Phosphorus-containing renewable polyester-polyols via ADMET polymerization: Synthesis, functionalization, and radical crosslinking

Article abstract of DOI:10.1002/pola.23887

An α,ω-diene containing hydroxyl groups was prepared from plant oil-derived platform chemicals. The acyclic diene metathesis copolymerization (ADMET) of this monomer with a phosphorus-containing α,ω-diene (DOPO II), also plant oil derived, afforded a series of phosphorus containing linear polyesters, which have been fully characterized. The backbone hydroxyls of these polyesters have been acrylated and radically polymerized to produce crosslinked polymers. The thermomechanical and mechanical properties, the thermal stability, and the flame retardancy of these phosphorus-based thermosets have been studied. Moreover, methyl 10-undecenoate has been used as chain stopper in selected ADMET polymerizations to study the effect of the prepolymers' molecular weights on the different properties of the final materials.

Full text of DOI:10.1002/pola.23887

Phosphorus-Containing Renewable Polyester-Polyols via ADMET
Polymerization: Synthesis, Functionalization, and Radical Crosslinking
1
1
LUCAS MONTERO DE ESPINOSA,1 MICHAEL A. R. MEIER,2 JUAN C. RONDA,1 MARINA GALIA, VIRGINIA CADIZ
`
´
1Department of Analytical and Organic Chemistry, Rovira i Virgili University, Campus Sescelades, Marcel.lı´ Domingo s/n,
43007 Tarragona, Spain
2Laboratory of Sustainable Organic Synthesis, Institute of Chemistry, University of Potsdam, Karl-Liebknecht-Str. 24-25,
14476 Golm/Potsdam, Germany
Received 30 October 2009; accepted 14 December 2009
DOI: 10.1002/pola.23887
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: An a,x-diene containing hydroxyl groups was pre-
pared from plant oil-derived platform chemicals. The acyclic
diene metathesis copolymerization (ADMET) of this monomer
with a phosphorus-containing a,x-diene (DOPO II), also plant oil
derived, afforded a series of phosphorus containing linear poly-
esters, which have been fully characterized. The backbone
hydroxyls of these polyesters have been acrylated and radically
polymerized to produce crosslinked polymers. The thermome-
chanical and mechanical properties, the thermal stability, and
the flame retardancy of these phosphorus-based thermosets
have been studied. Moreover, methyl 10-undecenoate has been
used as chain stopper in selected ADMET polymerizations to
study the effect of the prepolymers’ molecular weights on the
C
V
different properties of the final materials.
2010 Wiley Periodi-
cals, Inc. J Polym Sci Part A: Polym Chem 48: 1649–1660, 2010
KEYWORDS: ADMET; crosslinked polymer; crosslinking; flame
retardance; metathesis; polyesters; renewable resources
INTRODUCTION Recently, the use of plant oils as renewable
feedstock for the development of designed polymeric materi-
als has received particular attention due to environmental
concerns.1 The main components of vegetable oils are tri-
glycerides, consisting of glycerol and fatty acids. The chemi-
cal modification of their structure enables the synthesis of a
wide variety of monomers for the development of polymers
with specific properties,2 which are now being used in an
increasing number of industrial applications.
terial: starting from its synthesis, via fabrication, use, and
recycling to its final disposal. Therefore, the search for new
environmentally friendly flame-retardant polymeric materials
is of large current interest. Phosphorus-based polymers, for
instance, are an effective and well-established class of flame-
retardant materials.6 They have a good flame-retardant per-
formance and are preferred to the widely applied halogenated
flame retardants due to environmental and health reasons.7
Acyclic diene metathesis (ADMET) polymerization8 has pro-
ven to be a useful tool for the synthesis of polymers bearing
a wide variety of functional groups.9,10 The ADMET polymer-
ization of a,x-dienes affords strictly linear, unsaturated poly-
mers through a step-growth polycondensation, which is
driven by the release of ethylene. In a previous study,11 we
synthesized a series of phosphorus-containing linear polyest-
ers through ADMET copolymerization of a phosphorus-based
a,x-diene with different amounts of a castor oil-derived
diene. These polymers showed good flame retardancy and
potential application as flame-retardant coatings. However,
the low Tg of these polymers can be a limiting factor for
some applications. Taking into account the high functional
group tolerance of the so-called 2nd generation Grubbs me-
tathesis catalysts,12,13 ADMET polymerization enables the
introduction of functional groups that can act as crosslinking
points for the development of thermosets with improved me-
chanical properties.14 This work, thus, deals with the
Synthetic polymer materials are used in many areas and thus
the fire hazards associated with the use of these materials are
of great concern for both consumers and manufacturers.3
Because of the composition of triglycerides, plant oil-based
polymers are flammable, just like many other currently used
polymeric materials. The flammability of these materials is a
shortcoming in some applications. Therefore, the use of flame
retardants to reduce the combustibility of polymers is an im-
portant part of the development of plant oil-based polymeric
materials. In this way, the synthesis of flame-retardant poly-
mers from bromoacrylated plant oil triglycerides was
reported.4 However, it is known that bromine-containing
flame-retardant resins release hydrogen bromide during com-
bustion, which is toxic and corrosive.5 The concept of sustain-
able development requires fire-retardant technologies to be
developed that have a minimum impact on health and the
environment throughout the life cycle of the fire-resistant ma-
Correspondence to: M. A. R. Meier (E-mail: michael.meier@uni-potsdam.de) or J. C. Ronda (E-mail: juancarlos.ronda@urv.cat)
C
V
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 48, 1649–1660 (2010) 2010 Wiley Periodicals, Inc.
PHOSPHORUS-CONTAINING RENEWABLE POLYESTER-POLYOLS, MONTERO DE ESPINOSA ET AL.
1649

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
synthesis of plant oil-based linear polyesters containing
phosphorus and alcohol functionalities via ADMET. Further
crosslinking of these polymers has been achieved through
acrylation of the hydroxyl groups and subsequent radical
polymerization affording flame-retardant resins. The thermal
and flame-retardant properties of the obtained materials are
reported within this contribution.
HCl (5%), and once with 300 mL of brine. The organic layer
was separated and dried over MgSO4 and the solvent was
removed under reduced pressure. The reaction mixture was
subjected to column chromatography (hexane/ethyl acetate
6/1) obtaining 4.53 g of a mixture of 1,3- and 1,2-di-10-
undecenoylglycerol mixture (M1, 55% yield). Monoundece-
noylglycerol (2.38 g) and triundecenoylglycerol (5.60 g)
were obtained as by-products.
EXPERIMENTAL
FTIR (cmꢁ1): 3475 (OAH), 3077 (¼¼CAH), 1735 (C¼¼O),
1638 (C¼¼C), 1163 (CAO).
Materials
All chemicals were used as received unless otherwise speci-
fied. 10-undecenoic acid, 1,3-dichloro-2-propanol, ethyl vinyl
ether, and dicumyl peroxide were purchased from Aldrich.
Potassium carbonate and tetrabutylammonium hydrogen sul-
fate were purchased from Fluka, anhydrous magnesium sul-
fate and hydrochloric acid were purchased from Scharlab.
Benzylidene-bis(tricyclohexylphosphine)dichlororuthenium (C1,
Grubbs catalyst 1st generation), benzylidene[1,3-bis(2,4,6-tri-
methylphenyl)-2-imidazolidinylidene]dichloro(tricyclohexyl-
phosphine)ruthenium (C2, Grubbs catalyst 2nd generation),
and [1,3-Bis-(2,4,6-trimethylphenyl)-2-imidazolidinylidene]-
dichloro(o-isopropoxyphenylmethylene)ruthenium (C3, Hov-
eyda-Grubbs catalyst 2nd generation) were purchased from
Aldrich. Hexane, ethyl acetate, and methanol were pur-
1H NMR (CDCl3, TMS, d in ppm): 5.83–5.73 (m, CH2¼¼CH),
5.09–5.04 (m, CHAO of 1,2 isomer), 5.00 (d, CH2¼¼CH, Jtrans
¼
17.2 Hz), 4.89 (d, CH2¼¼CH, Jcis ¼ 10.4 Hz), 4.30 (dd, J ¼
12.00, 4.40 Hz, CH2ACO of 1,2 isomer), 4.24–4.03 (m,
OACH2ACHOHACH2AO of 1,3 isomer and CH2ACO of 1,2 iso-
mer), 3.70 (d, J ¼ 5.20 Hz, CH2AOH of 1,2 isomer), 2.65
(broad, OH), 2.33 (t, J ¼ 7.60 Hz, CH2ACOOCH2), 2.30 (t, J ¼
7.60 Hz, CH2ACOOCH), 2.01 (dt, J ¼ 7.07 Hz, CH2ACH¼¼CH2),
1.64–1.56 (m, CH2ACH2CO), 1.39–1.22 (m, CH2).
13C NMR (CDCl3, TMS, d in ppm): 174.11 (COOR of 1,3 iso-
mer), 173.98 (COOR of 1,2 isomer), 173.63 (COOR of 1,2 iso-
mer), 139.32 (CH2¼¼CH), 114.34 (CH2¼¼CH), 72.23 (CHAO of
1,2 isomer), 68.43 (CHAOH of 1,3 isomer), 65.18 (CH2AO of
1,3 isomer), 62.23 (CH2ACO of 1,2 isomer), 61.60 (CH2AOH
of 1,2 isomer), 34.43 (CH2ACO), 34.25 (CH2ACH¼¼CH2), 33.95
(CH2ACO), 29.46 (CH2), 29.44 (CH2), 29.37 (CH2), 29.35
(CH2), 29.25 (CH2), 29.22 (CH2), 29.21 (CH2), 29.03 (CH2),
29.07 (CH2), 25.08 (CH2ACH2CO), 25.03 (CH2ACH2CO).
R
chased from Scharlab. Tetrahydrofuran CHROMASOLVV Plus
(HPLC) was purchased from Aldrich. Triethylamine
(Aldrich) was dried by distillation over CaH2 and acryloyl
chloride (Aldrich) was distilled under vacuum before use.
Dimethylformamide and dichloromethane (Scharlab) were
dried over P2O5 and distilled immediately before use. Tolu-
ene was distilled from sodium/benzophenone. Thin layer
chromatography (TLC) was performed on silica gel TLC-
cards (60 F254, Merck). Compounds were visualized by
spraying with sulfuric acid/anisaldehyde ethanol solution
and heating at 200 ꢀC. For column chromatography, silica
gel 60 A.C.C. 40–63 lm (SDS) was used. Methyl 10-undece-
noate was prepared by esterification of 10-undecenoic acid
with methanol according to standard procedures. 10-[20,50-
Bis(10-undecenoyloxy)phenyl]-9,10-dihydro-9-oxa-10-phospha-
phenanthrene-10-oxide (DOPO II) was synthesized according
to a previously published procedure.15
ADMET Polymerization of M1 to Give P1
M1 (3g, 7.07 mmol) and Hoveyda-Grubbs 2nd generation cat-
alyst (22.1 mg, 0.035 mmol) were placed in a dry 10-mL
round-bottom flask under nitrogen atmosphere. The mixture
was stirred magnetically at 80 ꢀC under a constant flow of
nitrogen. After 12 h, the residue was dissolved in THF and
the metathesis reaction was stopped by adding ethyl vinyl
ether (500-fold excess to the catalyst) and stirring for 30
min at room temperature. P1 was precipitated from metha-
nol as a light brown sticky solid with 96% yield (2.86 g).
FTIR (cmꢁ1): 3540 (OAH), 1742 (C¼¼O), 1142 (CAO).
Synthesis of 1,3- and 1,2-Di-10-undecenoylglycerol
Mixture (M1)
1H NMR (CDCl3, TMS, d in ppm): 5.42–5.30 (m, CH¼¼CH),
5.10–5.05 (m, CHAO of 1,2 isomer), 4.31 (dd, J ¼ 11.60,
4.40 Hz, CH2ACO of 1,2 isomer), 4.24–4.04 (m,
OACH2ACHOHACH2AO of 1,3 isomer and CH2ACO of 1,2
isomer), 3.74–3.69 (m, CH2AOH of 1,2 isomer), 2.65 (broad,
OH), 2.35–2.29 (m, CH2ACO), 2.04–1.90 (CH2ACH¼¼CH),
1.70–1.56 (m, CH2ACH2CO), 1.38–1.21 (m, CH2).
10-Undecenoic acid (7.14 g, 38.7 mmol), potassium carbon-
ate (5.36 g, 38.7 mmol), and tetrabutylammonium hydrogen
sulfate (0.26 g, 0.77 mmol) were mixed in a dry 250-mL
two-necked flask under aꢀrgon. The potassium carbonate was
grinded and kept at 100 C for 24 h prior to use. Anhydrous
dimethylformamide (60 mL) and anhydrous toluene (60 mL)
were added. The mixture was heated to reflux and stirred
for 20 min. 1,3-Dichloro-2-propanol (1.85 mL, 19.4 mmol)
was then added with vigorous stirring and the reaction was
monitored with TLC (hexane/ethyl acetate 5/1). After com-
pletion of the reaction (approximately 4 h), the mixture was
allowed to cool down under a constant flow of argon. The
reaction mixture was diluted with 200 mL of toluene,
washed twice with 300 mL of water, twice with 300 mL of
13C NMR (CDCl3, TMS, d in ppm): 174.14 (COOR), 173.65
(COOR), 131.00–129.50 (CH¼¼CH), 72.23 (CHAO of 1,2 iso-
mer), 68.40 (CHAOH of 1,3 isomer), 65.19 (CH2AO of 1,3 iso-
mer), 62.24 (CH2ACO of 1,2 isomer), 61.57 (CH2AOH of 1,2
isomer), 34.44 (CH2ACO), 34.26 (CH2ACO), 32.76
(CH2ACH¼¼CH, trans), 29.77 (CH2), 29.62 (CH2), 29.41 (CH2),
29.28 (CH2), 29.14 (CH2), 28.91 (CH2), 28.76 (CH2), 27.35
(CH2ACH¼¼CH, cis), 25.04 (CH2ACH2CO), 24.90 (CH2ACH2CO).
1650
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 1 GPC, 1H NMR, and Thermal Characterization of the
(CH2ACO), 33.17 (C20), 32.59 (CH2ACH¼¼CH, trans), 29.72
(CH2), 29.59 (CH2), 29.47 (CH2), 29.32 (CH2), 29.25 (CH2),
29.12 (CH2), 28.95 (CH2), 28.79 (CH2), 27.19 (CH2ACH¼¼CH,
cis), 24.86 (CH2ACH2CO, C24), 24.03 (C21). 31P NMR (CDCl3,
162 MHz, d in ppm): 18.15.
Phosphorus-Containing ADMET Polyesters
Polymer (M1/M2)a
% Pb
Mn
PDIc
Tg (ꢀC)d/Tm (ꢀC)d
c
P1 (10.0/0.0)
P2 (7.5/2.5)
P3 (5.0/5.0)
P4 (2.5/7.5)
P5 (2.5/7.5)
P6 (2.5/7.5)
0.0
1.6
2.9
3.9
3.9
3.9
5,300
3,700
5,200
7,300
4,400
3,900
c
2.18
2.20
2.56
3.13
2.32
2.04
ꢁ25.8/41.3
ꢁ27.2/21.7
ꢁ12.4
Synthesis of Acrylated Polymers AP1–AP6
The acrylation of ADMET polymers P1–P6 was carried out
following a standard procedure: an anhydrous dichlorome-
thane solution of an ADMET polymer (8 mL of DCM per
gram of polymer) was placed in a round-bottom flask under
5.9
ꢁ16.6
ꢁ19.1
ꢀ
argon. The solution was cooled to 0 C and acryloyl chloride
a
Mol/mol ratio.
GPC data.
DSC data.
b
d
(1.5 mol-fold excess to hydroxyl groups), followed by trie-
thylamine (3 mol-fold excess to hydroxyl groups) were
added. The reaction mixture was allowed to reach room tem-
perature and vigorous stirring was maintained for 2 h. The
residue was added dropwise to stirring methanol, and the
pure acrylated polymers (AP1–AP6) were obtained in yields
Weight/weight percentages.
ADMET Copolymerization of M1 and DOPO II (M2)
M1 and M2 were mixed (3 g scale) in the desired molar ra-
tio (see Table 1) in a dry 10-mL round-bottom flask under
nitrogen atmosphere. If required, the respective amount
of end-capper (methyl 10-undecenoate) was added. Grubbs
2nd generation catalyst (0.5% mol related to dienes) was
added, and the mixture was stirred magnetically at 70 ꢀC
under a constant flow of nitrogen. After 12 h, the residue
was dissolved in 10 mL of THF, and the metathesis reaction
was stopped by adding ethyl vinyl ether (500-fold excess to
the catalyst) and stirring for 30 min at room temperature.
P2–P6 were precipitated from methanol with yields >95%.
The spectroscopic data is essentially the same for all
polymers.
between 60 and 96% as
precipitate.
a light brown sticky solid
Spectroscopic Data for AP1
FTIR (cmꢁ1): 1740 (C¼¼O, COOR), 1637 (C¼¼C, acrylate),
1173 (CAO), 808 (C¼¼CAH).
1H NMR (CDCl3, TMS, d in ppm): 6.43 (dd, J ¼ 17.2, 1.2 Hz,
COCH¼¼CH2), 6.41 (dd, J ¼ 17.2, 1.2 Hz, COCH¼¼CH02), 6.12 (dd,
J ¼ 17.2, 10.0 Hz, COCH¼¼CH2), 6.11 (dd, J ¼ 17.2, 10.0 Hz,
COCH0¼¼CH2), 5.88 (dd, J ¼ 10.0, 1.2 Hz, COCH¼¼CH2), 5.87 (dd,
J ¼ 10.0, 1.2 Hz, COCH¼¼CH02), 5.45–5.27 (m, CH¼¼CH and
CHAO of 1,2 isomer), 4.44–4.09 (m, OACH2ACHACH2AO of
1,3 isomer, CH2ACO of 1,2 isomer and CH2AO of 1,2 isomer),
2.34–2.28 (m, CH2ACO), 2.04–1.90 (m, CH2ACH¼¼CH), 1.72–
1.59 (m, CH2ACH2CO), 1.44–1.20 (CH2).
FTIR (cmꢁ1): 3450 (OAH), 1764 (C¼¼O, Ar COOR), 1735
(C¼¼O, COOR), 1607, 1595, 1582 and 1560 (Ar CAC), 1165
(CAO), 1116 (P¼¼O), 925 (PAO), 780, and 757 (Ar CAH).
1H NMR (CDCl3, TMS, d in ppm, number assignations related
to Scheme 2): 8.07–7.94 (m, H8,14,6), 7.69 (t, J ¼ 7.4 Hz, H9),
7.57 (dd, J ¼ 15.2, 7.6 Hz, H11), 7.45–7.36 (m, H10,16,4),
7.30–7.23 (m, H15,17), 7.14 (dd, J ¼ 8.6, 6.6 Hz, H3), 5.43–
5.31 (m, CH¼¼CH), 5.10–5.06 (m, CHAO of 1,2 isomer), 4.31
(dd, J ¼ 11.60, 4.40 Hz, CH2ACO of 1,2 isomer), 4.23–4.03
(m, OACH2ACHOHACH2AO of 1,3 isomer and CH2ACO of
1,2 isomer), 3.71 (d, J ¼ 4.4 Hz, CH2AOH of 1,2 isomer),
2.57 (t, J ¼ 7.6 Hz, H23), 2.35–2.29 (m, CH2ACO), 2.04-1.90
(CH2ACH¼¼CH), 1.78–1.56 (m, H24,20 and CH2ACH2CO),
1.44–0.92 (m, H21 and CH2).
13C NMR (CDCl3, TMS, d in ppm): 173.50 (COOR), 165.31
(COOR acrylate), 132.06 (CH¼¼CH2 acrylate), 130.51
(CH¼¼CH), 128.00 (CH¼¼CH2 acrylate), 69.47 (CHAOCOCH
¼¼CH2), 68.95 (CHAOCOR), 62.59 (CH2AOCOCH¼¼CH2), 62.22
(CH2AOCOR), 34.19 (CH2ACO), 32.79 (CH2ACH¼¼CH, trans),
29.82–28.98 (CH2), 27.38 (CH2ACH¼¼CH, cis), 24.22
(CH2ACH2CO).
Spectroscopic Data for AP2–AP6
FTIR (cmꢁ1): 3066 (Ar CAH), 1764 (C¼¼O, Ar COOR), 1736
(C¼¼O, COOR), 1637 (C¼¼C, acrylate), 1607, 1595, 1582, and
1560 (Ar CAC), 1180 (CAO), 1118 (P¼¼O), 922 (PAO), 808
(C¼¼CAH), 780 and 757 (Ar CAH).
13C NMR (CDCl3, TMS, d in ppm, number assignations related
to Scheme 2): 173.91 (COOR of 1,3 isomer), 173.76 (COOR
of 1,2 isomer), 173.46 (COOR of 1,2 isomer), 171.94 (C19),
170.85 (C22), 149.71 (C5), 149.22 (d, 9.15 Hz, C2), 147.82 (d,
16.80 Hz, C18), 135.09 (d, 5.33 Hz, C12), 133.28 (C9), 130.90
(C11), 130.73 (C16), 130.32 (CH¼¼CH), 128.61 (d, J ¼ 15.29
Hz, C10), 128.31 (C4), 128.01 (d, J ¼ 8.35 Hz, C6), 124.30 (d,
J ¼ 134.30 Hz, C7), 124.80–124.61 (C3, C14, C15), 123.21 (d, J
¼ 9.87 Hz, C8), 122.48 (d, J ¼ 145.26 Hz, C1), 121.02 (d, J ¼
11.47 Hz, C13), 120.66 (d, J ¼ 6.84 Hz, C17), 72.04 (CHAO of
1,2 isomer), 67.91 (CHAOH of 1,3 isomer), 64.98 (CH2AO of
1,3 isomer), 62.28 (CH2ACO of 1,2 isomer), 61.16 (CH2AOH
of 1,2 isomer), 34.26 (CH2ACO), 34.21 (C23), 34.08
1H NMR (CDCl3, TMS, d in ppm, number assignations related
to Scheme 2): 8.07–7.95 (m, H8,14,6), 7.69 (t, J ¼ 7.8 Hz, H9),
7.60–7.55 (m, H11), 7.45–7.36 (m, H10,16,4), 7.30–7.23 (m,
H
15,17), 7.15 (dd, J ¼ 8.6, 6.6 Hz, H3), 6.43 (dd, J ¼ 17.2, 1.2
Hz, COCH ¼ CH2), 6.41 (dd, J ¼ 17.2, 1.2 Hz, COCH ¼ CH02),
6.12 (dd, J ¼ 17.2, 10.0 Hz, COCH¼¼CH2), 6.11 (dd, J ¼ 17.2,
10.0 Hz, COCH0¼¼CH2), 5.88 (dd,
J
¼
10.0, 1.2 Hz,
COCH¼¼CH2), 5.87 (dd, J ¼ 10.0, 1.2 Hz, COCH¼¼CH02), 5.45–
5.27 (m, CH¼¼CH and CHAO of 1,2 isomer), 4.42–4.10 (m,
OACH2ACHACH2AO of 1,3 isomer, CH2ACO of 1,2 isomer
and CH2AO of 1,2 isomer), 2.57 (t, J ¼ 7.6 Hz, H23), 2.32–
PHOSPHORUS-CONTAINING RENEWABLE POLYESTER-POLYOLS, MONTERO DE ESPINOSA ET AL.
1651

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
2.28 (m, CH2ACO), 2.04–1.90 (CH2ACH¼¼CH), 1.78–1.54 (m,
H24,20 and CH2ACH2CO), 1.44–0.92 (m, H21 and CH2).
poly(methyl methacrylate) standards (Polymer Standards
Service PPS, Germany, Mp 102–981.000 Da) were used for
calibration. The IR analyses were performed on a FTIR-
680PLUS spectrophotometer with a resolution of 4 cmꢁ1 in
the transmittance mode. An attenuated-total-reflection acces-
sory with thermal control and a diamond crystal was used
to determine FTIR/ATR spectra. Calorimetric studies were
carried out on a Mettler DSC822 differential scanning calo-
rimeter using N2 as a purge gas (20 mL/min). Dynamic me-
chanical thermal analysis (DMTA) and tensile tests were per-
formed using a TA DMA 2928 in the controlled force-Tension
Film mode with a preload force of 0.1 N, an amplitude of 10
13C NMR (CDCl3, TMS, d in ppm, number assignations related
to Scheme 2): 173.47 (COOR), 172.08 (C19), 171.03 (C22),
165.29 (COOR acrylate), 149.85 (C5), 149.45 (d, 9.20 Hz, C2),
147.97 (d, 16.80 Hz, C18), 135.24 (C12), 133.32 (), 132.02
(CH¼¼CH2 acrylate), 131.03 (C11), 130.83 (C16), 130.48
(CH¼¼CH), 128.72 (d, J ¼ 14.48 Hz, C10), 128.37 (C4), 128.25 (d,
J ¼ 7.64 Hz, C6), 127.96 (CH¼¼CH2 acrylate), 124.72 (d, J ¼
135.00 Hz, C7), 124.92–124.83 (C3, C14, C15), 123.31 (d, J ¼
9.86 Hz, C8), 122.82 (d, J ¼ 144.86 Hz, C1), 121.23 (d, J ¼
11.47 Hz, C13), 120.88 (d,
J
¼
6.04 Hz, C17), 69.45
ꢀ
lm, and at a fixed frequency of 1 Hz in the ꢁ100 to 200 C
(CHAOCOCH¼¼CH2), 68.91 (CHAOCOR), 62.57 (CH2AOCOCH
¼¼CH2), 62.19 (CH2AOCOR), 34.39 (CH2ACO, C23), 34.18
(CH2ACO), 33.34 (CH2ACO, C20), 32.76 (CH2ACH¼¼CH, trans),
29.78–28.97 (CH2), 27.37 (CH2ACH¼¼CH, cis), 24.99
(CH2ACH2CO, C24), 24.19 (CH2ACH2CO, C21).
ꢀ
range and at a heating rate of 3 C/min. Rectangular samples
with dimensions 10 ꢂ 5 ꢂ 0.5 mm3 were used. The tensile
assays were performed by triplicate on rectangular samples
(10 ꢂ 5 ꢂ 0.5 mm3) measuring the strain while applying a
ꢀ
ramp of 0.5 N/min at 30 C. A preload force of 0.05 N and a
Curing Reactions and Extraction of Soluble Parts
soak time of 3 min were used. Thermal stability studies
were carried out on a Mettler TGA/SDTA851e/LF/1100 with
N2 as purge gas. The studies were performed in the 30–800
A dichloromethane solution (0.3 g/mL) of each acrylated poly-
ester (APs) and dicumyl peroxide (2 mol % related to acrylate
groups) was cast on a glass plate of 7.5 ꢂ 2.5 cm2. The samples
were heated to 40 ꢀC for 2 h to remove the solvent and then
the temperature was raised to 150 ꢀC at 1 ꢀC/min and main-
tained for 12 h. All samples were subjected to soxhlet extraction
with previously distilled dichloromethane to determine their
soluble fractions. Each sample (0.5 g) was extracted with 125
mL of dichlorometane. Previously to extractions, the samples
were grinded to maximize the extraction efficiency.
ꢀ
ꢀC temperature range at a scan rate of 10 C/min. The limit-
ing oxygen index (LOI) is the minimum concentration of oxy-
gen determined in a flowing mixture of oxygen and nitrogen
that will just support the flaming combustion of materials.
LOI values were measured in vertical tests on a Stanton Red-
croft instrument provided with an oxygen analyzer. The
dimensions of the polymer films were 100 ꢂ 5 ꢂ 0.5 mm3.
RESULTS AND DISCUSSION
Instrumentation
1H NMR 400 MHz and 13C NMR 100.6 MHz NMR spectra
were obtained using a Varian Gemini 400 spectrometer with
Fourier transform. CDCl3 was used as solvent and TMS as in-
ternal reference. Molecular weights were determined on a
Shimadzu gel permeation chromatography (GPC) system
equipped with a LC-20AD pump, RID-10A refractive index
detector, SIL-20A autosampler, and a CTO-20A column oven
set to 50 ꢀC. A PLgel 5 lm Mixed-D column from Polymer-
labs in THF at a flow rate of 1 mL/min was used. Linear
The versatility of ADMET polymerization enables the synthe-
sis of linear polymers with functional groups that can be
used as crosslinking points.14,16 Taking into account the
good flame retardant properties obtained in our previous
work with DOPO II (M2) as phosphorus containing comono-
mer,11 we decided to use it for the preparation of flame-re-
tardant thermosets. To introduce crosslinking points into the
polymer backbone, a hydroxyl-containing a,x-diene (M1,
Scheme 1) was synthesized. M1 and M2 were copolymerized
SCHEME 1 Synthesis of 1,3- and
1,2-di-10-undecenoylglycerol mix-
ture (M1) and ADMET polymer-
ization in presence of Hoveyda-
Grubbs 2nd generation catalyst
(C3).
1652
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 1 1H NMR spectra of (a) M1 (mixture of isomers), (b) P1, and (c) P3.
in different molar ratios via ADMET to give a series of linear
polyesters with different phosphorus contents. The hydroxyl
groups of these polyesters were then esterified with acryloyl
chloride to introduce polymerizable groups, and finally, the
resulting acrylated polyesters were crosslinked via radical
group10 proved the viability of ADMET polymerization of
alcohol-containing dienes using Grubbs 1st generation
catalyst.
We first tested different metathesis catalysts in the polymer-
ization of M1. The ADMET polymerizations were run in bulk
while a continuous flow of nitrogen was passed through the
reaction mixture to remove the ethylene, which is released
during the metathesis reaction. When 1% mol Grubbs 1st
generation catalyst (C1) was used, oligomerization and poor
conversion (by GPC) at 80 ꢀC after 24 h occurred, probably
due to catalyst degradation in presence of the primary alco-
hol of 1,2-diundecenoylglycerol.19 On the other hand, when
1% mol Grubbs (C2) and Hoveyda-Grubbs (C3) 2nd genera-
tion catalysts were used at 80 ꢀC, THF-insoluble products
were obtained. Moreover, the polymerization products were
not soluble in common organic solvents, suggesting that
some kind of crosslinking reaction might have taken place.
This poor solubility of hydroxyl-functionalized ADMET poly-
mers was also observed by Valenti et al., and the only pro-
posed reason for this behavior was the high molecular
weight of the investigated polymers.10 However, when the
catalyst load was loweredꢀ to 0.5% mol, THF soluble poly-
mers were obtained at 80 C with C1 and C2. Similar results
were found in the bulk copolymerization of M1 and M2 at
80 ꢀC in 1:1 molar ratio. C1 gave oligomerization, while C2
and C3 produced insoluble polymers with 1% mol catalyst.
When the polymerizations were conducted at 70 ꢀC with
0.5% mol of C2 and C3, soluble polymers were obtained
with monomer conversions over 98% (by GPC).
polymerization to obtain
thermosets.
a family of flame-retardant
For the synthesis of M1, 1,3-dichloro-2-propanol was reacted
with two equivalents of 10-undecenoic acid in presence of
potassium carbonate and tetrabutylammonium hydrogen sul-
fate (TBAH) as phase-transfer catalyst (Scheme 1). It must
be pointed out that 1,3-dichloro-2-propanol can be obtained
directly from glycerol17 and 10-undecenoic acid is obtained
from castor oil pyrolysis, thus, both reagents can be plant oil
derived and M1 can be considered as 100% renewable. The
1H NMR analysis of the reaction mixture revealed the pres-
ence of monoundecenoylglycerol, 1,3- and 1,2-diundecenoyl-
glycerols and triundecenoylglycerol as byproducts. In the
reaction conditions, glycidyl undecenoate is formed as inter-
mediate (detected by 1H NMR), and eventually, the epoxide
is opened by another undecenoic acid molecule, leading to a
mixture of both diacylglycerols. The crude reaction mixture
was subjected to column chromatography and the resulting
mixture of 1,3- and 1,2-di-10-undecenoylglycerol (60:40, as
1
determined by H NMR, see Fig. 1) was used in the ADMET
polymerizations.
It is known that ADMET polymerizations can be carried out
with heteroatom-containing dienes, if the heteroatom is not
situated close to the double bonds.18 In both M1 and M2,
the terminal olefins are nine carbon atoms spaced from the
functional groups. Moreover, previous work in Wagener’s
On the basis of the results obtained in the initial experi-
ments, we synthesized a series of polymers (P1–P4) with
different M1/M2 molar ratios (Scheme 2) with the aim of
PHOSPHORUS-CONTAINING RENEWABLE POLYESTER-POLYOLS, MONTERO DE ESPINOSA ET AL.
1653

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 2 Synthesis of phosphorus-containing polyesters via ADMET copolymerization in presence of Grubbs 2nd generation cat-
alyst (C2), with (bottom) and without (top) methyl 10-undecenoate as chain stopper.
determining the effect of the phosphorus content on the
properties of the ADMET polymers and the final materials.
We also wanted to study the effect of the molecular weight
of the ADMET polymers on the flame retardant properties of
the crosslinked materials. For this purpose, we performed
ADMET polymerizations with 10 and 20% of methyl 10-
undecenoate as chain stopper for the highest phosphorus
content (P5 and P6, Scheme 2). As reported previously, this
procedure results in an efficient end-capping and reduction
of the molecular weight.11,20 Thus, six polyesters (P1–P6)
with different M1/M2 molar ratios were synthesized (see
Table 1).
ever, this fact indicates that the real molecular weights are
probably higher than those obtained by GPC analysis.
The thermal characterization of the ADMET polyesters was
carried out with differential scanning calorimetry (DSC). The
DSC traces stemming from the second heating run (20 ꢀC/
min) of P1–P4 are shown in Figure 2, and the data are col-
lected in Table 1. The thermal analysis of P1 showed a glass
transition (Tg) at ꢁ25.8 ꢀC and a melt followed by a cold
crystallization and a second melt (Tm) at 41.3 ꢀC. However,
Figure 1 shows the 1H NMR spectra of M1 [as a mixture of
isomers; Fig. 1(a)], P1 [Fig. 1(b)], and P3 [Fig. 1(c)] as rep-
resentative examples. The ADMET polymerizations could be
confirmed by the disappearance of the terminal olefin signals
at 5.8 and 4.9 ppm together with the appearance of a mul-
tiplet at 5.3 ppm in both P1 and P3. The molecular weights
of the ADMET polyesters (P1–P6) were measured by GPC,
using PMMA standards (Table 1). The Mn of the M1 ADMET
polymer (P1) was 5300 Da, corresponding to a polymeriza-
tion degree about 12. From P2 to P4, Mn increases as the
content of comonomer M2 does. P5 and P6, with the same
composition as P4, showed lower Mn according to the
increasing percentage of chain stopper used in the copoly-
merization. The determination of the molecular weights by
integration of the chain end group signals in the 1H NMR
spectrum was not reliable for most of these polymers due to
the poor intensity of the terminal double bond signals. How-
FIGURE 2 DSC traces of the phosphorus-containing ADMET
polyesters P1–P4.
1654
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 2 TGA Results of the Phosphorus-Containing ADMET Polyesters
TGA (N2)
TGA (Air)
Tmax (ꢀC)b
% Pa
T5% loss (ꢀC)
Tmax (ꢀC)b
Char800 (%)
T5% loss (ꢀC)
Char800 (%)
ꢀ
ꢀ
Polymer
C
C
P1
P2
P3
P4
P5
P6
a
0.0
1.6
2.9
3.9
3.9
3.9
315
297
311
313
312
309
316/432
307/413/465
413/465
462
0.0
6.0
7.4
8.8
8.2
7.9
b
295
296
315
324
313
312
316/431/496
306/413/457/653
412/448/603
452/577
0.0
0.0
4.1
9.1
3.0
2.2
413/463
413/465
420/453/564
393/453/564
Weight/weight percentages.
Temperature of maximum weight loss rate.
observations with a polarizing microscope did not reveal
two different melting processes when using different heating
rates or annealing conditions. As M2 is added as comonomer,
polymer crystallization becomes more difficult due to the
bulky aromatic core of M2. Moreover, the increase in the aro-
matic content causes restrictions in the segmental mobility
and an increase in the Tg occurs. The DSC trace of P2, with
only a low content of M2, reveals a Tg of ꢁ27.2 ꢀC and just
1H NMR spectra are shown in Figure 4, where the character-
istic set of signals at 6.3, 6.1, and 5.7 ppm confirm the pres-
ence of the acrylate groups. Moreover, in the 13C NMR spec-
trum, the complete disappearance of the signals belonging to
the non acrylated polymer P3 and the appearance of the sig-
nals of AP3 confirms full functionalization.
ꢀ
one Tm of 21.7 C. Finally, P3 and P4 show only glass transi-
tions at ꢁ12.4 ꢀC and 5.9 ꢀC, respectively. As previously
mentioned, the addition of methyl 10-undecenoate as chain
stopper in the synthesis of P4 affords lower molecular
ꢀ
weights. As a result, the Tgs obtained for P5 (ꢁ16.6 C) and
ꢀ
P6 (ꢁ19.1 C) are found below that of P4.
The thermal stability of the ADMET polyesters was studied
by thermogravimetric analysis (TGA) under nitrogen and air
atmospheres (data in Table 2). Good thermal stability is
observed for P1–P6 under nitrogen [Fig. 3(a)] with 5%
weight loss around 310 ꢀC. For P1, the main degradation
ꢀ
step takes place at 430 C but as M2 is introduced, the ther-
mal stability increases and a new degradation step appears
around 460 ꢀC. Moreover, an increase in the aromatic and
phosphorus content carries an increase in the char obtained
ꢀ
at 800 C from P1 (0.0%) to P4 (8.8%). The thermal stabil-
ity and char obtained at 800 ꢀC of P5 and P6 is lowered
with respect to P4 due to their lower molecular weight. A
similar trend is observed in the thermal degradation behav-
ior under air [Fig. 3(b)]. Five percent weight loss around
310 ꢀC and two main degradation steps related with the
M1/M2 composition at 430 and 450 ꢀC were observed
under these conditions. Under air, the residues at 800 ꢀC
increase with the phosphorus content reaching a maximum
value of 9.1% for P4.
In the next step, the hydroxyl functionalized polyesters P1 to
P6 were reacted with acryloyl chloride in the presence of
triethylamine (Scheme 3). The acrylated polyesters (AP1 to
AP6) were isolated by slow addition of the reaction mixture
to methanol in yields between 60 and 96%. Nonquantitative
yields were probably due to the increased solubility of the
polyesters in methanol after acrylation of the hydroxyl
groups that led to partial fractionation. Two representative
FIGURE 3 TGA measurements (a) under nitrogen and (b) under
air of the phosphorus-containing ADMET polyesters P1–P6.
PHOSPHORUS-CONTAINING RENEWABLE POLYESTER-POLYOLS, MONTERO DE ESPINOSA ET AL.
1655

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 3 Synthesis of acrylated phosphorus-containing polyesters AP1–AP6.
Based on our previous experience in the crosslinking of acry-
late derivatives,21,22 dicumyl peroxide was chosen as radical
initiator for the crosslinking reaction of the acrylated ADMET
polyesters. DSC curing runs between APs and dicumyl perox-
ide (2% molar to acrylate groups) showed similar exotherms
with onsets about 150 ꢀC for all the acrylated polyesters.
Once the curing conditions were established, APs and
dicumyl peroxide (2% mol) were dissolved in dichlorome-
thane (0.3 g/mL) and the resulting solution was cast on a
glass plate. The samples were heated at 40 ꢀC for 2 h to
remove the solvent and then the temperature was raised to
150 ꢀC at 1 ꢀC/min and maintained for 12 h. The cured
materials (samples I to VI), were obtained as light brown
transparent films. The curing extent of the crosslinking reac-
tions was studied with FTIR spectroscopy by following the
disappearance of the acrylate group bands. Figure 5 shows
the FTIR spectra of AP3 and sample III as a representative
example of the crosslinking reaction. The bands at 1637
cmꢁ1 (C¼¼C stretching), 1403 and 1294 cmꢁ1 (¼¼CAH in-
plane deformation) and 983 and 808 cmꢁ1 (¼¼CAH out-of-
plane deformation) completely disappear confirming the po-
lymerization of the acrylate groups. The reactivity of noncon-
jugated internal double bonds toward radical polymerization
is very low,23,24 thus, the double bonds of the polyester
backbone were not expected to polymerize. This is con-
firmed in the ATR-FTIR spectrum after crosslinking by the
presence of the band at 967 cmꢁ1, which is associated to the
C¼¼CAH out-of-plane deformation. The crosslink extent was
also investigated by extracting the soluble fraction of sam-
ples I–VI. Sample I presents a soluble fraction of 0.9%,
which indicates a very high crosslinking degree. The soluble
fractions increase as the M2 content does reaching a value
of 20.3% for sample IV due to the decreasing number of
crosslink points available in the linear prepolymers. The
soluble fractions of samples V (16.8%) and VI (17.7%) are
similar to that of sample IV showing that similar crosslinking
degrees were achieved independently of the prepolymers
molecular weights.
FIGURE 4 1H NMR spectrum of AP3 and enlargement of the 75–60 ppm region of 13C NMR spectra of P3 and AP3.
1656
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
fested by a higher tan delta peak.27 This fact is confirmed by
the height of the tan delta peak, which increases from sam-
ple I to sample IV. The effect of the prepolymer molecular
weight on the dynamic mechanical properties of the cross-
linked materials is observed in samples IV, V, and VI. The Tg
value drops from sample IV to samples V and VI as a conse-
quence of the lower molecular weights of P5 and P6 with
respect to P4.
The thermal degradation behavior of the crosslinked materi-
als under nitrogen and air atmospheres is shown in Figure
7, and the data are collected in Table 3. Under nitrogen [Fig.
7(a)], all samples preseꢀnt good thermal stability with 5%
weight loss around 340 C. From the maxima of weight loss
rate, it can be inferred that two general degradation mecha-
nisms are taking place, which can be related to the prepoly-
mer composition. Sample I contains the M1 homopolymer
and degrades in one ꢀsingle step with the maximum weight
loss rate around 430 C. As M2 is addꢀed, a new maximum of
weight loss rate appears around 460 C, which becomes the
main degradation step as the M2 content increases. The char
at 800 ꢀC increases from sample I (5.0%) to sample IV
(11.2%) as the phosphorus content does and slightly
decreases for samples V and VI as a result of their lower
crosslink density that causes a lower thermal stability. The
thermal degradation behavior under air atmosphere is pre-
sented in Figure 7(b). 5% weight loss around 330 ꢀC is
observed for all samples followed by a main degradation
process and an oxidative degradation step. Sample I presents
FIGURE 5 FTIR-ATR spectra of (a) AP3 and (b) sample III. The
absorption bands associated to the acrylate group are
indicated.
The soluble fractions were analyzed by 1H and 31P NMR
spectroscopy. 1H NMR showed total reaction of the acrylate
double bonds and the presence of a triplet ꢃ2.4 ppm belong-
ing to the methylene protons adjacent to the carbonyls of
the polymerized acrylates. This clearly indicates that the
soluble fractions are composed of low molecular weight
polymers instead of non polymerized acrylates. However, the
aromatic region showed little variations suggesting some
change taking place in the DOPO moiety. This was further
confirmed by examining the 31P NMR spectra, where a new
singlet at 32.6 ppm appeared together with DOPO signal
(18.2 ppm). This new signal matches with the opened DOPO
form, having a phosphinic acid functionality (hydrolysis of
the PAO bond). Although the DOPO ring was closed through-
out the synthesis of M2, ADMET polymerizations and acryla-
tion reactions, it is known that a certain amount of hydrated
DOPO can be found when using DOPꢀO as a reagent, and
dehydration at temperatures over 200 C under vacuum are
usually necessary prior to use.25 It is, thus, possible that
part of the pendant DOPO moieties were hydrated by long
exposure of the samples to air. As the DOPO PAO bond is
cleaved in the early stages of thermal degradation (300
ꢀC),26 the flame retardant action of these materials will not
be affected in any case and thus, the presence of the opened
form of DOPO does not interfere with the aim of this study.
ꢀ
a one-step degradation mechanism around 430 C before the
oxidative degradation. As explained above, the introduction of
M2 as comonomer increases the thermal stability and a sec-
ond degradation step appears at a higher temperature (ꢃ450
ꢀC). For the M2 containing samples, the degradation rate is
ꢀ
retarded over 500 C with formation of an intermediate char.
The amount of char formed at these temperatures, which
increases proportionally to the phosphorus content, indicates
how efficiently the burning surface would protect the rest of
the polymer under real fire conditions. Moreover, the char at
ꢀ
800 C increases with the phosphorus content from sample I
to sample IV and decreases for samples V and VI for the
Figure 6 shows the dynamic thermomechanical analysis ꢀof
samples I to VI as the tan delta plots from ꢁ100 to 150 C.
The main peak of the tan delta plots, that is the alpha relaxa-
tion, is related to the glass transition temperature. As
expected, from sample I to sample IV, the maximum of the
tan delta peak shifts to higher temperatures as M2 content
increases due to the increasing aromatic fraction. On the
other hand, an increase in M2 content means more space
between acrylate groups, which is accompanied by a loss of
crosslink density, and a decreased crosslink density is mani-
FIGURE 6 Tan delta plots of the crosslinked polymers I–VI.
PHOSPHORUS-CONTAINING RENEWABLE POLYESTER-POLYOLS, MONTERO DE ESPINOSA ET AL.
1657

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
The mechanical properties of the crosslinked materials were
investigated in tensile assays. The mechanical parameters
obtained are collected in Table 4 and selected stress–strain
curves of samples I to VI are compared in Figure 8. When
examining the behavior of samples I to IV, two different fac-
tors must be taken into account. There is an increase in the
aromatic content that leads to an increase of the network ri-
gidity, but at the same time there is a decrease in the cross-
link density. Because of the combination of both factors, the
modulus decreases from sample I to sample IV as a result of
the decreasing crosslink density, but at the same time, the
tensile strength increases due to the increasing aromatic
content. The decrease of crosslink density also determines
the elongation at break, that increases from sample I to sam-
ple IV reaching a maximum value of 142% for sample IV.
However, the variation of these parameters is not completely
linear; the modulus increases for sample III, suggesting a
higher influence of the aromatic fraction on mechanical
behavior of this sample. Moreover, sample IV shows a
decrease in tensile strength due to its lower crosslink den-
sity when compared with samples I–III, which is confirmed
by the differences in soluble fractions. The effect of the pre-
polymer molecular weight on the mechanical properties is
clearly observed when comparing the stress/strain curves of
samples IV to VI. Although the aromatic content remains
constant, the tensile strength and modulus decrease. More-
over, the elongation at break is reduced twofold.
The flame retardancy of the crosslinked materials was eval-
uated using the LOI vertical test with films of thickness
between 0.4 and 0.5 mm. Table 3 contains the LOI values
obtained for samples I to VI. The phosphorus-free sample I
gave a LOI value of 18.4, a low index, which is related to its
high aliphatic content. The LOI values clearly increase with
the phosphorus content from sample II to sample IV reach-
ing a value of 25.7 for the later. The effect of the prepolymer
molecular weight at a constant phosphorus content on the
flame retardancy can be observed by comparing samples IV,
V, and VI. The LOI value drops from 25.7 in sample IV to
21.9 and 21.7 in samples V and VI, respectively. During com-
bustion, the thermal scission of the polymer backbone leads
to fragments of different sizes. In the case of low crosslink
FIGURE 7 TGA measurements (a) under nitrogen and (b) under
air of the crosslinked polymers I–VI.
reasons explained earlier. These crosslinked systems show an
increased thermal stability with onset degradation tempera-
tures 30 ꢀC above the non-crosslinked polymers, both under
nitrogen and air atmospheres. The crosslinking retards the
release of volatiles and favors char formation.
TABLE 3 Dynamic Thermomechanical Characterization, TGA Results, and LOI Values of Crosslinked Polymers I–VI
TGA (N2)
TGA (Air)
b
ꢀC
ꢀC
Tg
Soluble
Fraction (%)a
T5% loss
(ꢀC)
Tmax
(ꢀC)c
Char800
(%)
T5% loss
(ꢀC)
Tmax
(ꢀC)c
Char800
(%)
Sample
% Pa
(ꢀC)
LOI
I
0.0
1.5
2.7
3.8
3.8
3.8
35.4
47.0
47.3
52.2
40.0
47.9
0.9
4.1
343
328
346
358
344
337
434
5.0
9.8
331
324
328
337
325
329
428
0.0
0.1
2.6
7.6
4.8
6.0
18.4
21.8
24.0
25.7
21.9
21.7
II
423/462
425/463
463
418/452/564
418/449/619
449/592
III
IV
V
VI
a
8.9
10.9
11.2
10.7
10.2
20.3
16.8
17.7
425/462
425/462
419/449/609
419/452/631
c
Weight/weight percentages.
Temperatures of maximum weight loss rate.
b
Maxima of the tan d peak.
1658
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
renewable chain stopper. The crystallinity of these polyesters
decreased as the amount of DOPO-based comonomer (M2)
was increased and totally amorphous polymers were
obtained for the higher M2 contents. Extensive acrylation of
the hydroxyl groups in the polyesters backbone followed by
radical polymerization afforded thermosetting polymers with
high crosslinking degrees. These plant oil-based thermosets
show glass transition temperatures ranging from 35 to 52
ꢀC, good thermal stability, and relatively good flame retard-
ancy despite their high aliphatic (fatty acid) content.
TABLE 4 Mechanical Properties of the Crosslinked
Polymers I–VI
Sample
Modulus (MPa)
TS (MPa)a
Elongation (%)b
I
65.8
31.5
54.2
16.8
10.3
10.1
2.37
3.07
4.50
2.16
1.46
1.96
b
5
II
15
31
142
64
62
III
IV
V
VI
a
The authors express their thanks to MICINN [Ministerio de
Ciencia e Innovacio´n (MAT2008-01,412)] for the financial sup-
port for this work. M. A. R. Meier kindly acknowledges the fi-
nancial support from the German Federal Ministry of Food,
Agriculture and Consumer Protection (represented by the
Fachagentur Nachwachsende Rohstoffe; FKZ 22026905).
Tensile strength.
Elongation at break.
density, small volatile fragments are rapidly produced and
released to the flame, thus feeding it. As the prepolymer mo-
lecular weights decrease from sample IV to sample VI, the
crosslink density also does and as a result lower LOI values
are obtained. The effect of the crosslink density on the flame
retardant properties is further confirmed when comparing
the LOI values of samples II and VI. Despite its lower phos-
phorus content (1.5%), sample II gave the same LOI than
sample VI (3.8%). The molecular weights of P2 and P6 are
similar; however, the higher hydroxyl content of P2 is re-
sponsible for the higher crosslink density of sample II. As a
result, a similar flame retardant behavior is obtained for
both samples.
REFERENCES AND NOTES
1 Guner, F. S.; Yagci, Y.; Erciyes, A. T. Prog Polym Sci 2006,
¨
31, 633–670.
2 Meier, M. A. R.; Metzger, J. O.; Schubert, U. S. Chem Soc
Rev 2007, 36, 1788–1802.
3 Irvine, D. J.; McCluskey, J. A.; Robinson, I. M., Polym Degrad
Stab, 2000, 67, 383–396.
4 Eren, T.; Kusefoglu, S. H.
J Appl Polym Sci 2004, 91,
¨
2700–2710.
5 Purser, D. In Fire Retardant Materials; Horrocks, A. R.; Price,
CONCLUSIONS
D., Eds.; CRC Press: Boca Raton, FL, 2001; pp 69–127.
A plant oil-based a,x-diene containing hydroxyl groups (M1)
has been successfully polymerized via ADMET polymeriza-
tion with Hoveyda-Grubbs 2nd generation catalyst, reaching
high molecular weights. This monomer has also been copoly-
merized with an a,x-diene bearing a DOPO pendant group
using Grubbs 2nd generation catalyst. In this way, phospho-
rus-containing polyesters with molecular weights up to 7000
Da have been obtained. Moreover, while maintaining a con-
stant phosphorus content, the molecular weigh of the poly-
esters has been reduced using methyl 10-undecenoate as
6 Jain, P.; Choudary, V.; Varma, I. K. J Macromol Sci Polym
Rev 2002, 42, 139–183.
7 Lu, S. Y.; Hamerton, I. Prog Polym Sci 2002, 27, 1661–1712.
8 Schwendeman, J. E.; Church, A. C.; Wagener, K. B. Adv
Synth Catal 2002, 344, 597–613.
9 Baughman, T. W.; Wagener, K. B. Adv Polym Sci 2005, 176,
1–42.
10 Valenti, D. J.; Wagener, K. B. Macromolecules 1998, 31,
2764–2773.
11 Montero de Espinosa, L.; Ronda, J. C.; Galia`, M.; Ca´diz, V.;
Meier, M. A. R. J Poym Sci Part A: Polym Chem 2009, 47,
5760–5771.
12 Weskamp, T.; Schattenmann, W. C.; Spiegler, M.; Herrmann,
W. A. Angew Chem Int Ed 1998, 37, 2490–2493.
13 Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org Lett 1999,
1, 953–956.
14 Matloka, P. P.; Kean, Z.; Greenfield, M.; Wagener, K. B.
J Poym Sci Part A: Polym Chem 2008, 46, 3992–4011.
15 Lligadas, G.; Ronda, J. C.; Galia`, M.; Ca´diz, V. J Polym Sci
Part A: Polym Chem 2006, 44, 5630–5644.
16 Delgado, P. A.; Zuluaga, F.; Matloka, P.; Wagener, K. B.
J Polym Sci Part A: Polym Chem 2009, 47, 5180–5183.
17 Zheng, Y.; Chen, X.; Shen, Y. Chem Rev 2008, 108,
FIGURE 8 Stress–strain curves of the crosslinked polymers I–VI.
5253–5277.
PHOSPHORUS-CONTAINING RENEWABLE POLYESTER-POLYOLS, MONTERO DE ESPINOSA ET AL.
1659

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
18 Wagener, K. B.; Brzezinska, K.; Anderson, J. D.; Younkin, T.
R.; Steppe, K.; de Boer, W. Macromolecules 1997, 30,
7363–7369.
23 Valverde, M.; Andjelkovic, D.; Kundu, P. P.; Larock, R. C.
J Appl Polym Sci 2008, 107, 423–430.
24 Henna, P. H.; Andjelkovic, D. D.; Kundu, P. P.; Larock, R. C.
19 Dinger, M. B.; Mol, J. C. Organometallics 2003, 22,
J Appl Polym Sci 2007, 104, 979–985.
1089–1095.
25 Liu, Y. L.; Wei, W. L.; Chen, W. Y. Polym Int 2003, 52,
20 Rybak, A.; Meier, M. A. R. ChemSusChem 2008, 1, 542–547.
1275–1279.
21 Montero de Espinosa, L.; Ronda, J. C.; Galia`, M.; Ca´diz, V.
26 Guo, W.; Leu, W.-T.; Hsiao, S.-H.; Liou, G.-S. Polym Degrad
J Polym Sci Part A: Polym Chem 2009, 47, 1159–1167.
Stab 2006, 91, 21–30.
22 Montero de Espinosa, L.; Ronda, J. C.; Galia`, M.; Ca´diz, V.
27 Sepe, M. P. In Thermal Analysis of Polymers; Rapra Review
J Polym Sci Part A: Polym Chem 2009, 47, 4051–4063.
Reports 95; Rapra, 1995; p 28.
1660
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA