Effect of linker structure on salicylic acid-derived poly (anhydride - Esters)

Article abstract of DOI:10.1021/ma048051u

A series of salicylic acid-derived poly(anhydride-esters) were synthesized by melt polymerization methods, in which the structures of the molecule ("linker") linking together the two salicylic acids were varied. To determine the relationship between the l

Full text of DOI:10.1021/ma048051u

Macromolecules 2005, 38, 6895-6901
6895
Effect of Linker Structure on Salicylic Acid-Derived
Poly(anhydride-esters)
Almudena Prudencio, Robert C. Schmeltzer, and Kathryn E. Uhrich*
Department of Chemistry and Chemical Biology, Rutgers University,
Piscataway, New Jersey 08854-8087
Received September 21, 2004; Revised Manuscript Received June 5, 2005
ABSTRACT: A series of salicylic acid-derived poly(anhydride-esters) were synthesized by melt polym-
erization methods, in which the structures of the molecule (“linker”) linking together the two salicylic
acids were varied. To determine the relationship between the linker and the physical properties of the
corresponding poly(anhydride-ester), several linkers were evaluated including linear aliphatic, aromatic,
and aliphatic branched structures. For the linear aliphatic linkers, higher molecular weights were obtained
with longer linear alkyl chains. The most sterically hindered linkers yielded lower molecular weight
polymers. The thermal decomposition temperature increased with the alkyl chain length, but the glass
transition temperature decreased, due to the enhanced flexibility of the polymer. The highest glass
transition temperatures were obtained by using aromatic linkers as a result of increased π-π interactions.
Water contact angles determined the relative hydrophobicity of the polymers, which correlated to hydrolytic
degradation rates; i.e., the highest contact angle values yielded the slowest degrading polymers.
Introduction
simply homopolymers of salicylic acid34,35 (i.e., no linker
molecules) such that release of salicylate cannot be
readily modified.
Based on our success with salicylic acid-derived poly-
(anhydride-esters),5,24,25,36-43 alternate linkers to com-
pound 3f were investigated for inclusion into the
polymeric backbone to modify the physical properties
of the polymers as well as their degradation profiles.
The linkers were chosen on the basis of their chemical
structure and ranged from linear aliphatic and branched
aliphatic to aromatic compounds (Table 1).
By incorporating these linker compounds, the polymer
backbone delivers a high drug load of the chemically
incorporated drug (salicylic acid), from 62 up to 74 wt
%. This paper further describes how the polymers
incorporating the linkers of Table 1 degraded into
salicylic acid (2) and the corresponding carboxylic acid
(3a-j) (Table 1) as a function of the linker molecule
structure. The relationship between the linker structure
and the physical properties of the corresponding poly-
mer were assessed with respect to drug delivery;
ultimately, our goal is to control polymer degradation
and concurrent drug release by controlling the polymer
composition.
Biodegradable polyanhydrides have been investigated
for use in medical applications such as biomaterials1-3
for tissue scaffolds4,5 and drug delivery devices.3,6-10
Polyanhydrides can undergo in vitro and in vivo hydro-
lytic bond cleavage to form water-soluble biocompatible
degradation products.11-13 In addition, the degradation
rates of these polymers can be controlled by modifying
the polymer composition.14-16 Similar to poly(ortho
esters), polyanhydrides generally undergo predomi-
nantly surface erosion,13,17-20 in contrast to many other
degradable polymers which are predominantly bulk-
eroding.15,16,21-23 This aspect is particularly relevant for
controlled drug delivery as the degradation rate is
surface area dependent and can be modified by manipu-
lating the sample geometry.14
Previously, our laboratory reported the synthesis of
poly(anhydride-esters) (1f) comprised of salicylic acid
(2) as novel degradable biomaterials.24-26 An important
feature of homopolymer 1f is that upon hydrolysis of
the ester and anhydride bonds in the polymer backbone
salicylic acid (2), a therapeutically active compound, and
sebacic acid (3f) are released as outlined in Scheme 1.
Polymer 1f is unique among polymeric biomaterials
in that the drug is chemically incorporated into the
polymer backbone, not attached as a side group.27-32
This characteristic has three significant advantages.
First, a high load of drug can be incorporated into the
polymer; for example, ∼62 wt % for the polymer with
sebacic acid linker (1f) is the drug, salicylic acid. Second,
drug release is a function of the biocompatible linker14
(e.g., compound 3f in Scheme 1), which we define as the
structure that connects the two salicylate units in
polymer 1f. Third, the polymer completely degrades
because of the hydrolytically labile anhydride and ester
bonds within the backbone. Examples of other salicy-
late-based polymers are found with low concentrations
of salicylate (∼20 wt %) in the backbone33 or which are
Experimental Section
Materials. Tetrahydrofuran (THF), pyridine, acetic anhy-
dride, methylene chloride, and diethyl ether were purchased
from Fisher (Fair Lawn, NJ). All other fine chemicals and
solvents were obtained from Aldrich (Milwaukee, WI) and used
as received.
Methods. Proton nuclear magnetic resonance (1H NMR)
spectra were recorded on a Varian 200 MHz spectrometer. The
samples (5-10 mg) were dissolved in deuterated solvents
(DMSO-d6), which was also used as the internal reference.
Infrared (IR) spectra were measured on a Thermo Nicolet/
Avatar 360 FT IR spectrometer, by depositing samples onto
NaCl plates (if liquid) or solvent-casting samples from meth-
ylene chloride onto NaCl plates (if solid).
Weight-averaged molecular weights (Mw) were determined
by gel permeation chromatography (GPC) on a Perkin-Elmer
liquid chromatography system consisting of a series 200
* Corresponding
rutchem.rutgers.edu.
author.
E-mail
uhrich@
10.1021/ma048051u CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/15/2005

6896 Prudencio et al.
Macromolecules, Vol. 38, No. 16, 2005
Scheme 1. Hydrolytic Degradation of Salicylic Acid-Derived Poly(anhydride-esters), 1
Table 1. “Linker” Structures of R in Polymer 1
Sessile-drop contact angles were measured on polymer-
coated coverslips with a model 100 goniometer (Rame-Hart,
Mountain Lakes, NJ). Polymers (100 mg) were dissolved in
methylene chloride (1 mL), and 2-3 drops were added to the
glass coverslips that were spun at 1800 rpm for 40 s using a
photoresist spinner (Headway Research, Garland, TX).
Poly(anhydride-esters) Precursors: Diacid Synthesis
(5). Diacids were prepared by reaction of salicylic acid with
the appropriate acyl chloride in the presence of a base
(pyridine). The preparation of 5a is provided as an example.
Salicylic acid (2) (3.1 g, 22 mmol) was dissolved in THF (40
mL) and pyridine (3.6 mL, 44 mmol). Glutaryl chloride (4a)
(1.4 mL, 11 mmol) dissolved in tetrahydrofuran (10 mL) was
added dropwise over 5 min to the stirring reaction mixture at
room temperature to afford a suspension. The reaction was
stirred for 2 h at room temperature, poured over water (400
mL), and acidified to pH ∼ 2 using concentrated hydrochloric
acid while stirring. The off-white solid (diacid, 5) that formed
was isolated by vacuum filtration, washed with water (3 ×
100 mL), and dried overnight under vacuum at room temper-
ature.
1,5-Bis(o-carboxyphenoxy)pentanoate (5a). Yield: 71%
(off-white powder). 1H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH),
7.65 (t, 2H, ArH), 7.40 (t, 2H, ArH), 7.25 (d, 2H, ArH), 2.75 (t,
4H, CH2), 2.50 (m, 2H, CH2). IR (NaCl, cm-1): 3600-3100 (OH,
COOH), 1760 (CdO, ester), 1700 (CdO, COOH). Anal. Calcd:
C, 61.3%; H, 4.3%; O, 34.4%. Found: C, 61.2%; H, 4.3%; O,
34.5%; mp 132-133 °C.
1,6-Bis(o-carboxyphenoxy)hexanoate (5b). Yield: 92%
(white powder). 1H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH), 7.65
(t, 2H, ArH), 7.40 (t, 2H, ArH), 7.20 (d, 2H, ArH), 2.65 (t, 4H,
CH2), 1.75 (m, 4H, CH2). IR (NaCl, cm-1): 3600-3000 (OH,
COOH), 1730 (CdO, ester), 1650 (CdO, COOH). Anal. Calcd:
C, 62.2%; H, 4.7%; O, 33.1%. Found: C, 61.6%; H, 4.8%; O,
34.0%; mp 174-176 °C.
1,7-Bis(o-carboxyphenoxy)heptanoate (5c). Yield: 80%
(white powder). 1H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH), 7.65
(t, 2H, ArH), 7.50 (t, 2H, ArH), 7.20 (d, 2H, ArH), 2.60 (t, 4H,
CH2), 1.70 (m, 4H, CH2), 1.50 (m, 2H, CH2). IR (NaCl, cm-1):
3650-3100 (OH, COOH), 1700 (CdO, ester), 1650 (CdO,
COOH). Anal. Calcd: C, 63.0%; H, 5.0%; O, 32.0%. Found: C,
62.8%; H, 5.0%; O, 32.3%; mp 131-133 °C.
1,8-Bis(o-carboxyphenoxy)octanoate (5d). Yield: 89%
(white powder). 1H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH), 7.65
(t, 2H, ArH), 7.40 (t, 2H, ArH), 7.20 (d, 2H, ArH), 2.60 (t, 4H,
CH2), 1.65 (m, 4H, CH2), 1.40 (m, 4H, CH2). IR (NaCl, cm-1):
3650-3100 (OH, COOH), 1760 (CdO, ester), 1700 (CdO,
COOH). Anal. Calcd: C, 63.8%; H, 5.3%; O, 30.9%. Found: C,
63.5%; H, 5.4%; O, 30.3%; mp 141-143 °C.
refractive index detector, a series 200 LC pump, and an ISS
200 advanced sample processor. A Dell OptiPlex GX110
computer running Perkin-Elmer TurboChrom 4 software was
used for data collection and processing and to automate the
analysis via Perkin-Elmer Nelson 900 series interface and 600
series link. Polymers were dissolved in methylene chloride (5
mg/mL) and filtered through 0.45 µm poly(tetrafluoroethylene)
(PTFE) syringe filters (Whatman, Clifton, NJ) before elution.
Samples were resolved on a Jordi divinylbenzene mixed-bed
GPC column (7.8 × 300 mm) (Alltech Associates, Deerfield,
IL) at 25 °C, with methylene chloride as eluent at a flow rate
of 0.5 mL/min. Molecular weights were calibrated relative to
narrow molecular weight polystyrene standards (Polysciences,
Dorval, Canada).
Thermal analyses were performed on a Perkin-Elmer sys-
tem consisting of Pyris 1 DSC and TGA 7 analyzers with TAC
7/DX instrument controllers. Perkin-Elmer Pyris software was
used for data collection on a Dell OptiPlex GX110 computer.
For DSC, samples (5 mg) were heated under dry nitrogen gas.
Data were collected at heating and cooling rates of 10 °C/min
with a two-cycle minimum. Glass transition temperatures were
calculated as half Cp extrapolated. For TGA, samples (10 mg)
were heated under dry nitrogen gas. Data were collected at a
heating rate of 10 °C/min. Decomposition temperatures were
defined as the onset of decomposition.
1,9-Bis(o-carboxyphenoxy)nonanoate (5e). Yield: 90%
(off-white powder). 1H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH),
7.65 (t, 2H, ArH), 7.40 (t, 2H, ArH), 7.20 (d, 2H, ArH), 2.55 (t,
4H, CH2), 1.65 (m, 4H, CH2), 1.40 (m, 6H, CH2). IR (NaCl,
cm-1): 3650-3100 (OH, COOH), 1760 (CdO, ester), 1700 (Cd
Elemental analyses were provided by QTI (Whitehouse, NJ).
Melting points below 200 °C were obtained with a Mel-Temp
apparatus at a heating rate of 1 °C/min, while those above
200 °C were determined on the Pyris 1 DSC (see above).

Macromolecules, Vol. 38, No. 16, 2005
Salicylic Acid-Derived Poly(anhydride-esters) 6897
O, COOH). Anal. Calcd: C, 64.5%; H, 5.6%; O, 29.1%. Found:
C, 64.0%; H, 5.6%; O, 29.3%; mp 101-103 °C.
1.35 (b, 8H, CH2). IR (NaCl, cm-1): 1830 (CdO, anhydride),
1730 (CdO, ester). Td ) 345 °C.
1,10-Bis(o-carboxyphenoxy)decanoate (5f). Yield: 91%
(white powder). 1H NMR (DMSO-d6): δ 7.90 (d, 2H, ArH), 7.60
(t, 2H, ArH), 7.40 (t, 2H, ArH), 7.20 (d, 2H, ArH), 2.55 (t, 4H,
CH2), 1.65 (m, 4H, CH2), 1.35 (b, 8H, CH2). IR (NaCl, cm-1):
3500-3000 (OH, COOH), 1760 (CdO, ester), 1700 (CdO,
COOH). Anal. Calcd: C, 65.2%; H, 5.9%; O, 28.9%. Found: C,
64.5%; H, 5.8%; O, 29.7%; mp 128-129 °C.
p-Bis(o-carboxyphenoxy)benzoate Monomer (6g).
Yield: quantitative (white solid). 1H NMR (DMSO-d6): δ 8.35
(s, 4H, ArH), 8.05 (d, 2H, ArH), 7.75 (t, 2H, ArH), 7.50 (t, 2H,
ArH), 7.45 (d, 2H, ArH), 2.25 (s, 6H, CH3). IR (NaCl, cm-1):
1810 (CdO, anhydride), 1740 (CdO, ester). Td ) 354 °C.
m-Bis(o-carboxyphenoxy)benzoate Monomer (6h).
Yield: quantitative (white solid). 1H NMR (DMSO-d6): δ 8.85
(s, 1H, ArH), 8.55 (d, 2H, ArH), 8.15 (d, 2H, ArH), 7.90 (t, 1H,
ArH), 7.75 (t, 2H, ArH), 7.60 (t, 2H, ArH), 7.55 (d, 2H, ArH),
2.25 (s, 6H, CH3). IR (NaCl, cm-1): 1800 (CdO, anhydride),
1740 (CdO, ester). Td ) 329 °C.
2,2′-Bis(o-carboxyphenoxy)isopropanoate Monomer
(6i). Yield: quantitative (white solid). 1H NMR (DMSO-d6):
δ 8.10 (d, 2H, ArH), 7.85 (t, 2H, ArH), 7.55 (t, 2H, ArH), 7.35
(d, 2H, ArH), 2.35 (s, 6H, CH3), 1.70 (s, 6H, CH3). IR (NaCl,
cm-1): 1810 (CdO, anhydride), 1760 (CdO, ester). Td ) 275
°C.
p-Bis(o-carboxyphenoxy)benzoate (5g). Yield: 92% (white
1
solid). H NMR (DMSO-d6): δ 8.35 (s, 4H, ArH), 8.05 (d, 2H,
ArH), 7.75 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.45 (d, 2H, ArH).
IR (NaCl, cm-1): 3300-3000 (OH, COOH), 1740 (CdO, ester),
1700 (CdO, COOH). Anal. Calcd: C, 65.0%; H, 3.5%; O, 31.5%.
Found: C, 64.6%; H, 3.4%; O, 31.3%; mp 212 °C.
m-Bis(o-carboxyphenoxy)benzoate (5h). Yield: 90%
1
(white solid). H NMR (DMSO-d6): δ 8.80 (s, 1H, ArH), 8.45
(d, 2H, ArH), 8.00 (d, 2H, ArH), 7.85 (t, 1H, ArH), 7.75 (t, 2H,
ArH), 7.45 (t, 2H, ArH), 7.40 (d, 2H, ArH). IR (NaCl, cm-1):
3650-3100 (OH, COOH), 1740 (CdO, ester), 1700 (CdO,
COOH). Anal. Calcd: C, 65.0%; H, 3.5%; O, 31.5%. Found: C,
64.3%; H, 3.7%; O, 32.0%; mp 415 °C.
3,3′-Bis(o-carboxyphenoxy)pentanoate Monomer (6j).
Yield: quantitative (white solid). 1H NMR (DMSO-d6): δ 8.10
(d, 2H, ArH), 7.85 (t, 2H, ArH), 7.60 (t, 2H, ArH), 7.40 (d, 2H,
ArH), 2.35 (s, 6H, CH3), 2.20 (q, 4H, CH2), 1.0 (t, 6H, CH3). IR
(NaCl, cm-1): 1810 (CdO, anhydride), 1760 (CdO, ester). Td
) 284 °C.
2,2′-Bis(o-carboxyphenoxy)isopropanoate (5i). Yield:
1
84% (white solid). H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH),
7.70 (t, 2H, ArH), 7.45 (t, 2H, ArH), 7.25 (d, 2H, ArH), 1.70 (s,
6H, CH3). IR (NaCl, cm-1): 3300-3000 (OH, COOH), 1755 (Cd
O, ester), 1700 (CdO, COOH). Anal. Calcd: C, 61.3%; H, 4.3%;
O, 34.1%. Found: C, 61.2%; H, 4.3%; O, 34.1%; mp 188 °C.
3,3′-Bis(o-carboxyphenoxy)pentanoate (5j). Yield: 80%
Polymer Synthesis (1). Monomer (6) (3.0 g) was placed
in a 100 mL two-necked round-bottom flask with 24/40 joints
with a vacuum joint in one neck and a Teflon vacuum-stirring
adapter in the other. The reaction flask was heated to 160-
180 °C using a temperature controller (Cole Parmer) in a
silicone oil bath under high vacuum (<2 mmHg). During this
time, the melt was actively stirred at ∼100 rpm by an overhead
stirrer (T-line Laboratory Stirrer, Talboys Engineering, Mon-
trose, PA). Polymerization was complete when the viscosity
of the melt remained constant and/or solidified (3-6 h). The
polymer was cooled to room temperature and isolated by
precipitation in methylene chloride/diethyl ether (5 mL/100
mL). Polymers were soluble in organic solvents such as
tetrahydrofuran (THF), dimethylformamide (DMF), dimethyl
sulfoxide (DMSO), and chlorinated solvents such as chloroform
and methylene chloride.
1
(white solid). H NMR (DMSO-d6): δ 7.95 (d, 2H, ArH), 7.65
(t, 2H, ArH), 7.45 (t, 2H, ArH), 7.25 (d, 2H, ArH), 2.20 (q, 4H,
CH2), 1.00 (t, 6H, CH3). IR (NaCl, cm-1): 3600-3100 (OH,-
COOH), 1750 (CdO, ester), 1700 (CdO, COOH). Anal. Calcd:
C, 63.0%; H, 5.0%; O, 32.0%. Found: C, 63.1%; H, 5.0%; O,
32.2%; mp 157-158 °C.
Monomer Synthesis (6). The diacid (5) (3.0 g) was
activated with stirring in an excess of acetic anhydride (50
mL) at room temperature until the initial suspension becomes
a clear solution (∼3-12 h). The excess acetic anhydride is
removed by rotoevaporation at room temperature under
vacuum to afford the monomer.
1,5-Bis(o-carboxyphenoxy)pentanoate Monomer (6a).
Yield: quantitative (colorless oil). 1H NMR (DMSO-d6): δ 8.10
(d, 2H, ArH), 7.80 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.35 (d, 2H,
ArH), 2.75 (t, 4H, CH2), 2.40 (s, 6H, CH3), 2.00 (m, 2H, CH2).
IR (NaCl, cm-1): 1810 (CdO, anhydride), 1760 (CdO, ester).
Td ) 217 °C.
Poly[1,5-bis(o-carboxyphenoxy)pentanoate] (1a). Tpolym
) 160 °C. Yield: quantitative (brown solid). 1H NMR (DMSO-
d6): δ 8.10 (b, 2H, ArH), 7.80 (b, 2H, ArH), 7.45 (b, 2H, ArH),
7.35 (b, 2H, ArH), 2.65 (b, 4H, CH2), 1.90 (b, 2H, CH2). IR
(NaCl, cm-1): 1760, 1700 (CdO, anhydride), 1730 (CdO,
ester). Mw ) 12 600, PDI ) 1.8. Td ) 243 °C, Tg ) 50 °C.
Contact angle: 75°.
Poly[1,6-bis(o-carboxyphenoxy)hexanoate] (1b). Tpolym
) 160 °C. Yield: quantitative (brown solid). 1H NMR (DMSO-
d6): δ 8.15 (b, 2H, ArH), 7.75 (b, 2H, ArH), 7.40 (b, 4H, ArH),
2.60 (b, 4H, CH2), 1.60 (b, 4H, CH2). IR (NaCl, cm-1): 1780,
1700 (CdO, anhydride), 1740 (CdO, ester). Mw ) 18 400, PDI
) 2.6. Td ) 292 °C, Tg ) 59 °C. Contact angle: 77°.
Poly[1,7-bis(o-carboxyphenoxy)heptanoate] (1c). Tpolym
) 180 °C. Yield: quantitative (light brown solid). 1H NMR
(DMSO-d6): δ 8.15 (b, 2H, ArH), 7.75 (b, 2H, ArH), 7.40 (b,
2H, ArH), 7.25 (b, 2H, ArH), 2.60 (b, 4H, CH2), 1.60 (b, 6H,
CH2). IR (NaCl, cm-1): 1780, 1720 (CdO, anhydride), 1740
(CdO, ester). Mw ) 29 700, PDI ) 2.6. Td ) 313 °C, Tg ) 40
°C. Contact angle: 88°.
Poly[1,8-bis(o-carboxyphenoxy)octanoate] (1d). Tpolym
) 180 °C. Yield: quantitative (colorless solid). 1H NMR
(DMSO-d6): δ 8.10 (b, 2H, ArH), 7.80 (b, 2H, ArH), 7.50 (b,
2H, ArH), 7.30 (b, 2H, ArH), 2.60 (b, 4H, CH2), 1.50 (b, 4H,
CH2), 1.20 (b, 4H, CH2). IR (NaCl, cm-1): 1775, 1725 (CdO,
anhydride), 1760 (CdO, ester). Mw ) 30 200, PDI ) 1.6. Td )
341 °C, Tg ) 30 °C. Contact angle: 87°.
Poly[1,9-bis(o-carboxyphenoxy)nonanoate] (1e). Tpolym
) 180 °C. Yield: quantitative (pale yellow solid). 1H NMR
(DMSO-d6): δ 8.10 (b, 2H, ArH), 7.75 (b, 2H, ArH), 7.45 (b,
2H, ArH), 7.25 (b, 2H, ArH), 2.65 (b, 4H, CH2), 1.55 (b, 4H,
CH2), 1.25 (b, 6H, CH2). IR (NaCl, cm-1): 1790, 1730 (CdO,
anhydride), 1750 (CdO, ester). Mw ) 35 700, PDI ) 1.9. Td )
342 °C, Tg ) 24 °C. Contact angle: 88°.
1,6-Bis(o-carboxyphenoxy)hexanoate Monomer (6b).
Yield: quantitative (yellowish oil). 1H NMR (DMSO-d6): δ 8.10
(d, 2H, ArH), 7.80 (t, 2H, ArH), 7.55 (t, 2H, ArH), 7.35 (d, 2H,
ArH), 2.65 (t, 4H, CH2), 2.35 (s, 6H, CH3), 1.75 (m, 4H, CH2).
IR (NaCl, cm-1): 1800 (CdO, anhydride), 1760 (CdO, ester).
Td ) 274 °C.
1,7-Bis(o-carboxyphenoxy)heptanoate Monomer (6c).
1
Yield: quantitative (yellow oil). H NMR (DMSO-d6): δ 8.05
(d, 2H, ArH), 7.80 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.35 (d, 2H,
ArH), 2.65 (t, 4H, CH2), 2.35 (s, 6H, CH3), 1.70 (m, 4H, CH2),
1.50 (m, 2H, CH2). IR (NaCl, cm-1): 1820 (CdO, anhydride),
1760 (CdO, ester). Td ) 310 °C.
1,8-Bis(o-carboxyphenoxy)octanoate Monomer (6d).
Yield: quantitative (white solid). 1H NMR (DMSO-d6): δ 8.05
(d, 2H, ArH), 7.80 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.35 (d, 2H,
ArH), 2.65 (t, 4H, CH2), 2.35 (s, 6H, CH3), 1.70 (m, 4H, CH2),
1.45 (m, 4H, CH2). IR (NaCl, cm-1): 1810 (CdO, anhydride),
1760 (CdO, ester). Td ) 359 °C.
1,9-Bis(o-carboxyphenoxy)nonanoate Monomer (6e).
Yield: quantitative (colorless oil). 1H NMR (DMSO-d6): δ 8.10
(d, 2H, ArH), 7.80 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.35 (d, 2H,
ArH), 2.65 (t, 4H, CH2), 2.35 (s, 6H, CH3), 1.65 (m, 4H, CH2),
1.35 (m, 6H, CH2). IR (NaCl, cm-1): 1820 (CdO, anhydride),
1730 (CdO, ester). Td ) 345 °C.
1,10-Bis(o-carboxyphenoxy)decanoate Monomer (6f).
Yield: quantitative (colorless oil). 1H NMR (DMSO-d6): δ 8.05
(d, 2H, ArH), 7.80 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.30 (d, 2H,
ArH), 2.65 (t, 4H, CH2), 2.35 (s, 6H, CH3), 1.65 (m, 4H, CH2),

6898 Prudencio et al.
Macromolecules, Vol. 38, No. 16, 2005
Scheme 2. Synthetic Scheme for Salicylic Acid-Derived Poly(anhydride-esters), 1
Poly[1,10-bis(o-carboxyphenoxy)decanoate] (1f). Tpolym
) 180 °C. Yield: quantitative (pale tan solid). 1H NMR
(DMSO-d6): δ 8.10 (b, 2H, ArH), 7.75 (b, 2H, ArH), 7.30 (b,
4H, ArH), 2.45 (b, 4H, CH2), 1.55 (b, 4H, CH2), 1.25 (b, 8H,
CH2). IR (NaCl, cm-1): 1790, 1740 (CdO, anhydride), 1760
(CdO, ester). Mw ) 43 000, PDI ) 1.6. Td ) 351 °C, Tg ) 27
°C. Contact angle: 88°.
hydroxide and/or 1 N hydrochloric acid solutions, and pH
measurements were performed on a pH meter (VWR Scientific,
San Francisco, CA).
In Vitro Hydrolytic Degradation. Hydrolytic degradation of
the polymers was performed by placing the disks in 20 mL
Wheaton PET plastic scintillation vials (Fisher, Fair Lawn,
NJ) with 10 mL of PBS and incubating them at 37 °C with
agitation using a controlled environment incubator-shaker
(New Brunswick Scientific Co., Edison, NJ) at 60 rpm for 30
days. Every 24 h, the buffer solution was replaced by fresh
solution (10 mL), and the spent media was analyzed by UV (λ
) 303 nm) with a DU 520 UV/vis spectrophotometer (Beckman
Instruments, Fullerton, CA) to specifically monitor the release
of salicylic acid. Measurements were taken at the maximum
absorbance of salicylic acid (λ ) 303 nm) that did not overlap
with the UV absorbance of the linkers. UV data (average of
three samples per time point) were calibrated against salicylic
Poly[p-bis(o-carboxyphenoxy)benzoate] (1g). Tpolym
)
180 °C. Yield: quantitative (beige solid). 1H NMR (DMSO-
d6): δ 8.15 (b, 6H, ArH), 7.80 (b, 2H, ArH), 7.50 (b, 4H, ArH).
IR (NaCl, cm-1): 1790, 1740 (CdO, anhydride), 1740 (CdO,
ester). Mw ) 17 100, PDI ) 1.8. Td ) 403 °C, Tg ) 118 °C.
Contact angle: 81°.
Poly[m-bis(o-carboxyphenoxy)benzoate] (1h). Tpolym
)
180 °C. Yield: quantitative (beige solid). 1H NMR (DMSO-
d6): δ 8.80 (s, 1H, ArH), 8.50 (d, 2H, ArH), 8.05 (d, 2H, ArH),
7.90 (t, 1H, ArH), 7.75 (t, 2H, ArH), 7.50 (t, 2H, ArH), 7.40 (d,
2H, ArH). IR (NaCl, cm-1): 1800, 1740 (CdO, anhydride), 1740
(CdO, ester). Mw ) 8800, PDI ) 1.7. Td ) 405 °C, Tg ) 106
°C. Contact angle: 77°.
acid solutions of known concentrations (8.4 × 10-4, 1.7 × 10-3
,
3.3 × 10-3, and 6.7 × 10-3 mg/mL). For reference, solutions of
the individual linkers (4) as well as the diacids (5) were
measured to ensure that the UV spectra of the linkers (4) did
not overlap with that of salicylic acid and that the diacids (5)
were not present in the degradation media.
Poly[2,2′-bis(o-carboxyphenoxy)isopropanoate] (1i).
Tpolym ) 160 °C. Yield: quantitative (beige solid). 1H NMR
(DMSO-d6): δ 8.10 (d, 2H, ArH), 7.95 (t, 2H, ArH), 7.55 (t,
2H, ArH), 7.45 (d, 2H, ArH), 1.70 (s, 6H, CH3). IR (NaCl, cm-1):
1780, 1720 (CdO, anhydride), 1760 (CdO, ester). Mw
)
Results and Discussion
39 800, PDI ) 1.5. Td ) 261 °C, Tg ) 44 °C. Contact angle:
95°.
Diacid Synthesis. An outline of the poly(anhydride-
ester) (1) synthesis, starting from salicylic acid (2) and
the appropriate acyl chloride of the linker (4), is shown
in Scheme 2.
Poly[3,3′-bis(o-carboxyphenoxy)pentanoate] (1j). Tpolym
1
) 160 °C. Yield: quantitative (beige solid). H NMR (DMSO-
d6): δ 8.10 (d, 2H, ArH), 7.90 (t, 2H, ArH), 7.60 (t, 2H, ArH),
7.40 (d, 2H, ArH), 2.20 (q, 4H, CH2), 1.0 (t, 6H, CH3). IR (NaCl,
cm-1): 1775, 1725 (CdO, anhydride), 1750 (CdO, ester). Mw
) 16 500, PDI ) 1.6. Td ) 276 °C, Tg ) 23 °C. Contact angle:
93°.
Degradation Studies. Sample Preparation. Polymer pel-
lets were prepared by pressing ground polymers (∼150 ( 5
mg) into 13 mm diameter × 1 mm thick disks in an IR pellet
die (International Crystal Laboratories, Garfield, NJ) with a
benchtop hydraulic press (Carver model M, Wabash, IN). A
pressure of 10 000 psi was applied for 5 min at room temper-
ature. No change in polymer color was observed upon applying
pressure.
The salicylate-based monomer precursor (5) was
formed by direct coupling of salicylic acid (2) with the
corresponding diacyl chloride (4) in an appropriate
solvent (THF) containing pyridine at room tempera-
ture.26 Pyridine deprotonates salicylic acid (2) and also
acts as a catalyst to form an acylpyridinium ion,44 which
reacts with the free phenolate. Thus, formation of the
acylpyridinium ion, which is known to react more
rapidly with alcohols than acyl chlorides,44,45 eliminates
the need to protect the carboxylic acid of the salicylic
acid. This method also eliminates further purification,
except for washing with the appropriate solvent to
remove starting materials and reaction’s byproducts,
because of the large solubility differences. This one-step
Degradation Media Preparation. Degradation media con-
sisted of phosphate buffer solution (PBS) containing 0.1 M
potassium hydrogen phosphate and 0.1 M potassium dihydro-
gen phosphate. The pH was adjusted to 7.4 with 1 M sodium

Macromolecules, Vol. 38, No. 16, 2005
Salicylic Acid-Derived Poly(anhydride-esters) 6899
Figure 2. Influence of linear aliphatic chain length, (CH2)n,
on poly(anhydride-ester) 1a-1f properties.
Table 2. Influence of Aromatic and Branched Linker
Structures on Poly(anhydride-esters) 1g-1j
salicylic
contact
Tg
Td
mol
wt
R
acid (%) angle (deg) (°C) (°C)
terephthalic, g
isophthalic, h
dimethylmalonic, i
diethylmalonic, j
67.5
67.5
73.6
68.5
81
77
95
93
118 403 17 100
106 405 8 800
44 261 39 800
23 276 16 500
Figure 1. Melting point temperatures of diacid (5a-5f) as a
function of alkyl chain length, (CH2)n.
for the glutaric linker (three methylenes) to 351 °C for
the sebacic linker (eight methylenes) (Figure 2).
Glass transition temperatures decreased as the linear
linker length increased due to the enhanced flexibility,
ranging from 59 to 24 °C. The drug loading of the
polymer was also modified; drug loads ranged from 60
to 74% for the longest alkyl chain (Figure 1, n ) 8) to
the shortest alkyl chain (Figure 1, n ) 3), respectively.
As the number of methylenes increases, the amount of
salicylic acid chemically incorporated in the polymer
backbone decreases.
Influence of Aromatic/Branched Linkers on Poly-
(anhydride-ester) Properties. Polymers with aro-
matic (1g, 1h) and branched alkyl linker structure (1i,
1j) were also synthesized and their properties evaluated.
For consistency, similar polymerization conditions were
used rather than optimizing the reaction conditions for
each polymer. As previously demonstrated5 by our
laboratory and expected for melt condensation polym-
erization, molecular weights can be optimized for each
polymers by systematically modifying polymerization
time and temperature. Among these systems, the poly-
mers with the most sterically hindered linker groups
were the most difficult to polymerize (1h, 1j), thus
yielding lower molecular weights as shown in Table 2.
Polymers with aromatic linkers showed the highest
decomposition temperatures likely due to the increased
π-π interactions between the polymer chains. Steric
bulk decreased the packing efficiency of the polymer
chains, resulting in greater free volume and lowered
glass transition temperatures for the polymers with the
more sterically bulky linker groups. Thus, for the
aromatic linkers, changing the linker substitution pat-
tern from para-substitution (1g) to meta-substitution
(1h) decreases the Tg by 12 °C.
method26 was used to prepared a variety of diacids (5)
with different linkers that will ultimately undergo melt
condensation to yield poly(anhydride-esters). Three
classes of linker molecules (Table 1) were evaluated to
assess the role of the linker chemical structure in the
polymer properties: (1) linear aliphatic, (2) branched
aliphatic, and (3) aromatic compounds.
Isolated yields for the diacids (5) based on linear
aliphatic linkers (5a-5f) ranged from 71 to 91%, and
an odd/even effect was observed in the melting point
(Figure 1). The trend of odd-membered alkanes display-
ing slightly lowered melting points than even-membered
systems is well documented in simple alkyl chains.46
For the diacids based on aromatic and branched
aliphatic linkers (5g-5j), the yields ranged from 80 to
92%. As anticipated, diacids with aromatic linkers (5g,
5h) displayed higher thermal stabilities (i.e., melting
points) than the diacids with aliphatic linkers (5a-5f
and 5i, 5j). Melting points decreased with increased
steric hindrance of the linker molecule due to less
effective intermolecular interactions.
Influence of Linear Aliphatic Linkers on Poly-
(anhydride-ester) Properties. The activation of the
diacid (5a-5f) with acetic anhydride yielded the anhy-
dride monomer (6a-6f), which was melt-polymerized
at 160-180 °C under vacuum (<2 mmHg) for 3-6 h to
yield the poly(anhydride-ester) (1a-1f). The chemical,
physical, and thermal properties of polymers with linear
alkyl chain linkers are summarized in Figure 2.
Molecular weights significantly increased as the alkyl
chain length of the linker increased from three to eight
methylenes, while the polydispersity index (PDI) re-
mained relatively unchanged due to homogeneous mix-
ing of the polymer melt.26 Similarly, decomposition
temperatures (Td) for the polymers are influenced by
the alkyl linker chain length, increasing from 243 °C
Degradation Studies. In vitro hydrolytic degrada-
tion of the polymers was measured by quantifying the

6900 Prudencio et al.
Macromolecules, Vol. 38, No. 16, 2005
angle values and also displayed the slowest degradation
relative to the other poly(anhydride-esters) (1a-1h).
Conclusion
The specific biomedical application for which a poly-
mer is suited depends on several factors, including in
vitro degradation rate, thermal and mechanical proper-
ties, and biocompatibility of the polymer.47 This work
demonstrates that the properties of salicylate-derived
poly(anhydride-esters), 7, were readily modified by
varying the compound linking together the salicylic
acids as summarized below. First, polymers with higher
molecular weight values were obtained by using longer
linear aliphatic linkers. This effect is likely due to
enhanced mobility in the molten state during polymer-
ization. Second, the glass transition temperature (Tg)
concurrently decreased with increasing aliphatic chain
length. Overall, polymers containing aromatic linkers
(1g and 1h) displayed the highest Tg values, even
relative to polymers with branched aliphatic molecules
(1i and 1j). Third, the drug loading of the polymer was
modified by changing the linker, obtaining drug loads
ranging from 60 to 74 wt %. Fourth, hydrolytic degrada-
tion of the polymers to release salicylic acid was
demonstrably a function of the linker structure; poly-
mers with aliphatic linkers generally degraded faster
in vitro than polymers comprised of aromatic or branched
aliphatic linkers. Last, water contact angle values of the
undegraded polymer is a good indicator of in vitro
degradation rate; the polymer (1i) with the highest
contact angle (i.e., most hydrophobic) degraded over a
longer time period.
Figure 3. In vitro degradation profile of poly(anhydride-
esters) with linear aliphatic linkers, 1a-1f.
Figure 4. In vitro degradation profile of poly(anhydride-
esters) with aromatic linkers, 1g and 1h.
We are currently evaluating specific parameters such
as water solubility and log P values of the polymer
degradation products to further optimize the polymer
degradation profiles.
Acknowledgment. The authors thank the National
Institutes of Health (DE 13207) for financial support.
References and Notes
(1) Uhrich, K.; Gupta, A.; Thomas, T.; Laurencin, C.; Langer, R.
Macromolecules 1995, 28, 2184-2193.
(2) Uhrich, K.; Larrier, D.; Laurencin, C.; Langer, R. J. Polym.
Sci., Polym. Chem. Ed. 1996, 34, 1261-1269.
(3) Langer, R. Acc. Chem. Res. 2000, 33, 94-101.
(4) Burkoth, A.; Anseth, K. Biomaterials 2000, 21, 2395-2404.
(5) Campo, C.; Anastasiou, T.; Uhrich, K. Polym. Bull. (Berlin)
1999, 42, 61-68.
Figure 5. In vitro degradation profile of poly(anhydride-
esters) with branched aliphatic linkers, 1i and 1j.
salicylic acid concentration in the degradation media
using UV spectrometry. Polymers 7a-7j were com-
pressed into disks and then degraded at 37 °C in pH
7.4 phosphate buffered solution over 30 days. Polymers
with linear aliphatic linkers (1a-1f) showed similar
release profiles (Figure 3): short lag times (0-3 days)
before salicylic acid was detected in solution and com-
plete degradation of the disk within 8-16 days. In
contrast to polymers with linear aliphatic linkers (1a-
1f), polymers with aromatic linkers (1g, 1h) required
longer degradation time (up to 20 days) (Figure 4)
and also displayed a longer lag time (4 days). Polymers
with branched aliphatic linkers (1i, 1j) degraded much
slower and were only partially degraded in 30 days, with
less than 20% of the drug released at this time point
(Figure 5).
(6) Anastasiou, T.; Uhrich, K. Macromolecules 2000, 33, 6217-
6221.
(7) Mathiowitz, E.; Jacob, J.; Jong, Y.; Carino, G.; Chickering,
D.; Chaturvedi, P.; Santos, C.; Vijayaraghavan, K.; Mont-
gomery, S.; Bassett, M.; Morrell, C. Nature (London) 1997,
386, 410-414.
(8) Chasin, M.; Langer, R. Biodegradable Polymers as Drug
Delivery Systems; Marcel Dekker: New York, 1990.
(9) Brem, H.; Piantadosi, S.; Burger, P.; Walker, M.; Selker, R.;
Vick, N.; Black, K.; Sisti, M.; Brem, S.; Mohr, G.; Muller, P.;
Morawetz, R.; Schold, S. Lancet 1995, 345, 1008-1012.
(10) Hanes, J.; Chiba, M.; Langer, R. Biomaterials 1998, 19, 163-
172.
(11) Leong, K.; Kost, J.; Mathiowitz, E.; Langer, R. Biomaterials
1986, 7, 364-371.
(12) Leong, K.; D’Amore, P.; Maletta, M.; Langer, R. J. Biomed.
Mater. Res. 1986, 20, 51-64.
(13) Sanders, A.; Li, B.; Bieniarz, C.; Harris, F. Polym. Prepr.
1999, 40, 888.
Water contact angles were measured to evaluate the
relative hydrophobicity of the polymers as a factor that
may influence hydrolytic degradation rate of the poly-
mers. Indeed, polymers containing the branched ali-
phatic linkers (1i, 1j) displayed the highest contact
(14) Whitaker-Brothers, K.; Uhrich, K. E. J. Biomed. Mater. Res.,
Part A, in press.
(15) Gopferich, A.; Tessmar, J. Adv. Drug Del. Rev. 2002, 54, 911-
931.
(16) Gopferich, A. Biomaterials 1996, 17, 103-114.

Macromolecules, Vol. 38, No. 16, 2005
Salicylic Acid-Derived Poly(anhydride-esters) 6901
(17) Von Burkersroda, F.; Schedl, L.; Gopferich, A. Biomaterials
2002, 23, 4221-4231.
(33) Kricheldorf, H.; Gerken, A.; Yulchibaev, B.; Friedrich, C. J.
Polym. Sci., Polym. Chem. Ed. 2000, 38, 2013-2022.
(18) Tamada, J.; Langer, R. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 552-556.
(34) Liming, T.; Rabenstein, M.; Kricheldorf, H. Macromol. Chem.
Phys. 2001, 202, 1497-1504.
(19) Quick, D.; Macdonald, K.; Anseth, K. J. Controlled Release
2004, 97, 333-343.
(20) Park, E.; Maniar, M.; Shah, J. J. Controlled Release 1996,
40, 111-121.
(35) Shalaby, S.; Koelmael, D.; Arnold, S. U.S. Patent 5,082,925,
1992.
(36) Schmeltzer, R.; Schmalenberg, K.; Uhrich, K. Biomacromol-
ecules 2005, 6, 359-367.
(21) Gopferich, A. Macromolecules 1997, 30, 2598-2604.
(22) Vogelhuber, W.; Rotunno, P.; Magni, E.; Gazzaniga, A.;
Spruss, T.; Bernhardt, G.; Buschauer, A.; Go¨pferich, A. J.
Controlled Release 2001, 73, 75-88.
(37) Kim, Y.; Uhrich, K. Polym. Prepr. 2004, 45, 505.
(38) Whitaker-Brothers, K.; Uhrich, K. E. J. Biomed. Mater. Res.
2004, 70A, 309-318.
(23) Uhrich, K.; Cannizzaro, S.; Langer, R.; Shakesheff, K. Chem.
Rev. 1999, 99, 3181-3198.
(39) Anastasiou, T.; Uhrich, K. J. Polym. Sci., Polym. Chem. Ed.
2003, 41, 3667-3679.
(24) Erdmann, L.; Uhrich, K. Biomaterials 2000, 20, 1941-1946.
(25) Erdmann, L.; Macedo, B.; Uhrich, K. Biomaterials 2000, 21,
2507-2512.
(40) Anastasiou, T.; Beaton, M.; Uhrich, K. Polym. Prepr. 2001,
42, 121.
(41) Bedell, C.; Deng, M.; Anastasiou, T.; Uhrich, K. J. Appl.
Polym. Sci. 2001, 80, 32-38.
(42) Anastasiou, T.; Uhrich, K. Polym. Prepr. 2000, 41, 1366.
(26) Schmeltzer, R.; Anastasiou, T.; Uhrich, K. Polym. Bull.
(Berlin) 2003, 49, 441-448.
(27) San Roman, J.; Bujan, J.; Bellon, J.; Gallardo, A.; Escudero,
M.; Jorge, E.; de Haro, J.; Alvarez, L.; Castillo-Olivares, J.
J. Biomed. Mater. Res. 1996, 32, 19-27.
(43) Krogh-Jespersen, E.; Anastasiou, T.; Uhrich, K. Polym. Prepr.
2000, 41, 1048.
(28) Sato, H.; Kojima, J.; Nakajima, A.; Morita, T.; Noishiki, Y.;
Gu, Z.; Li, F.; Feng, X. J. Biomater. Sci. Polym. Ed. 1991, 2,
1-13.
(44) Fersht, A.; Jencks, W. J. Am. Chem. Soc. 1970, 92, 5442-
5452.
(45) Hoefle, G.; Steglich, W.; Vorbrueggen, H. Angew. Chem., Int.
(29) Elvira, C.; Gallardo, A.; San Roman, J. Polymer 1999, 40,
6911-6924.
Ed. Engl. 1978, 17, 569-583.
(30) Chafi, N.; Montheard, J.; Vergnaud, J. Int. J. Pharm. 1989,
52, 203-211.
(46) Solomons, T. Organic Chemistry; Wiley & Sons: New York,
1980.
(31) Rivas, B.; Moreno-Villoslada, I.; Elvira, C.; San Roman, J. J.
Membr. Sci. 2001, 192, 187-191.
(47) Jagur-Grodzinski, J. J. React. Funct. Polym. 1979, 39,
99-138.
(32) Elvira, C.; San Roman, J. J. Mater. Sci.: Mater. Med. 1997,
8, 743-746.
MA048051U