J. V., Helminen, A. O., and Korhonen, H., Degradable polyesters through chain linking for packaging and biomedical applications. Macromol. Biosci. 4 (2004) 208-217. Seppälä, J. V., Karjalainen, T. M., Rich, J. M., Korhonen, H. J., Ahola, N. M., Biodegradable material, WO 01/60425 A1, 2001 Seppälä, J. V., Hakala, R. A., Korhonen, H. J., New biodegradable polymers, WO 2007/085702 A1, 2007
11
NOMENCLATURE
Abbrevations
ASA alkenylsuccinic anhydride
8-ASA (+/-)-2-octen-1-ylsuccinic anhydride
12-ASA 2-dodecen-1-ylsuccinic anhydride
18-ASA n-octadecenylsuccinic anhydride
BD 1,4-butanediol
CL ε-caprolactone
CPP p-(carboxyphenoxy)propane
CPH p-(carboxyphenoxy)hexane
DLLA DL-lactide
DS degree of substitution
DSC differential scanning calorimetry
FA fumaric acid
FDA Food and Drug Administration
FTIR Fourier transform infrared spectroscopy
4-HBA 4-hydroxybenzoic acid
4-HMBA 4-(hydroxyl-methyl)benzoic acid
LA lactide
LLA L-lactide
MAAH methacrylic anhydride
MALDI-TOF matrix assisted laser desorption/ionization – time of flight
MW molecular weight
MWD molecular weight distribution
NMR nuclear magnetic spectroscopy
PBS phosphate buffer solution
PCL poly(ε-caprolactone)
PDLLA poly(DL-lactide)
PERYT pentaerythritol
PGL(-06,-10) polyglycerine(-06, -10)
PLLA poly(L-lactide)
ROA ricinoleic acid
ROP ring-opening polymerization
SA sebacic acid
SAH succinic anhydride
SEC size exclusion chromatography
SnOct2 stannous octoate
12
UV ultra violet
X conversion
Postfixes -A carboxylic acid terminated
-AH anhydride
-OH hydroxyl terminated
Symbols
∆H melting enthalpy (J/g)
Mn number average molecular weight (g/mol)
Mw weight average molecular weight (g/mol)
Tg glass transition temperature (°C)
Tm melting temperature (°C)
Designation of polymer samples
The polymer samples are designated in terms of the monomer (PCL for poly(ε-caprolactone),
PLLA for poly(L-lactide), and PDLLA for poly(D,L-lactide)); the amount of co-initiator (BD5 or
BD10 for 1,4-butanediol 5 or 10 mol-%); the number of carbons in the alkenyl chains of the
different succinic anhydrides (0, 8, 12, or 18); and the type of functionality (OH and A for
hydroxyl- and acid-terminated prepolymers, AH for poly(ester-anhydrides)). For example, acid-
terminated D,L-lactide oligomer polymerised with 5 mol-% of 1,4-butanediol and functionalised
with n-octadecenylsuccinic anhydride is designated PDLLA-BD5-18-A.
13
1 INTRODUCTION
1.1 Background
Degradable polymers have been extensively studied in the last few decades. Many of the
biodegradable polymers have good film forming properties, making them suitable for
applications including food containers, agriculture film, waste bags, and general use as packaging
material.1,2 In the biomedical field, biodegradable polymers have been used in orthopedic and
pharmaceutical applications such as sutures, drug delivery vehicles and surgical implants.3 When
used in medical applications, biodegradable polymers must be biocombatible and they must show
an appropriate rate of degradation. In addition, they must fulfil many requirements that depend on
the targeted application. These requirements include adequate mechanical, physical and thermal
properties and resistance to sterilization.4
Biodegradable polymers enjoy the advantages of self-elimination, avoiding the need to remove
the polymer from the site of implantation after its use. Biodegradable synthetic polymers offer a
number of advantages over other materials in biomedical applications. In drug delivery, targeted
delivery or localized drug delivery offers the advantage of reduced body burden and systemic
toxicity of the drugs, especially useful for highly toxic drugs like anticancer agents. Some of the
therapeutic agents with a relatively short half-life, specifically proteins, peptides and other
biologically unstable biomolecules, can also be delivered locally with minimal loss in therapeutic
activity.5,6 In tissue engineering, synthetic polymers are preferred over natural ones since they are
presumed to be free of immunogenicity and their physicochemical properties are more
predictable, reproducible, and easy to modify. Other advantages include the ability to tailor
mechanical properties and degradation kinetics to suit various applications. Synthetic polymers
are also attractive because they can be fabricated into various shapes with desired pore
morphologic features conducive to tissue in-growth. Furthermore, polymers can be designed with
chemical functional groups that can induce tissue in-growth.7-9
The growing need for biodegradable polymers for use in the emerging technologies including
drug delivery systems, tissue engineering and regenerative medicine, implantable devices, and
14
gene therapy has resulted in the development of a range of biodegradable polymers. Polyesters,
such as polylactide, polyglycolide and poly(ε-caprolactone), are widely studied bulk erodable
polymers with numerous medical and pharmaceutical applications. Polyanhydrides, on the other
hand, are among the most promising polymers for controlled drug delivery, since they display
surface erosion if the hydrophobicity of the monomers used in the polyanhydride synthesis is
high enough.10 As a means of improving the versatility of the two types of polymer, polyesters
and polyanhydrides have been combined into various poly(ester-anhydrides). This thesis is
focused on the preparation and properties of poly(ester-anhydrides) in which biodegradable
polyesters have been used as precursors for thermoplastic or crosslinkable poly(ester-anhydrides).
The syntheses result in polyesters that contain linkages whose degradation rate differs from ester
linkages and thus new types of degradation behaviour and polymer properties can be achieved.
These materials are expected to find applications in the biomedical field.
1.2 Polyesters
Polyesters are the best characterized and most widely studied biodegradable polymers. The
mechanism of degradation in polyester materials is classified as bulk degradation with random
hydrolytic scission of the polymer backbone. Biodegradable polyesters have been used in a
number of medical applications. The major applications include resorbable sutures, drug delivery
systems and orthopaedic fixation devices such as pins, rods and screws.11 Among the families of
synthetic polymers, the polyesters have been attractive for these applications because of their ease
of degradation by hydrolysis, degradation products being resorbed through the metabolic pathway
and the potential to tailor the structure to alter degradation rates12. Polyesters have also been
considered for development of tissue engineering applications. In addition to medical
applications, biodegradable polyesters are increasingly used in high-volume applications such as
packaging, films and fibers. 1,13,14
The most common way to obtain high molecular weight polyesters is through ring-opening
polymerization (ROP) of cyclic esters. ROP can be applied in the preparation of polyesters such
as polyglycolide, polylactides with different stereo structures, poly(ε-caprolactone) and poly(δ-
15
valerolactone), and polycarbonates.15 Polyglycolide is a highly crystalline (45-55%) thermoplastic
material with a melting temperature of 220-225 °C and a glass transition temperature of 36-40
°C. A major application of PGA is in resorbable sutures. Lactide is the cyclic dimer of lactic acid,
which exists in three stereoisomeric forms: L-lactide, D-lactide, and meso-lactide which contains
both L- and D-lactyl units in the ring. In addition, DL-lactide is a 50:50 mixture of L- and D-
lactides. Poly(L-lactide) has a melting temperature of 170-180 °C and glass transition
temperature in the range of 60-65 °C. PLLA exhibits high tensile strength and low elongation,
and high molecular weight PLLA therefore has sufficient strength for use a load bearing material
in medical applications. Poly(DL-lactide) is more attractive for use in drug delivery systems
because it is an amorphous polymer with Tg of 55-60 °C, and it thus degrades much faster than
PLLA. PCL is a ductile semicrystalline polymer with a melting temperature of 54-64 °C and a
glass transition temperature of -60 °C. PCL has good permeability to many therapeutic drugs and
has been studied for long-term contraceptive delivery. 16-19
The insertion mechanism has proven to be the most efficient method for the ROP of cyclic esters.
The most widely used initiators reacting by the insertion mechanism are carboxylates and
alkoxides of Sn, Ti, Zn, and Al. Among the initiators, Tin(II) 2-ethylhexanoate (SnOct2) is
probably the most widely used initiator in the polymerization of cyclic esters. SnOct2 is known
for its fast rate of polymerization, low degree of racemisation even at high temperatures, and
acceptance by the FDA.20 In addition, SnOct2 is commercially available, easy to handle, and
soluble in common organic solvents and cyclic ester monomers.21 When stannous octoate is
used, it first reacts with compounds containing hydroxyl groups forming tin alkoxide, which then
acts as an actual initiator in the polymerization.22-25 The structure of the polymer depends on the
alcohol used as a co-initiator. Mono- and difunctional alcohols yield linear polymers, whereas
alcohols with hydroxyl functionality higher than two give comb-shaped, star-shaped, hyper-
branched, or dendritic polymers. 26-32
The propagation is stopped via a chain transfer with another alcohol molecule, which causes the
polymerization to yield hydroxyl terminated polymers with molecular weight depending on the
ratio of monomer to co-initiator. Thus, the number average degree of polymerization of the
polyester formed in the cyclic ester monomer/SnOct2/co-initiator system is given by the
16
([monomer]0-[monomer])/[co-initiator]0. In the case of polymerization carried out with ‘pure’
SnOct2, the polyester chain growth is co-initiated with impurities present in SnOct2 and in
monomers. Due to these impurities, Mn ~ 106 g/mol seems to be the limit of Mn of the aliphatic
polyesters prepared by ROP.25,33
Chain extending of polyester oligomers offers a different route to high molecular weight
polymers. Chain extenders such as diisocyanates, bis(2-oxazolines), bis(epoxides), and bis(ketene
acetals) are bifunctional low-molecular weight monomers that in small amounts increase the
molecular weight of polymers in rapid reaction. The chain extenders also introduce new
functional groups and flexibility to the manufacture of polymers, which can be exploited to
improve the physical and mechanical properties and biodegradability of the resulting polymers.34
1.3 Polyanhydrides
Polyanhydrides are another extensively studied class of biodegradable polymers with
demonstrated biocompatibility and excellent controlled release characteristics.35 Polyanhydrides
are a unique class of polymers for drug delivery because some of them demonstrate a near zero
order drug release and relatively rapid biodegradation in vivo.7 In the past 20 years
polyanhydrides have been extensively investigated for use in the controlled delivery of a number
of drugs including chemotherapeutics, antibiotics, anaesthetics, and polypeptides. Recently, novel
polyanhydrides for use in the field of orthopedics and polymeric drugs have been reported.36-39
To obtain a device that erodes heterogeneously, the polymer should be hydrophobic yet contain
water-sensitive linkages. One type of polymer system that meets these requirements is the
polyanhydrides. Polyanhydrides undergo hydrolytic bond cleavage to form degradation products
that can dissolve in an aqueous environment. Polyanhydrides are believed to undergo
predominantly surface erosion due to the high water lability of the anhydride bonds on the
surface and the hydrophobicity which prevents water penetration into the bulk. This process is
similar to the slow disappearance of a soap bar over time. The decrease in the device thickness
throughout the erosion process, maintenance of the structural integrity, and the nearly zero-order
17
degradation kinetics suggest that heterogeneous surface erosion predominates.6,35,40,41
The majority of polyanhydrides are prepared by melt polycondensation. Starting with a
dicarboxylic acid monomer, a prepolymer of a mixed anhydride is formed with acetic anhydride.
The final polymer is obtained by heating the prepolymer under vacuum to remove the acetic
anhydride by-product. By optimizing polymerization conditions such as prepolymer purity,
reaction time and temperature, and removal of the condensation product, polymers with weight
average molecular weights of about 100 000 g/mol have been prepared. Higher molecular
weights were achieved by using catalysts such as cadmium acetate and earth metal oxides but
their use in medical grade polymers is limited because of their potential toxicity. 42 When
alternative synthesis methods such as ring-opening polymerization or reaction between dibasic
acids and diacid chlorides have been used, polymerizations have yielded polymers with
considerably lower molecular weights than obtained by melt polycondensation. 43,44
Degradation rates of polyanhydrides can be altered over a thousand fold by simple changes in the
polymer backbone. Aliphatic polyanhydrides degrade within days, whereas aromatic
polyanhydrides degrade over several years.45,46 Degradation rates of copolymers of aliphatic and
aromatic monomers vary between these two extremes. The most extensively studied
polyanhydride is a copolymer of sebasic acid (SA) and p-(carboxyphenoxy)propane (CPP). The
biocombatibility of copolymers of SA and CPP has been well established. Evaluation of the
toxicity of polyanhydrides has shown that they possess excellent in vivo biocombatibility and the
FDA has approved the use of poly(CPP-SA) for the treatment of brain cancer. 35 Other aromatic
diacids that have been used as a hydrophobic component in polyanhydrides are p-
(carboxyphenoxy)hexane (CPH), different p-(carboxyphenoxy)alkanoic acids, isophthalic acid,
and terephthalic acid.45,47,48
Fatty acids are another important group of hydrophobic monomers used in polyanhydrides.
Polyanhydrides synthesized from dimers of unsaturated fatty acids such as erucic acid and oleic
acid undergo surface erosion.49 In vivo studies, however, have showed that these polymers
degrade to the semisynthetic fatty acid dimers that are not easily metabolized in vivo. Another
way is to convert ricinoleic acid into a diacid by esterification with succinic or maleic anhydride
18
to give a nonlinear diacid. These polymers hydrolyze into ricinoleic acid-containing acid ester
monomers, which further hydrolyze to ricinoleic acid and succinic or maleic acid. The
inflammatory response after subcutaneous implantation of polymer samples, arising 21-days
post-implantation, was minimal to mild, comparable to that noted with clinically used Vicryl-
absorbable surgical suture.50,51 The properties of polyanhydrides have also been modified by
using monofunctional fatty acids as chain terminators of the polymerization, resulting in fatty
acid terminated polyanhydrides. Increasing the chain length of fatty acid terminals was found to
decrease the degradation rate of the polymer.52
The often limited mechanical stability of polyanhydrides has been a handicap to their use as
biomaterials for orthopaedic applications such as a temporary replacement in bone defects. To
overcome these limitations, unsaturated polyanhydrides that allow crosslinking, such as fumaric
acid based polyanhydrides, have been synthesized.53 In recent years, unsaturated polyanhydrides
have been studied and developed more intensively. 54-60 Typically, cross-linked polyanhydrides
were synhesized from monomers such as SA, CPP, or CPH after conversion to mixed anhydrides
with methacrylic anhydride. The methacrylated monomers were then photopolymerized by using
UV light and appropriate photo initiators.
Polyanhydrides have also been modified by the inclusion of amino acids such as glycine and
alanine into the polymer backbone to increase the mechanical properties of the polyanhydrides.
The amino acids were first converted into dicarboxylic acids by condensation with trimellitic
anhydride and subsequently polymerized to form poly(anhydride-co-imides). As a result of the
imide bond, the poly(anhydride-co-imides) displayed improved mechanical strength.61,62
1.4 Poly(ester-anhydrides)
Polyesters and polyanhydrides differ significantly in the rate and the mode of degradation. As a
means of improving the versatility of the degradation behaviour of the two type of polymer,
polyesters and polyanhydrides have been combined into various poly(ester-anhydrides). These
polymers possess properties of individual polyesters and polyanhydrides and may therefore
19
provide extended advantages compared to either polymer alone.
Poly(ester-anhydrides) containing different molar ratios of ester and anhydride groups have been
prepared by a number of groups. For example, Uhrich et al. have prepared poly(ester-anhydrides)
composed of alkyl chains linked by ester groups to aromatic moieties of salicylic acid. The
polymer undergoes hydrolytic degradation to release salicylic acid and thus it acts as a polymeric
prodrug.63-66 Pinther and Hartman describe the synthesis of polyanhydrides containing aromatic
and various aliphatic moieties connected by ester bonds.67 Jiang and Zhu have prepared
monomers for alternate poly(ester-anhydrides) by derivatization of p-hydroxybenzoic acid at the
hydroxyl terminus with cyclic anhydrides.68 Zhang et. al have prepared cycloaliphatic poly(ester-
anhydrides) in which various diols are introduced by ester link into the polyanhydride main chain
of poly(1,4-cyclohexanedicarboxylic anhydride) to decrease its melting point.69 Krasko et. al.
describe the introduction of ester bonds along the polyanhydride chain by the random reaction of
a polyanhydride with ricinoleic or lithocholic acid followed by the polymerization of
poly(anhydride ester) oligomers into high molecular weight polyanhydrides.70,71 Furthermore,
Kricheldorf et. al have prepared several kinds of thermotropic poly(ester-anhydrides) containing
varying mole ratios of ester and anhydride groups.72,73
More rarely, studies have been carried out on poly(ester-anhydrides) where biodegradable
polyesters are used as precursors for high molecular weight polymers. Slivniak and Domb
prepared ABA triblock copolymers composed of sebasic acid polyanhydride with poly(lactic
acid) (PLA) terminals. The PLA terminals were reported to have a significant effect on polymer
degradation and drug release rate.74 Xiao and Zhu incorporated anhydride linkages into the
poly(trimethylene carbonate) backbone to impart a higher degradation rate to the polymer.
Poly[(tetramethylene carbonate)-co-(sebacic anhydride)] copolymer was described as having
degradation behaviour similar to that of surface erosion materials.75,76 Storey and co-workers
prepared carboxylic acid terminated prepolymers from poly(ε-caprolactone) and converted them
to higher molecular weight poly(ester-anhydrides). The polymers displayed a two-stage
degradation profile in which rapid degradation of anhydride linkages was followed by a slower
degradation of polyester oligomers.77 Similarly, degradation tests for poly(sebacic anhydride-co-
ε-caprolactone) multi-block copolymers demonstrated the acceleration of the degradation rate
20
with increasing sebacic acid concentration.78
Besides their exhibition of unique degradation profiles, poly(ester-anhydrides) have the potential
for surface modification that makes the delivery potential of poly(ester-anhydride) devices
similar to that of polyanhydrides.79 Surface modification of poly(ester-anhydrides) based on
poly(L-lactic acid) prepolymers has recently been demonstrated by Pfeifer et al., who covalently
attached cystamine to the surface of poly(ester-anhydride) micro- and nanospheres.80 They
proposed that by exploiting the lability of anhydride bonds towards amine-containing
compounds, surface modification with more biologically oriented materials such as proteins,
sugars, and lipids should be possible. Furthermore, the same authors have reported a study in
which poly(ester-anhydride) copolymers and poly(β-amino esters) were formulated with plasmid
DNA into micro- and nanospheres, which were assayed for encapsulation and cellular
transfection properties over a range of poly(ester-anhydride) copolymer ratios.81
21
1.5 Scope of the study
The polymerization of lactic acid, lactides and ε-caprolactone has been extensively studied in the
Laboratory of Polymer Technology at Helsinki University of Technology. The main target of this
thesis was to produce new types of polyesters that contain linkages whose degradation rate differs
from ester linkages, i.e. poly(ester-anhydrides) were prepared by incorporating labile anhydride
into the polyester backbone. These polymers are expected to find applications in the field of
targeted drug delivery and tissue engineering.
This thesis discusses the research reported in five appended publications. Polymerization
methods are reported in publications I, II and IV. The ring opening polymerization of lactide in
the presence of co-initiators with different number of hydroxyl groups has been investigated in
publication I. The molecular structure of the polymers was confirmed and the effect of the co-
initiators on the polymerization was studied. Publication II describes the method for coupling of
polyester precursors to high molecular weight linear poly(ester-anhydrides). Synthesis of
crosslinked poly(ester-anhydrides) is reported in publication V.
The properties and degradation of different types of poly(ester-anhydrides) was investigated in
publications III-V. Degradation of linear poly(ester-anhydrides) prepared from different
monomers and co-initiators was studied in publication III. Poly(L-lactide), poly(D,L-lactide) and
poly(ε-caprolactone) prepolymers were synthesised by ring-opening polymerization of cyclic
esters in the presence of 1,4-butanediol (BD) or ricinoleic acid (ROA) as co-initiator. In
publication IV, poly(ester-anhydrides) were prepared from alkenylsuccinic anhydride modified
poly(lactone) precursors. The presence of a hydrophobic alkenyl chain in the polyester precursor
had a marked effect on the thermal properties and hydrolysis behaviour of the poly(ester-
anhydrides). Crosslinked PCL, PDLLA, and PLLA based poly(ester-anhydrides) showed very
rapid degradation as reported in publication V.
22
2 SYNTHESIS
The reaction scheme for the preparation of linear and crosslinked poly(ester-anhydrides) is shown
in Figure 1. The reaction consisted of three steps. Hydroxyl-terminated prepolymers were
prepared by ring-opening polymerization of cyclic esters in the presence of a co-initiator
containing the hydroxyl group. In the next step, terminal hydroxyl groups were converted to
carboxylic acid functionality in the reaction of hydroxyl groups with succinic or alkenylsuccinic
anhydride. In the final step, carboxylic acid groups were converted to anhydrides with acetic
anhydride, and these intermediates were coupled into poly(ester-anhydrides) by melt
polycondensation. Alternatively, carboxylic acid-functional prepolymers were allowed to react
with methacrylic anhydride to form precursors for crosslinked poly(ester-anhydrides).
OO O
RO C (CH2)2
O
C
O
OH
R
O C
O
CH3
O C
O
(CH2)2
R
OO O
OH
OHO
+ -
- OO O
C
O
O C (CH2)2
O R
C
O
C
O
OO C (CH2)2
O R
+ C
OO C
OOHC
O
-
C
O
O C (CH2)2
O R
O C
O
CH
CH2
CH3
O C
O
CH
CH2
CH3C
O
O C (CH2)2
O R
Linear poly(ester-anhydride) Crosslinked poly(ester-anhydride)
LA / CLROP
R =
Figure 1. Reaction scheme for the synthesis of linear and crosslinked poly(ester-anhydrides).
23
2.1 Ring-opening polymerization
Polyester precursors for poly(ester-anhydrides) were prepared by ring-opening polymerization of
cyclic esters in the presence of Sn(II)2-ethylhexanoate (SnOct2) as initiator. SnOct2 first reacts
with co-initiators containing hydroxyl groups to form a tin alkoxide that acts as an actual initiator
in the polymerization.22-25 Polymerization yields hydroxyl terminated polymers with molecular
weight depending on the ratio of monomer to co-initiator. In this thesis, ring-opening
polymerization studies concerned three aspects. The first target was to study the initiation activity
of hydroxyl groups when co-initiators with different numbers of hydroxyl groups were used.I
Secondly, the effect of the number of hydroxyl groups on polymerization of high molecular
weight polyesters was investigated.I Thirdly, in order to modify the hydrophobicity of the
polymers, different alcohols were used as co-initiators in the preparation of polyester precursors
for poly(ester-anhydrides). II, III
2.1.1 Co-initiators with different numbers of hydroxyl groups
2.1.1.1 Initiation activity
The structure of the polymer depends on the alcohol used as a co-initiator. Mono- and
difunctional alcohols yield linear polymers, whereas alcohols with hydroxyl functionality higher
than two give comb-shaped, star-shaped, hyper-branched, or dendritic polymers. In order to
systematically study the initiation activity of co-initiators, linear, star-shaped and branched
polylactides were prepared using alcohols with different numbers of hydroxyl groups as co-
initiators (Figure 2 and 3).I Linear polymers were produced with 1,4-butanediol as a co-initiator.
Pentaerythritol has four primary hydroxyl groups in equivalent positions and it was used to yield
a four-arm star-shaped polymer structure. The co-initiators of main interest were polyglycerine-
06 (PGL-06) and polyglycerine-10 (PGL-10), which were used as novel co-initiators with the aim
of achieving a more branched structure than with pentaerythritol. According to the manufacturer
(Daicel Chemiacal Industries), PGL-06 consists of six glycerol units connected with ether bonds,
so it possesses 8 hydroxyl groups on average. Correspondingly, the number of hydroxyl groups in
PGL-10 is 12 on average.
24
+R(OH)xSnOct2O
O
O
O
H3C
CH3
OCHC
O
O HR
CH3 x
Figure 2. Ring-opening polymerization of lactide
CH2 OH (CH2)4 OHHO C
CH2
CH2
CH2
CH2
OH
OH
OH
HO HO CH2 CH CH2 O H
OHn,av
Benzyl alcohol 1,4-butanediol Pentaerythritol n=6 Polyglycerine-06
n=10 Polyglycerine-10
Figure 3. Co-initiators used in the ring-opening polymerization of lactide.I
To confirm the effect of co-initiator on the polymer structure, low molecular weight polylactides
were prepared and characterised by 1H-NMR. Calculations were performed by two methods. The
direct method was to compare characteristic peaks of the unreacted and reacted OH groups of the
co-initiator. Alternatively, initiation activity was estimated indirectly by comparing theoretical
chain lengths with chain lengths determined by NMR
Table 1 shows that the number of hydroxyl groups initiating polymerization was theoretical for
1,4-butanediol (2 OH groups) and near theoretical for pentaerythritol (4 OH groups). Similar
initiation activity for PERYT has been reported in other studies. Kim et al.26 have reported that
increasing the ratio of monomer to initiator increases initiation activity, and the formation of
four-armed oligomer occurred when the ratio of monomer to PERYT reached 32. Similarly,
Lang et al.82 have reported that PCL oligomers had one, two, three, or four arms when the ratio of
monomer to PERYT was 5:1, whereas a higher ratio of 40:1 produced mainly three- or four-
armed oligomers.
25
Table 1. Initiation activity of hydroxyl groups in co-initiators estimated with 1H NMR.I
Co-initiator BD PERYT PGL-06 PGL-10 Co-initiator/lactide 3/100 5/100 10/100 5/100 8/100 12.5/100 1/100 3/100 5/100 5/100 Numbers of initiating OH groups
Theoretical
2
2
2
4
4
4
8
8
8
12
Calculation via the ratio of measured to theoretical chain lengths:
-OH
1.8
1.9
1.9
3.4
3.5
3.5
6.2
5.9
4.0
8.3
-CH-OH 2.0 2.0 2.0 3.4 3.9 3.8 9.9 9.5 8.0 11.1 Calculation via the characteristic peaks of reacted and unreacted OH groups in co-initiator: –CH2-OH / CH2-O-polymer
2.0
2.0
2.0
3.8
3.7
3.7
- a
- a
- a
- a
a) Overlapping peaks
The direct method could not be applied for oligomers initiated with polyglycerines owing to
overlapping of the peaks. In the indirect method, the chain length can be evaluated from the ratio
of either the hydroxyl proton or the terminal methine proton to methine protons in the polymer
chain. Since the peaks of the co-initiator partly overlapped the signal from the terminal methine
proton, the measured chain lengths were calculated more reliably from the hydroxyl protons. The
initiation activities for polyglycerines were 4.0-6.2 for PGL-06 (8 OH-groups) and 8.3 for PGL-
10 (12 OH-groups). It seems that secondary OH-groups of PGL did not initiate polymerization as
efficiently as primary OH-groups, and thus measured values were somewhat lower than
theoretical values. However, it can be concluded that a substantial proportion of secondary
hydroxyl groups take part in the initiation, and that their participation increases as the co-initiator
content decreases. This is in accordance with the behavior of hyperbranched polyglycerol as co-
initiator in the polymerization of ε-caprolactone. Burgath et al. have reported that in the case of a
theoretical arm length of 10 CL units, the initiation activity of OH-groups was 75%. However,
when the monomer/co-initiator ratio was higher and theoretical arm length was 30 CL units, an
initiation activity of 96% was measured.83
26
2.1.1.2 High molecular weight polymers
Polymerization of high molecular weight poly(L-lactide) in the presence of co-initiators with
different number of hydroxyl groups is reported in publication I. Polymerizations were carried
out at 200°C in the presence of co-initiator and SnOct2. The numbers of hydroxyl groups in co-
initiators were between1 to 12. Polymerizations were carried out with a fixed co-initiator content
of 0.03 mol-%.
The dependence of molecular weight on polymerization time for different co-initiators is shown
in Figure 4a. Compared with polymerization by SnOct2 alone, initiation and propagation were
considerably enhanced by all the co-initiators. The polymerization rate increased with increasing
number of hydroxyl groups in the co-initiator and the fastest polymerization was obtained with
the co-initiator having the highest hydroxyl group content (PGL-10). Increasing hydroxyl group
content in the co-initiator also yielded polymer with a higher molecular weight.
.
0
100000
200000
300000
400000
0 10 20 30 40 50 60
Time / min
Mw
/ g/
mol
None Benzyl alcohol 1,4-Butanediol
Pentaerythritol PGL-06 PGL-10
0
20
40
60
80
100
0 10 20 30 40 50 60
Time / min
Con
vers
ion
/ %
None Benzyl alcohol 1,4-Butanediol
Pentaerythritol PGL-06 PGL-10
Figure 4. Dependence of (a) Mw and (b) conversion on time for different co-initiators.I
Similarly to molecular weights, monomer conversions showed that the fastest polymerization was
achieved with the highest hydroxyl group content in the co-initiator (Figure 4b). Hydroxyl end
groups have been previously reported to cause back-biting via intramolecular
transesterification.84,85 High hydroxyl group content in the polymer, however, did not cause a
drop in the conversion level or enhanced backbiting during extended polymerization. All the co-
27
initiators yielded conversion of about 95%, and the conversions stayed on the same level to the
end of the polymerization. These results suggest that the hydroxyl content of the co-initiator
affects the polymerization rate but not the monomer/polymer equilibrium.
Star-shaped polymers are reported to have lower melting temperatures and higher cold-
crystallisation temperatures than linear ones.86 The melting temperatures of our polymers
followed this trend. The star-shaped polylactides initiated with PERYT or PGLs exhibited Tm of
165-172°C, while the linear polymers exhibited Tm of 172-173°C.
2.1.2 Preparation of polyester precursors for poly(ester-anhydrides)
In the preparation of poly(ester-anhydrides), the first step was to prepare hydroxyl-terminated
prepolymers. Poly(L-lactide), poly(D,L-lactide), and poly(ε-caprolactone) prepolymers were
synthesized by ring-opening polymerization of cyclic esters in the presence of a co-initiator
containing the hydroxyl group. Different co-initiators used in the polymerizations are shown in
Figure 5. 1,4-Butanediol initiated prepolymers were prepared to study the variation in properties
of poly(ester-anhydrides) obtained from different monomers. Ricinoleic acid, a fatty acid that has
been used in copolyanhydrides to increase hydrophobicity of the polymer 50, 51, was investigated
to see if it could be used as a hydrophobic co-initiator in the preparation of prepolymers. For the
same reason, 4-hydroxybenzoic acid and 4-(hydroxyl-methyl)benzoic acid were tested as co-
initiators because aromatic hydroxyacids are commonly used as the hydrophobic component in
copolyanhydrides.
OH
O
HO CH2
O
HO OHHO (CH2)4 OH HO
O
OH
(CH2)5
CH3
1,4-butanediol ricinoleic acid 4-hydroxybenzoic acid 4-(hydroxyl-methyl)benzoic acid
(BD) (ROA) (4-HBA) (4-HMBA)
Figure 5. Co-initiators used in the preparation of polyester precursors for poly(ester-
anhydrides).II, III
28
It is well known that hydroxyl containing co-initiators allow variation in the molecular weight via
the monomer/initiator ratio. The molecular weights of 1,4-butanediol were near the theoretical
values according to 1H NMR. Like the 1,4-butanediol initiated polymers, the ricinoleic acid
initiated polymers followed the trend of lower molecular weight with higher co-initiator
concentration, which indicates that the hydroxyl group of ricinoleic acid is active in initiation.
However, ricinoleic acid contained about 15% non-hydroxy impurities incapable of co-initiating
the polymerization, which resulted in ROA initiated prepolymers with molecular weights slightly
higher than 1,4-butanediol initiated polymers.
When 4-hydroxybenzoic acid was used as an initiator, the measured molecular weights were
much higher than the theoretical values, suggesting that propagation is more rapid than initiation.
Probably, the less nucleophilic nature of the aromatic hydroxyl group prevents effective
initiation. However, the initiator bearing one methylene group between the aromatic ring and the
hydroxyl group allowed more successful polymerization, since polymerization initiated by 4-
(hydroxymethyl)benzoic acid yielded a molecular weight near theoretical. This is in accordance
with benzyl alcohol initiated polymerizations. Benzyl alcohol has a similar hydroxyl group
position and it has been used without problems as a co-initiator in the polymerization of lactide.87
2.2 COOH-terminated prepolymers
In the preparation of poly(ester-anhydrides), the second step was to convert terminal hydroxyl
groups of prepolymers to carboxylic acid functionality in the reaction of hydroxyl groups with
succinic or alkenylsuccinic anhydride. The reaction scheme of the functionalizations is presented
in Figure 6.
OH O C (CH2)2
O
C
O
OH
CxHy
OO O
CxHy
X = 0, 8, 12, 18
Figure 6. Functionalization of OH-terminated polyester to carboxylic acid functionality.
29
Although catalysts such as pyridine can be used 75 88 89, functionalization of hydroxyl groups to
carboxylic acids was carried out in bulk without a catalyst. Comparing the two types of
prepolymer, there was a clear difference in the reactivities of the hydroxyl groups of PCL and
PLA According to 1H NMR studies, hydroxyl groups of PCL reacted completely with succinic
anhydride within 3 hours at 160°C (Figure 7). For PLA, the degree of substitution was 85-93%
despite a longer reaction time of 6 hours. Lang et al.82 have suggested that the difference in the
reactivity of cyclic anhydride with hydroxyl groups in PCL and PLA is mainly due to the
different reactivities of anhydride with primary and secondary alcohols.
a)
b)
Figure 7. 1H NMR spectra of PCL-ROA15: a) OH-terminated prepolymer, b) COOH-terminated
prepolymer. III
Similarly to succinic anhydride, all the alkenylsuccinic anhydrides (ASA) reacted easily with the
hydroxyl groups of PCL. For ASA-substituted polylactides, reaction conditions were varied
according to the length of the alkenyl chain in ASA. Reactions of 8-ASA were carried out at 140
°C for 24 hours and a degree of substitution (DS) between 87 and 96% was achieved. In the case
of 12-ASA, reactions carried out at 160 °C for 8 hours gave DS of 88-91%. Under the same
reaction conditions, DS for 18-ASA-substituted polymers was 70-80%. However, when more 18-
ASA was added to the reaction mixture and the reaction was continued another 8 hours at 160
°C, a DS of 90% was achieved.
C '
A '+ B '
C'
OHO
B'
A'
O OHO
OO
30
2.3 Preparation of poly(ester-anhydrides)
2.3.1 Coupling of polyester precursors to linear poly(ester-anhydrides)
Typically, in the synthesis of polyanhydrides, diacid monomers are refluxed in acetic anhydride,
the excess of acetic anhydride is vaporized at a mild temperature, and traces of acetic anhydride
are removed by extraction. Finally, the purified monomers are melt polycondensed to high
molecular weight polymers. In this work, one-pot melt polycondensation was applied, i.e. the
polycondensation step followed immediately after the removal of excess acetic anhydride. Pfeifer
et al.80 have recently used a similar method in the preparation of PLA-based poly(ester-
anhydrides). The major advantage of the one-pot synthesis is the avoidance of problems
associated with the handling of sticky prepolymers with anhydride functionalities sensitive to
moisture.
As reported in publication II, the one-pot preparation method proved to be successful for the
synthesis of poly(ester-anhydride)s. Upon coupling of PCL-based prepolymer, size exclusion
chromatography (SEC) analyses showed an increase from 3600 to 70 000 g/mol in number
average molecular weight. Upon couplings, the mixing efficiency was found to greatly affect the
molecular weights of the polymers: the linking of the same prepolymer with the use of a
magnetic bar yielded an increase in Mn from 2900 g/mol to 18 000 g/mol, while a considerably
higher Mn of 38 000 g/mol was achieved with the overhead mechanical stirrer.V
In comparison with poly(ester-anhydrides) prepared from different monomers, polymers based on
ε-caprolactone tended to show slightly higher molecular weights than lactide-based polymers
(Table 2). This can be explained by differences in the change of hydroxy groups to acid
functionality: for PCL oligomers the functionalization was nearly 100%, while for PLA the
degree of substitution was around 90%. Another factor that may affect molecular weights is the
better thermal stability of PCL at high temperature (180 °C) where the polycondensations were
carried out for one hour.
Apart from the monomer, the type of co-initiator affected the average molecular weights of the
31
poly(ester-anhydrides). As seen in Table 2, poly(ester-anhydrides) based on prepolymers
initiated with 1,4-butanediol contained 6-10 prepolymer units, while those based on prepolymers
initiated with ricinoleic acid contained 4-6 prepolymer units. The lower number in the case of
ricinoleic acid was probably due to the presence of impurities in ricinoleic acid. In spite of
purification by CH2Cl2/H2O extraction, some monofunctional non-hydroxy fatty acids probably
remained and terminated the chain growth.
Table 2. Molecular weight of 1,4-butanediol and ricinoleic acid initiated poly(ester-anhydrides).III
Acid-terminated prepolymer Poly(ester-anhydride) Monomer Co-initiator
mol-%
Mna
g·mol-1
Mwa
g·mol-1
MWD Xb
%
DSc
%
Mna
g·mol-1
Mwa
g·mol-1
MWD
1,4-Butanediol initiated polymers: DLLA 5 4200 5600 1.3 97 85 23000 39000 1.7 DLLA 10 2000 2400 1.2 98 92 13000 24000 1.9 LLA 5 4800 6000 1.2 97 93 27000 61000 2.2 LLA 10 2100 2500 1.2 97 88 15000 29000 1.9 ε-CL 5 4400 7000 1.6 >99 98 45000 99000 2.3 ε-CL 10 2900 3600 1.3 >99 99 18000 53000 3.0 Ricinoleic acid initiated polymers: DLLA 5 3900 4700 1.2 82 87 16000 30000 1.8 DLLA 10 1500 1800 1.2 83 83 7000 11000 1.6 LLA 5 6000 7700 1.3 85 84 22000 41000 1.9 LLA 10 2900 3900 1.3 88 83 14000 26000 1.9 ε-CL 5 5200 8500 1.7 >99 >99 33000 67000 2.0 ε-CL 10 3900 5900 1.5 97 >99 20000 42000 2.1 ε-CL 15 2700 4400 1.6 >99 >99 18000 37000 2.0
a Determined by SEC with respect to polystyrene standards. b Monomer conversion, determined by 1H NMR. c Degree of substitution from OH functionality to COOH functionality, determined by 1H NMR.
In publication IV, OH-terminated prepolymers were functionalized with alkenylsuccinic
anhydrides with three different alkenyl chain lengths. According to the molecular weights
obtained, the length of the alkenyl chain in ASA did not affect polycondensation of the
poly(ester-anhydrides) in general (Table 3). However, there were two 18-ASA functionalized
PCL-based prepolymers whose coupling failed: coupling of prepolymers PCL-BD5-18-A and
PCL-BD10-18-A yielded rubbery gels, which swelled extensively in dichloromethane. Similar
gelation of pure polyanhydrides was earlier attributed to the formation of entangled cyclic
macromers during polycondensation.42
32
Table 3. Molecular weights and thermal properties of ASA-functionalized poly(ester-anhydrides).IV
Prepolymer
Poly(ester-anhydride)
Prepolymer Mn g/mol
Mw g/mol
PD DS
%
Tg
°C
Tm
°C
∆H
J/g
Mn g/mol
Mw g/mol
MWD Tg °C
Tm °C
∆H J/g
PCL-BD5-OH 4600 5500 1.2 - - 48 -85
PCL-BD5-0-A 5000 6700 1.3 >99 - 50 -83 45000 99000 2.3 - 56 -72 PCL-BD5-8-A 5100 6400 1.3 >99 - 33, 42 a -65 40000 82000 2.1 - 41 -45 PCL-BD5-12-A 5000 6500 1.3 >99 - 32, 41 a -60 39000 78000 2.0 - 40 -43 PCL-BD5-18-A 5600 7000 1.2 97 - 30, 40 a -65 - - - - - -
PCL-BD10-OH 2400 2800 1.2 - - 36 -84
PCL-BD10-0-A 2900 3600 1.3 >99 - 38 -58 38000 183000 4.9 - 47 -49 PCL-BD10-8-A 2700 3400 1.3 >99 - 15 -47 31000 68000 2.2 - 14 -37 PCL-BD10-12-A 2900 3700 1.3 >99 - 16 -49 26000 52000 2.0 - 20 -47 PCL-BD10-18-A 3200 4200 1.3 96 - 10 -43 - - - - - - PDLLA-BD5-OH 4400 5700 1.3 - 30 - - PDLLA-BD5-0-A 4200 5600 1.3 85 32 - - 23000 39000 1.7 33 - - PDLLA-BD5-8-A 4100 5700 1.4 90 15 - - 28000 53000 1.9 23 - - PDLLA-BD5-12-A 4000 5300 1.3 92 8 - - 18000 32000 1.7 19 - - PDLLA-BD5-18-A 4700 6100 1.3 90 13 - - 24000 49000 2.0 20 - - PDLLA-BD10-OH 2200 2700 1.2 - 32 - - PDLLA-BD10-0-A 2000 2400 1.2 92 16 - - 13000 24000 1.9 20 - - PDLLA-BD10-8-A 2300 2800 1.2 93 -10 - - 19000 36000 1.9 6 - - PDLLA-BD10-12-A 2500 2900 1.2 92 -3 - - 18000 32000 1.7 10 - - PDLLA-BD10-18-A 2800 3300 1.2 95 -6 - - 13000 31000 2.5 10 - - PLLA-BD5-OH 5100 6200 1.2 - 36 127 -29 PLLA-BD5-0-A 4800 6000 1.2 93 40 130 -37 27000 61000 2.2 42 132 -29 PLLA-BD5-8-A 4800 6200 1.3 95 20 123 -1 22000 47000 2.1 36 119 b -24 b PLLA-BD5-12-A 4800 6200 1.3 94 26 125 -16 25000 55000 2.2 32 117 b -23 b PLLA-BD5-18-A 5100 6400 1.3 92 24 126 -7 26000 81000 3.1 31 121 b -30 b PLLA-BD10-OH 2300 2900 1.3 - 22 - - PLLA-BD10-0-A 2100 2500 1.2 88 25 - - 15000 29000 1.9 32 - - PLLA-BD10-8-A 2500 3100 1.2 93 -7 - - 21000 43000 2.0 11 - - PLLA-BD10-12-A 2800 3300 1.2 90 -5 - - 18000 33000 1.8 10 - - PLLA-BD10-18-A 3000 3500 1.2 92 -5 - - 18000 57000 3.2 11 - - a Split peak b First heating in DSC
2.3.2 Preparation of crosslinked poly(ester-anhydrides)
Preparation of crosslinked polyesters has been intensively studied in our previous work.90-93 In
this study, carboxylic acid terminated polyester oligomers were allowed to react with methacrylic
anhydride (MAAH) to form crosslinkable polyester oligomers with anhydride functionalities. A
similar method has been previously used in the synthesis of crosslinked polyanhydrides and
poly(ether-anhydrides).55,56,94 Functionalization was carried out at 60°C for 24 hours with an
33
excess of MAAH. After purification, the degree of substitution from carboxylic acid functionality
to anhydride functionality was found to be 92-98%.
The crosslinking was performed with benzoyl peroxide at 120°C. Upon curing, the physical
appearance of polymers changed from sticky or waxy oligomers to hard polymer specimens that
showed considerable swelling in CH2Cl2. The glass transition temperatures of the crosslinked
polymers were 9-18°C higher than those of the corresponding oligomers.
3 DEGRADATION
The effect of the concentration of anhydride linkages and the composition of the original
polyester prepolymer on the degradation of poly(ester-anhydrides) has been studied in
publications III and IV. The degradation of poly(ester-anhydrides) was studied by monitoring the
change in molecular weight, hydrolysis of anhydride bonds, water absorption and mass loss. The
chemical structures of different poly(ester-anhydrides) studied are shown in Figure 8.
(CH2)4 LA/CL O C
O
(CH2)2 C
O
O C
O
CH3OC
O
(CH2)2C
O
O LA/CLC
O
H3C
n
C
O
O C
O
CH3
n
O LA/CL O C
O
(CH2)2
O
OC
O
H3C
n
OC
O
CHCH2C
O
O LA/CLC
O
H3C
R
(CH2)4 LA/CL O C
O
CHCH2 C
O
O C
O
CH3
R
R =
Figure 8. Chemical structures of poly(ester-anhydrides: a) poly(ester-anhydrides) from different
monomersIII , b) ricinoleic acid initiated poly(ester-anhydrides)III , c) alkenylsuccinic anhydride
functionalized poly(ester-anhydrides)IV.
a)
b)
c)
34
3.1 Hydrolysis of anhydride bonds
Hydrolysis of anhydride bonds was studied by SEC and FTIR. Similar to the fast decrease in
molecular weights reported for polyanhydrides 95 30, poly(ester-anhydrides) showed a rapid drop
in molecular weights and polymers degraded back to the molecular weight level of the
prepolymers during the first 3 days (Figure 9a). FTIR confirmed that the decrease in molecular
weights during the first days was due to the hydrolysis of anhydride linkages (Figure 9b). A rapid
disappearance of the anhydride peak was detected independently of the monomer used. In
addition, the use of ricinoleic acid or alkenylsuccinic anhydride did not affect the rate of
hydrolysis of anhydride linkages. It thus seems that an increase in the hydrophobicity of polymers
was not sufficient to markedly slow the penetration of water into the specimen.
Figure 9. a) Molecular weight changes of 1,4-butanediol poly(ester-anhydrides) during immersion in PBS
(pH 7.0) at 37 °C. b) FTIR spectra of PCL-ROA10-AH exposed to PBS for different times.
3.2 Poly(ester-anhydrides) from different monomers
The rate of polymer mass loss after hydrolysis of anhydride linkages depends on the hydrolysis
rate and solubility of the polyester precursors. In addition to chemical composition, solubility was
greatly affected by the molecular weight and thermal properties of the prepolymers. The effect of
20001950 1900 1850 1800 1750 1700 1650 1600 15501500Wavenumber (cm-1)
0 day
1 day
2 days
3 days
Anhydride
0
5000
10000
15000
20000
25000
30000
35000
40000
0 1 2 3 4 5 6 7
Time / days
Mn /
g/m
ol
DLLA-BD5 DLLA-BD10 LLA-BD5LLA-BD10 CL-BD5 CL-BD10
35
molecular weight of prepolymers on the thermal properties of the polymers is seen in Figure 10,
i.e. an increase in co-initiator content resulted in lower thermal transitions for both the
prepolymers and the poly(ester-anhydrides).
0
10
20
30
40
50
60
0 5 10 15 20
Co-initiator / mol-%
Tm
/ °C
Prepolymer Poly(ester-anhydride)
Figure 10. Thermal properties of ricinoleic acid initiated ε-caprolactone based poly(ester-
anhydrides) and prepolymers.III
The mass loss differences attributable to the monomer used are shown in Figures 11a,b.
Independently of the monomer used, poly(ester-anhydrides) prepared from prepolymers with 10
mol-% of 1,4-butanediol showed rapid mass loss accompanied by high water absorption.
Prepolymers for these poly(ester-anhydrides) had thermal transitions below 37 °C, which caused
the hydrolysis of anhydride linkages, resulting in fast dissolving oligomers.
A decrease in 1,4-butanediol content from 10 mol-% to 5 mol-% led to longer prepolymer blocks
and higher thermal transitions. The effect of poly(ester-anhydride) crystallinity on degradation is
seen in a comparison of the two lactide-based polymers. PLLA-BD5-AH was a partly crystalline
polymer with a melting temperature of 130°C, and it showed practically no water absorption or
mass loss over a three-week test period. In contrast, PDLLA-BD5-AH was an amorphous
polymer that showed a lag time of two weeks with no significant changes in weight, but after that
a rapid mass loss occurred within one week. Clearly, the molecular weight of oligomers
decreased gradually until, after two weeks, oligomers were short enough to dissolve in the
36
dissolution medium.
In the case of ε-caprolactone poly(ester-anhydrides), the prepolymer of PCL-BD5-AH had Tm of
50 °C and there was no mass loss or water absorption during immersion. In contrast, the melting
temperature for the prepolymer of PCL-BD10-AH was 36 °C. This is slightly below the
immersion temperature and fast mass loss and high water absorption were observed.
Figure 11. a) Water absorption and b) mass loss of 1,4-butanediol initiated poly(ester-anhydrides) during
immersion in PBS (pH 7.0) at 37 °C.III
3.3 Ricinoleic acid initiated poly(ester-anhydrides)
The use of ROA in poly(ester-anhydrides) is reported in publication III. Ricinoleic acid is a fatty
acid that has been used in copolyanhydrides as the hydrophobic component of the polymer.50, 51
We supposed that, by using ROA as co-initiator, more hydrophobic polyester precursors could be
prepared and new possibilities for tailoring the degradation of poly(ester-anhydrides) would be
realised.
Mass loss for the ricinoleic acid initiated polymers is shown in Figure 12. For LLA-based
0
10
20
30
40
50
60
70
80
90
100
0 7 14 21 28
Time/days
Ma
ss lo
ss /
%
DLLA-BD5 DLLA-BD10 LLA-BD5
LLA-BD10 CL-BD5 CL-BD10
0
50
100
150
200
250
300
0 7 14 21 28
Time/days
Wa
ter
abs
orpt
ion
/ %
DLLA-BD5 DLLA-BD10 LLA-BD5
LLA-BD10 CL-BD5 CL-BD10
37
poly(ester-anhydrides), water absorption and mass loss increased steadily with time, and within
10 weeks about 20 % of the weight was lost. In comparison with 1,4-butanediol initiated
polymers, mass loss was much slower for PLLA-ROA10-AH than for the PLLA-BD10-AH,
exhibiting mass loss within few days. The difference in degradation is due to the different
thermal properties of the precursors: ROA initiated prepolymer showed crystallinity and had Tm
of 125 °C, whereas the prepolymer prepared with BD was amorphous. Differences in thermal
properties in turn are due to the different number of hydroxyl groups in the co-initiator, i.e. ROA
contains one initiating OH while BD contains two, which means that average chain lengths are
theoretically only half as long in BD as in ROA initiated prepolymers. For the same reason, PCL-
ROA10-AH exhibited mass loss of 10% in 10 weeks, while BD initiated polymer exhibited rapid
mass loss within one week.
In the case of PCL-based poly(ester-anhydrides), mass loss in hydrolysis was greatest for the
polymer with highest ROA content, although it should be the most hydrophobic one. Since the
polymer had the lowest melting temperature, it seems that for ROA, as for BD initiated polymers,
chain length and thermal properties are the most critical factors affecting the solubility of
polyester precursors and thus the rate of degradation.
Figure 12. Mass loss of ricinoleic acid initiated poly(ester-anhydrides) during immersion in PBS (pH 7.0)
at 37 °C.III
0
10
20
30
40
50
60
70
80
90
100
0 7 14 21 28 35 42 49 56 63 70
Time / days
Ma
ss lo
ss /
%
DLLA-ROA5 DLLA-ROA10 LLA-ROA5 LLA-ROA10
CL-ROA5 CL-ROA10 CL-ROA15
38
3.4 Alkenylsuccinic anhydride functionalized poly(ester-anhydrides)
In publication IV, the hydrophobicity of polyester precursors was increased by using
alkenylsuccinic anhydrides (ASAs) with different alkenyl chain lengths in the conversion of
hydroxyl group end functionality of the prepolymers to carboxylic acid functionality. The
increase in the hydrophobicity of prepolymers is clearly seen in the contact angles: succinic
anhydride functionalized PCL-based prepolymer clearly showed a lower contact angle than the
two ASA-functionalized prepolymers (Figure 13). In comparison with ASA-functionalized
polymers, lengthening of the alkenyl chain resulted in a considerable increase in the contact
angle.
0
10
20
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12 14 16
Time / s
Con
tact
ang
le /
°
0 8 18 Series1
Figure 13. Effect of alkenyl chain length on contact angles of PCL-based prepolymers (unpublished data).
The presence of alkenyl chain in the polyester precursor had a marked effect on the thermal
properties and hydrolysis behaviour of the poly(ester-anhydrides). For poly(ester-anhydrides)
prepared from low molecular weight prepolymers with thermal transitions below 37 °C, the
presence of hydrophobic alkenyl chains in the polyester precursors slowed the rate of mass loss.
As seen in Figures 14a and 14b, poly(ester-anhydrides) without an alkenyl chain showed a rapid
mass loss within a few days, while ASA-functionalized polymers exhibited a much smaller mass
loss over four weeks of immersion. The differences in length of the alkenyl chain, as such, had
little effect on the mass loss behaviour of the ASA-functionalized poly(ester-anhydrides).
39
Figure 14. Mass loss of a) PDLLA-BD10-AHs and b) PLLA-BD10-AHs with different alkenyl chain
lengths during immersion in PBS (pH 7.0) at 37 °C.IV
Poly(ester-anhydrides) prepared from higher molecular weight prepolymers showed different
mass loss behaviour. Among the polyester precursors and poly(ester-anhydrides), thermal
transitions of alkenylsuccinic anhydride-functionalized polymers tended to be 10-15 °C lower
than those of the corresponding polymers without an alkenyl chain. Due to lower crystallinities
and thermal transitions, alkenyl chain-containing poly(ester-anhydrides) showed a faster mass
loss than for poly(ester-anhydrides) without the alkenyl chain (Figures 15a,b).
Figure 15. Mass loss of a) PLLA-BD5-AHs and b) PDLLA-BD5-AHs with different alkenyl chain lengths
during immersion in PBS (pH 7.0) at 37 °C.IV
0
10
20
30
40
50
60
70
80
90
100
0 14 28 42 56 70 84
Time / days
Mas
s lo
ss /
%
0 8 12 18
0
10
20
30
40
50
60
70
80
90
100
0 7 14 21 28
Time / days
Mas
s lo
ss /
%
0 8 12 18
0
10
20
30
40
50
60
70
80
90
100
0 7 14 21 28
Time / days
Mas
s lo
ss /
%
0 8 12 18
0
10
20
30
40
50
60
70
80
90
100
0 7 14 21 28
Time / days
Mas
s lo
ss /
%
0 8 12
40
3.5 Crosslinked poly(ester-anhydrides)
Crosslinked polyesters have been investigated in our previous studies.90-93 To increase the
degradation rate of the crosslinked polyesters, anhydride bonds were incorporated into the
polyester precursors. Crosslinked poly(ester-anhydrides) based on polylactide and poly(ε-
caprolactone) precursors degraded in a few days in PBS at 37 °C. Similarly with linear
poly(ester-anhydrides), crosslinked poly(ester-anhydrides) showed a two-stage degradation in
which a rapid cleavage of anhydride bonds was followed by the degradation and dissolution of
constituent oligomers.
Comparing different poly(ester-anhydrides), polylactide based polymers showed higher water
absorption and faster cleavage of anhydride bonds than PCL based poly(ester-anhydrides). After
one day in hydrolysis, water absorption of PLA samples was nearly 100 % and characteristic
anhydride absorbances at 1820 cm-1 were not detected in FTIR. The degradation of the PCL
based polymer networks was different from that of the polylactide networks. The water
absorption was significantly lower in PCL networks than in PLA networks, owing to the higher
hydrophobicity of the PCL blocks. PCL-BD5-AH showed mass loss during the first 3 days and
after that no further mass loss occurred during a two week test period (Figure 16). In PCL-BD10-
AH, the mass loss was nearly linear and completed in two days. According to FTIR, anhydride
bonds were still present in PCL-BD10-AH after 24 and 40 hours of immersion, at which point
mass loss was about 40% and 90% respectively. In addition, the dimensions of the specimen
decreased steadily as shown in Figure 17. Hence, PCL-BD10-AH showed clear signs of surface
erosion: a linear mass loss but a practically intact core.
41
0
20
40
60
80
100
0 1 2 3 4
Time (days)
Mas
s lo
ss (
wt.%
) PCL10-AH
PCL10-COOH
PCL5-AH
PCL5-COOH
(b)
Figure 16. The mass loss of the PCL based poly(ester-anhydride) networks and the corresponding COOH-
terminated oligomers.V
Figure 17. The disk degradation profile for poly(ester-anhydride) network PCL-BD10-AH.V
3.6 On-going research: a case study for the use of crosslinked poly(ester-anhydrides) as
porogen materials
Poly(ester-anhydrides) studied in this thesis are expected to find use in tissue engineering and in
controlled release applications. In order to demonstrate the new possibilities that poly(ester-
anhydrides) can offer in these fields, the on-going study for the use of poly(ester-anhydride)
fibres as porogen materials is briefly presented. In this study, fast eroding poly(ester-anhydride)
fibres were selectively leached from more slowly degrading polyester matrix to form a
predetermined pore structure within the matrix material. The chemical structures for the
42
poly(ester-anhydride) fibres and for the polyester matrix are shown in Figure 18.
OCH2C C
O
(CH2)5 O C
O
CH
CH2
CH3 O C
O
(CH2)2 C
O
O C
O
CH
CH2
CH3(CH2)5C
O
OCH2C
44
matrix porogen
Figure 18. Chemical structures of a) matrix and b) porogen materials.
Porogen fibres were prepared by photo curing and they were used in the form of elastic fibre mat
or a mesh. The fibres were half-moon shaped with a thickness of 100 µm and diameters of 450-
850 µm along the wider edge. In hydrolysis, poly(ester-anhydride) porogens dissolved from the
photo cured polyester matrix within one week. As seen in scanning electron microscope
micrographs, the shapes and dimensions of the pores formed in hydrolysis closely corresponded
to the original porogen fibres. Furthermore, micro-CT images showed that porosity mimicked the
form and dimensions of the porogen mat or mesh throughout the matrix sample. The porosities
estimated by micro-CT were 30-39 %, while the amount of porogens added in the matrix was 30
wt-%.
Figure 19. a) SEM micrograph of the porogen fibre, b) SEM micrograph of the porous matrix, c) micro-
CT image of the pore structure within the matrix porogenized with poly(ester-anhydride) mesh.
43
The results show that interconnecting porosity with predetermined dimensions can be obtained by
using fast eroding crosslinked poly(ester-anhydrides) as porogen materials. Other expected
advantages include the possibility for in situ porogenisation and the possibility for preparation of
complex 3-D porogen networks by stereo lithography techniques. In addition, preparation of
poly(ester-anhydride) porogens by photo curing at room temperature allows adding of
temperature sensitive active agents such as proteins into the porogen material. Thus, it should be
possible to prepare bioactive scaffolds with controlled porosity for tissue engineering
applications.
44
4 CONCLUSIONS
The main focus of this work was to prepare polyesters that contain anhydride linkages
incorporated along the polyester backbone. The polymerization procedure consisted of ring-
opening polymerization and subsequent coupling of polyester precursors to higher molecular
weight poly(ester-anhydrides). Polyester precursors were prepared from L-lactide, DL-lactide and
ε-caprolactone. In addition to the different monomers used, structure of the prepolymers was
modified using ricinoleic acid and alkenylsuccinic anhydrides with different chain lengths as
hydrophobic components in the syntheses of prepolymers. The major findings of the thesis are
summarized in the following:
• The use of alcohols with different numbers of hydroxyl groups as co-initiators enabled the
preparation of linear, star-shaped and branched polylactides by ring-opening polymerization. By
using polyglycerines as novel co-initiators, highly branched polylactides with a number of arms
exceeding 8 were obtained. The increasing number of hydroxyl groups in the co-initiator was
found to increase the polymerization rate in the preparation of high molecular weight
polylactides.
• Polyester precursors for poly(ester-anhydrides) were prepared from different monomers and co-
initiators. 1,4-Butaniediol initiated prepolymers were synthesized to study the variation in
properties of poly(ester-anhydrides) obtained from L-lactide, DL-lactide and ε-caprolactone.
More hydrophobic prepolymers were synthesized using ricinoleic acid and 4-(hydroxyl-
methyl)benzoic as co-initiators in the preparation of polyester precursors.
• Terminal hydroxyl groups of polyester prepolymers were converted to carboxylic acid
functionality in the reaction of hydroxyl groups with succinic or alkenylsuccinic anhydride.
Functionalization of hydroxyl groups to carboxylic acids was carried out in bulk without a
catalyst. PCL showed higher reactivity than PLA in functionalizations.
• Preparation methods for linear and crosslinkable poly(ester-anhydrides) were developed.
Linear poly(ester-anhydrides) were prepared by one-pot melt polycondensation. Crosslinkable
45
precursors for poly(ester-anhydrides) were prepared through the reaction of carboxylic acid-
functional prepolymers with methacrylic anhydride.
• In dissolution, all the poly(ester-anhydrides) showed a hydrolysis of anhydride linkages within a
few days. After hydrolysis of the anhydride bonds, mass loss of the poly(ester-anhydrides)
depended on the composition of the original polyester prepolymer
• Mass loss of poly(ester-anhydrides) was greatly affected by the molecular weight and thermal
properties of the prepolymers. For poly(ester-anhydrides) prepared from prepolymers with
thermal transitions below 37 °C, hydrolysis of anhydride linkages was accompanied by rapid
mass loss caused by fast dissolution of the degradation products. When thermal transitions of
prepolymers were above the hydrolysis temperature, the poly(ester-anhydrides) showed a clear
two-stage degradation: a rapid hydrolysis of anhydride linkages was followed by slower
hydrolysis and mass loss of the remaining polyester oligomer
• Several factors affected the solubility of the degradation products at the same time. As
expected, alkenysuccinic anhydride functionalization had a slowing effect on the mass loss of
poly(ester-anhydrides) prepared from low molecular weight prepolymers due to increased
hydrophobicity. However, the mass loss behaviour was opposite for poly(ester-anhydrides)
prepared from higher molecular weight prepolymers: the lower thermal transitions and lower
crystallinities of ASA-functionalized polymers overrode the greater hydrophobicity and alkenyl
chain-containing poly(ester-anhydrides) showed faster mass loss than poly(ester-anhydrides)
without the alkenyl chain.
Overall, the results demonstrate the potential for preparing biodegradable polyesters with greatly
modified degradation profiles through incorporation of anhydride linkages into the polyester
backbone. Further development of materials is in progress, and poly(ester-anhydrides) have
already shown promising results in the field of drug release and tissue engineering. When PCL-
based crosslinked poly(ester-anhydride) was used as a matrix for the release of high molecular
weight dextran as a model compound, zero order release kinetics was observed within three days.
Adjustability of the degradation times of poly(ester-anhydrides) from days to months is the
46
biggest challenge for polymer synthesis in the future. In tissue engineering, fast eroding
crosslinked polymer fibers have been used for generating an interconnecting pore network with
predetermined dimensions into the scaffold. In addition, fibres containing bioactive glass have
been preliminarily tested for the preparation of bioactive scaffolds.
47 REFERENCES
1 Van de Velde, K. and Kiekens P., Polymer Testing, Biopolymers: Overview of several properties and consequences on their applications, Polymer Testing 21 (2002) 433-442. 2 Auras, R., Harte, R. and Selke, S., An overview of polylactides as packaging materials, Macromol. Biosci. 4 (2004) 835-864. 3 Chandra, R. and Rustgi R., Biodegradable polymers, Prog. Polym. Sci. 23 (1998) 1273-1335 4 Ikada Y and Tsuji H., Biodegradable polyesters for medical and ecological applications, Macromol. Rapid Commun. 21 (2000) 117-132. 5 Jain, J. P., Modi, S., Domb, A. J., and Kumar, N. Role of polyanhydrides as localized drug carriers, J. Controlled Release 103 (2005) 541-563. 6 Uhrich, K. E., Cannizzaro, S. E., Langer, R. S. and Shakesheff, S. M., Polymeric systems for controlled drug release, Chem. Rev. 99 (1999) 3181-3198. 7 Gunatillake, P. A. and Adhikari R., Biodegradable synthetic polymers for tissue engineering, European cells and materials 5 (2003) 1-16. 8 Martina M. and Hutmacher D. W., Biodegradable polymers applied in tissue engineering reseach: a review, Polym. Int. 56 (2007) 145-157. 9 Sokolsky-Papkov, M., Agashi, K., Olaye, A., Shakesheff, K. and Domb, A. J., Polymer carriers for drug delivery in tissue engineering, Advanced drug delivery reviews 59 (2007) 187-206 10 Domb, A. J., Elmalak, O., Shastri, V. R., Ta-Shma, Z., Masters, D. M., Ringel, I., Teomim, D., and Langer, R., Polyanhydrides. In Handbook of biodegradable polymers, Domb, A. J., Kost, J., Wiseman, D. M. (eds.), Harwood Academic Publishers, Amsterdam, (1997) 135.159. 11 Bendix, D., Chemical synthesis of polylactide and its copolymers for medical applications, Polym. Degrad. Stab. 59 (1998) 129-135 12 Amass, W., Amass, A., and Tighe, B., A review of biodegradable polymers: uses, current developments in the synthesis and chracterization of biodegradable polyesters, blends of biodegradable polymers and recent advances in biodegradation studies, Polym. Int. 47 (1998) 89-144. 13 Lunt, J., Large-scale production, properties and comercial applications of polylactic acid polymers, Polym. Degrad. Stab. 59 (1998) 145-152. 14 Vink, E. T. H., Rabago, K. R., Glassner, D. A., Springs, B., O'Connor, R. P., Kolstad, J., Gruber P. R., The sustainability of NatureWorksTM polylactide polymers and IngeoTM polylactide fibers: an update of the Future, Macromol. Biosci. 4 (2004) 551-564. 15 Löfgren, A., Albertsson, A.-C., Recent advances in ring-opening polymerization of lactones and related compounds, J.M.S.-Rev.Macromol.Chem.Chys. 35 (1995) 379-418. 16 Albertsson, A.-C., Varma, I. K., Recent developments in ring opening polymerization of lactones for biomedical applications, Biomacromolecules, 4 (2003) 1466-1486. 17 Mehta, R., Kumar, V., Bhunia, H., and Upadhyay S. N., Synthesis of poly(lactic acid): a review, J. Macromol. Sci., Part C: Polymer Reviews 45 (2005) 325-349. 18 Södergård, A., Stolt, M. Properties of lactic acid based polymers and their correlation with composition, Prog.
48 Polym. Sci. 27 (2002) 1123-1163. 19 Bastioli, C., Biodegradable materials – present situation and future perspectives, Macromol. Symp. 135 (1998) 193-204. 20 Witzke, D. R, Narayan, R., and Kolstad, J. J., Reversible kinetics and thermodynamics of the homopolymerization of L-lactide with 2-ethylhexanoic acid Tin(II) salt, Macromolecules 30 (1997) 7075-7085. 21 Möller, M., Kånge, R., and Hedrick, J. L., Sn(OTf)2 and Sc(OTf)3: Efficient and versatile catalyst for the controlled polymerization of lactones, J. Polym. Sci.: Part A: Polym. Chem. 38 (2000) 2067-2074. 22 Kowalski, A, Duda, A, and Penczek, S., Kinetics and mechanism of cyclic esters polymerization initiated with tin(II) octoate, 1, Macromol. Rapid Commun. 19 (1998) 567-572 23 Kowalski, A, Duda, A, and Penczek S. Mechanism of cyclic ester polymerization initiated with tin(II) octoate,. 2. Macromolecus fitted with tin(II) alkoxide species observed directly in MALDI-TOF spectra, Macromolecules 33 (2000) 689-695 24 Kricheldorf, H. R., Kreiser-Saunders, I., Stricker, A., Polylactones 48. SnOct2-initiated polymerizations of lactide: a mechanistic study, Macromolecules 33 (2000) 702-709 25 Duda, A., Penczek, S., and Kowalski, A., Polymerizations of ε-caprolactone and L,L-dilactide initiated with stannous octoate and stannous butoxide – a comparison, Macromol. Symp. 153 (2000) 41-53 26 Kim, S. H., Han, Y. K, Kwang-Duk, A., Kim, Y. H., Taihyun, C., Preparation of star-shaped polylactide with pentaerythritol and stannous octoate, Macromol. Chem. 194 (1993) 3229-3236. 27 Hao, Q., Li, F., Li, Q., Li, Y., Jia, L., Yang, J., Fang Q., and Cao, A., Preparation and crystallization kinetics of new structurally well-defined star-shaped biodegradable poly(L-lactide)s initiated with diversed natural sugar alcohols, Biomacromolecules 6 (2005) 2236-2247. 28 Biela, T., Polanczyk, I., One-pot synthesis of starshaped aliphatic polyesters with hyperbranched cores and their characterization with size exclusion chromatography, J. Pol. Sci: Part A: Pol. Chem. 44 (2006) 4214-4221. 29 Ouchi, T., Ichimura, S., and Ohya, Y., Synthesis of branched poly(lactide) using polyglycidol and thermal, mechanicl properties of its solution-cast film, Polymer 47 (2006) 429-434. 30 Dailey, L. A., Wittmar, M., and Kissel, T., The role of branched polyesters and their modifications in the development of modern drug delivery vehicles, J. Controlled Release 101 (2005) 137-149. 31 Zhao, Y.-L., Cai, Q., Jiang, J., Shuai, X.-T., Bei, J.-Z., Chuan-Fu, C., and Xi, F., Synthesis and thermal properties of novel star-shaped poy(L-lactide)s with starburst PAMAM-OH dendrimer macroinitiator, Polymer 43 (2002) 5819-5825. 32 Trollsås M., Atthoff, B., Claesson, H., and Hedrick J. L., Dendritic homopolymers and block copolymers: tuning the morphology and properties, J. Pol. Sci: Part A: Pol. Chem.42 (2004) 1174-1188. 33 Biela, T., Kowalski, A., Libiszowski J., Duda, A., and Penczek, S., Progress in polymerization of cyclic esters: mechanisms and synthetic applications, Macromol. Symp. 240 (2006) 47-55. 34 Seppälä, J. V., Korhonen, H., Kylmä, J., and Tuominen, J., General methodology for chemical synthesis of polyester, in Biopolymers Vol 3b Polyesters II – Properties and chemical synthesis, Doi, Y. and Steinbüchel, A. (Eds.), Wiley-VCH, Germany 2002, pp. 327-369. 35 Katti, D. S., Lakshami, S., Langer, R., and Laurencin, C. T., Toxicity, biodegradation and elimination of polyanhydrides, Advanced Drug Delivery Reviews 54 (2002) 933-961.
49 36 Li, L. C., Deng, J., and Stephens, D., Polyanhydride implant for antibiotic delivery – from the bench to the clinic, Advanced Drug Delivery Reviews 54 (2002) 963-986. 37 Torres, M., Determan, A. S., Anderson, G. L., Mallapragada, S. K., and Narasimhan, B., Amphiphilic polyanhydrides for protein stabilization and release, Biomaterials 28 (2007) 108-116. 38 Laayre, A.-M., Gref, R., Richard, J., Requier, D., Chacun, H., Domb, A. J., and Couvreur, P., Nanoencapsulation of a crystalline drug, Int. J. Pharmaseutics 298 (2005) 323-327. 39 Guo W.-X., Huang, K.X., Tang, R., and Chi, Q., New polyanhydrides derived from C12, C13, C14, C15 dibacic acid: synthesis and characterization, Polymer 45 (2004) 5743-5748. 40 Göpferich, A., Tessmar, J., Polyanhydride degradation and erosion, Advanced Drug Delivery Reviews 54 (2002) 911-931. 41 Von Burkersroda, F., Schedl, L., and Göpferich, A., Why degradable polymers undergo surface erosion or bulk erosion, Biomaterials 23 (2002) 4221-4231. 42 Domb, A. J., Langer, R., Polyanhydrides. I. Preparation of high molecular weight polyanhydrides, J. Pol. Sci: Part A: Pol. Chem. 25 (1987) 3373-3386. 43 Kumar, N., Langer, R. S., Domb, A. J., Polyanhydrides: an overview, Advanced Drug Delivery Reviews 54 (2002) 889-910. 44 Domb, A. J., Elmalak, O., Shastri, V. R., Ta-Shma, Z., Masters, D. M., Ringel, I., Teomim, D., and Langer, R., Polyanhydrides, in Handbook of biodegradable polymers, Kost, J. and Wiseman, D. M. (Eds.), Harwood Academic Publishers, Singapore 1997, pp. 135-159. 45 Leong, K. W., Brott, B. C., and Langer, R., Bioerodible polyanhydrides as drug-carrier matrices. I: Characterization, degradation and release chracteristics, J. Biomed. Mat. Res. 19 (1985) 941-955. 46 Domb, A. J., Nudelman, R., In vivo and in vitro elimination of aliphatic polyanhydrides, Biomaterials 16 (1995) 319-323. 47 Domb, A. J., Gallardo, C. F., and Langer, R., Polyanhydrides. 3. Poly(anhydrides) based on aliphatic-aromatic diacids, Macromolecules 22 (1989) 3200-3204. 48 Domb, A. J., Synthesis and characterization of biodegradable aromatic anhydride copolymers, Macromolecules 25 (1992) 12-17. 49 Domb, A. J., Maniar, M., Absorbable biopolymers derived from dimer fatty acids, J. Pol. Sci: Part A: Pol. Chem. 31 (1993) 1275-1285. 50 Teomim, D., Nyska, A., and Domb, A. J., Ricinoleic acid-based biopolymers, J. Biomed. Mater. Res. 45 (1999) 258-287. 51 Teomim, D., Mäder, K., Betolila A., Magora, A., and Domb, A. J., γ-Irradiation stability of saturated and unaturated aliphatic polyanhydrides - Ricinoleic acid based polymers, Biomacromolecules 2 (2001) 1015-1022. 52 Teomim, D. and Domb, A. J, Nonlinear fatty acid terminated polyanhydrides, Biomacromolecules 2 (2001) 37-44. 53 Domb, A., Mathiowitz, E., Ron, E., Giannos, S., and Langer, R., Polyanhydrides IV. Unsaturated and crosslinked polyanhydrides, J. Pol. Sci: Part A: Pol. Chem. 29 (1991) 571-579. 54 Muggli, D. S., Burkoth, A. K., Keyser. S. A., Lee, H. R., and Anseth, K. S., Reaction behavior of biodegradable, photo-cross-linkable polyanhydrides, Macromolecules 31 (1998) 4120-4125.
50 55 Young, J. S., Gonzales, K. D., and Anseth, K. S., Photopolymers in orthopedics: characterization of novel crosslinked polyanhydrides, Biomaterials 21 (2000) 1181-1188. 56 Quick, D. J., Macdonald, K. K., and Anseth, K. S., Delivering DNA from crosslinked, surface eroding polyanhydrides, J. Controlled Release 97 (2004) 333-343. 57 Shastri, V. P., Padera, R. F., Tarcha, P., and Langer, R., A preliminary report on the biocompability of photopolymerizable semi-interpenetrating anhydride networks, Biomaterals 25 (2004) 715-721. 58 Nagata, M., Ioka, E., Synthesis and degradation of novel photocrosslinkable aromatic copolyanhydrides, Eur. Polym. J. 42 (2006) 2617-2622. 59 Nagata, M., Ioka, E., Photocrosslinkable degradable copolyanhydrides made from sebacic acid and 4-hydroxycinnamic acid., Reactive & Functional Polymers 63 (2005) 163-170. 60 Cheng, G., Aponte, M. A., and Ramirez, C. A., Cross-linked amino acid-containing polyanhydrides for controlled release applications, Polymer 45 (2004) 3157-3162. 61 Staubli, A., Ron, E., Langer, R., Hydrolytically degradable amino acid containing polymers, J. Am. Chem. Soc. 112 (1991) 2283-2290. 62 Staubli, A., Mathiowitz, E., Lucarelli, M, and Langer, R., Characterization of hydrolytically degradable amino acid containing poly(anhydride-co-imides), Macromolecules 24 (1991) 2283-2290. 63 Schmeltzer, R. C., Anastasiou, T. J., and Uhrich, K. E., Optimized synthesis of salicylate-based poly(anhydride-esters), Polymer Bulletin 49 (2003) 441-448. 64 Schmeltzer, R. C., Schmalenberg, K. E., and Uhrich, K. E., Synthesis and cytotoxicity of salicylate-based poly(anhydride esters), Biomacromolecules 6 (2005) 359-367. 65 Prudencio, A., Schmeltzer, R. C., and K. E., and Uhrich, K. E., Effect of linker structure on salicyli acid-derived poly(anhydride-esters), Macromolecules 38 (2005) 6895-6901. 66 Schmeltzer, R. C., and Uhrich, K. E., Synthesis and characterization of anticeptic-based poly(anhydride-esters), Polymer Bulletin 57 (2006) 281-291. 67 Pinther, P., Hartmann, M., Synthesis of polyanhydrides containing ester groups, Makromol. Chem., Rapid Commun. 11 (1990) 403-408. 68 Jiang, H. L., Zhu, K. J., Synthesis, characterization and in vitro degradation of a new family of alternate poly(ester-anhydrides) based on aliphatic and aromatic diacids, Biomaterials 22 (2001) 211-218. 69 Zhang, T., Xu, M., Chen, H., and Yu, X., Synthesis, degradation, and drug delivery of cycloaliphatic poly(ester-anhydrides), J. Appl. Pol. Sci. 86 (2002) 2509-2514. 70 Krasko, M. Y., Shikanov, A., Ezra, A.., and Domb, A. J., Poly(ester anhydride)s prepared by the insertion of ricinoleic acid into poly(sebacic acid), J. Polym. Sci: Part A: Polym. Chem. 41 (2003) 1059-1069. 71 Krasko, M. Y., Ezra, A.., and Domb, A. J., Lithocholic-acid-containing poly(ester-anhydrides), Polym. Adv. Technol. 14 (2003) 832-838. 72 Kricheldorf, H. R., Lübbers, D., New polymer syntheses, 44, Synthesis of polyanhydrides from silylated dicarboxylic acids and thionyl chloride, Makromol. Chem., Rapid Commun. 11 (1990) 261-266. 73 Kricheldorf, H. R., Lübbers, D., Polyanhydrides 1. Synthesis of thermotropic poly(ester-anhydride)s from
51 terephthalic acid and hydroxyacids by means of acetic anhydride, Eur. Polym. J. 12 (1992) 1397-1400. 74 Slivniak, R., Domb, A. J., Stereocomplexes of enantiomeric lactic acid and sebacic acid ester-anhydride triblock copolymers, Biomacromolecules 3 (2002) 754-760. 75 Xiao, C., Zhu, K. J., Synthesis and in vitro degradation of poly(alkylene carbonate anhydride)], Macromol. Rapic Commun. 21 (2000) 1113-1115. 76 Xiao, C., Zhu, K. J., Synthesis and in vitro degradation properties of poly[(tetramethylene carbonate)-co-(sebacic anhydride)], Polym. Int. 50 (2001) 414-420. 77 Storey, R. F., Taylor, A. E., Synthesis of novel biodegradable poly(ester-anhydride)s, J.M.S.-Pure Appl. Chem. 34 (1997) 265-280. 78 Miao, H., Fan, Y., Liu, Y., Hao, J., and Deng, X., Biodegradable poly(sebacic anhydride-co-ε-caprolactone) multi-block copolymers: Synthesis, characterization, crystallinity and crystalline morphology, Eur. Polym, J. 43 (2007) 1055-1064. 79 Gao, J., Niklason, L., Zhao, X.-M., and Langer, R., Surface modification of polyanhydride microspheres, J. Pharm. Sci. 87 (1998) 246-248. 80 Pfeifer, B. A., Burdick, J. A., and Langer, R., Formulation and surface modification of poly(ester-anhydride) micro- and nanosphres, Biomaterials 26 (2005) 117-124. 81 Pfeifer, B. A., Burdick, J. A., and Langer, R., Poly(ester-anhydride): poly(β-amino ester) micro- and nanospheres: DNA encapsulation and cellular transfection, Int. J. Pharm. 34 (2005) 210-219. 82 Lang, M. and Chu, C.-C., Functionalized multiarm poly(ε-caprolactone)s: Synthesis, structure analysis, and network formation, J. Appl. Sci. 86 (2002) 2296-2306. 83 Burgath, A., Sunder, A., Neuner, A., Mülhaupt R., and Frey, H., Multi-arm star block copolymers based on ε-caprolactone with hyperbranched polyglycerol core, Macromol. Chem. Phys. 201 (2000) 792-797. 84 McNeill, I. C. and Leiper, H. A., Degradation studies of some polyesters and polycarbonates – 2. Polylactide: Degradation under isothermal conditions, Thermal degradation mechanism and photolysis of the polymer, Polym. Degrad. Stab. 11 (1985) 309-326. 85 Jamshidi, K., Hyon, S.-H., and Ikada, Y., Thermal characterization of polylactides, Polymer 29 (1988) 2229-2234. 86 Pitet, L. M., Hait, S. B., Lanyk, T. J., and Knauss, D. M., Linear and branched architectures from the polymerization of lactide with glycidol, Macromolecules 40 (2007) 2327-2334. 87 Kricheldorf, H. R., Kreiser-Saunders, I., Boetttcher, C., Polylactones: 31. Sn(II)octoate-initiated polymerization of L-lactide: a mechanistic study, Polymer 36 (1995) 1253-1259. 88 Orita, A., Tanahashi, C., Kakuda, A., and Otera, J., Highly powerful and practical acylation of alcohols with acid anhydride catalyzed Bi(OTf)3, J. Org. Chem. 66 (2001) 8926-8934. 89 Gimenez, V., Mantcon, J. C., and Cadiz, V., Poly(vinyl alcohol) modified with carboxylic acid anhydrides: crosslinking through carboxylic groups, J. Appl. Pol. Sci. 65 (1997) 1643 – 1651. 90 Turunen, M. P. K., Korhonen, H., Tuominen, J., and Seppälä, J. V., synthesis, characterization and crosslinking of functional star-shaped poly(ε-caprolactone), Polym. Int. 51 (2001) 92-100. 91 Helminen, A., Korhonen, H., and Seppälä, J. V., Biodegradable crosslinked polymers based on triethoxysilane
52 terminated polylactide oligomers, Polymer 42 (2001) 3345-3353. 92 Helminen, A., Korhonen, H., and Seppälä, J. V., Structure modification and crosslinking of methacrylated polylactide oligomers, J. Appl. Polym. Sci 86 (2002) 3616-3624. 93 Helminen, A., Korhonen, H., and Seppälä, J. V., Crosslinked poly(ε-caprolactone/D,L-lactide) copolymers with elastic properties, Macromol. Chem. Phys. 203 (2002) 2630-2639. 94 Kim, B. S., Hrkach, J. S., and Langer, R., Synthesis and characterization of novel degradable photocrosslinked poly(ether-anhydride) networks, J. Polym. Sci. Part. A: Polym. Chem. 38 (2000) 1277-1282. 95 D’Emanuele, A., Hill, J., Tamada, J. A., Domb, A. J., and Langer, R., Molecular weight changes in polymer erosion, Pharmaceutical Reserch 9 (1992) 1279-1283.
ISBN 978-951-22-9219-6ISBN 978-951-22-9220-2 (PDF)ISSN 1795-2239ISSN 1795-4584 (PDF)
