Polyester from dimethylketene and acetaldehyde: Direct copolymerization and β-lactone ring-opening polymerization

Article abstract of DOI:10.1002/pola.24854

Two ways to obtain aliphatic polyesters (PEs) from dimethylketene and acetaldehyde were investigated. On the one hand, a direct anionic copolymerization was carried out in toluene at -60 °C. The resulting polymer was mainly composed of PE units. On the other hand, a two-step process involving the synthesis of 3,3,4-trimethyl-2-oxetanone by [2+2] cycloaddition, followed by its ring-opening polymerization, with various initiators and solvents, led to the expected PE. Molecular weights up to 9000 g mol ^-1 (measured by nuclear magnetic resonance (NMR)), with narrow polydispersity around 1.2, were obtained. These polymers were found stable up to 274 °C under nitrogen and a broad and complex endothermic peak attributed to crystallinity was observed near 139 °C by differential scanning calorimetry (DSC). The crystallinity, measured by X-ray diffraction, was close to 0.45.

Full text of DOI:10.1002/pola.24854

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Polyester from Dimethylketene and Acetaldehyde: Direct
Copolymerization and b-Lactone Ring-Opening Polymerization
1
,2
1
2
1
Marc Brestaz, Nicolas Desilles, Guillaume Le, Claude Bunel
1
INSA de Rouen, Laboratoire Polym e` res Biopolym e` res Surfaces, CNRS UMR 6270 & FR 3038, avenue de l’Universit e´ ,
BP8, 76801 Saint-Etienne du Rouvray Cedex, France
2
ARKEMA, Cerdato, 27470 Serquigny, France
Correspondence to: N. Desilles (E-mail: nicolas.desilles@insa-rouen.fr)
Received 27 May 2011; accepted 22 June 2011; published online 12 July 2011
DOI: 10.1002/pola.24854
ABSTRACT: Two ways to obtain aliphatic polyesters (PEs) from
dimethylketene and acetaldehyde were investigated. On the
one hand, a direct anionic copolymerization was carried out
around 1.2, were obtained. These polymers were found sta-
ꢁ
ble up to 274 C under nitrogen and a broad and complex
endothermic peak attributed to crystallinity was observed
ꢁ
ꢁ
in toluene at ꢀ60 C. The resulting polymer was mainly com-
near 139 C by differential scanning calorimetry (DSC). The
posed of PE units. On the other hand, a two-step process
involving the synthesis of 3,3,4-trimethyl-2-oxetanone by
crystallinity, measured by X-ray diffraction, was close to 0.45.
VC 2011 Wiley Periodicals, Inc. J Polym Sci Part A: Polym
[
2þ2] cycloaddition, followed by its ring-opening polymeriza-
Chem 49: 4129–4138, 2011
tion, with various initiators and solvents, led to the expected
ꢀ
1
PE. Molecular weights up to 9000 g mol (measured by nu-
clear magnetic resonance (NMR)), with narrow polydispersity
KEYWORDS: anionic polymerization; dimethylketene; polyesters;
ring-opening polymerization; synthesis
2
6–29
30–32
INTRODUCTION Ketenes are derivatives of carboxylic acids
which contain two consecutive double bonds (C ¼¼ C ¼¼ O). First
isolated by Staudinger in 1905 with the synthesis of diphe-
tives,
benzaldehyde and derivatives,
formalde-
3
3
34
hyde, and furfural. The preferred initiators are anionic
ones, such as alkaline metals with aromatic organic com-
1
27,30,34
33,34
nylketene, ketenes have been greatly studied and several
pounds (sodium-naphtalene
ary phosphines, lithium alcoholates), and organometallic
compounds
polymer presents a strictly alternated structure between ke-
tene and the carbonyl compound to form an ester unit
depicted in Scheme 2. Among the researches undertaken
during the 1960s, only brief analyses such as infrared spec-
troscopy, LiAlH4 reduction, elementary analysis, and X-ray
), bases
(amines, terti-
2
–7
reviews have been published.
These substances, contain-
2
6–32,34
ing two double bonds in adjacent position, show a particular
high reactivity so that most of them are unstable products
that cannot be isolated and must be handled very carefully.
The chemistry of ketenes is dominated by addition reactions
with nucleophiles, electrophiles, or reagents having labile
(n-BuLi, diethylzinc ZnEt ). The obtained
2
2
,8–10
hydrogen atoms,
cycloaddition.
and they characteristically undergo
One of the most interesting properties of
1
5,26–32,35
1
1,12
diffraction were presented.
these compounds is their ability to be used as monomers.
Indeed, in the absence of ketenophiles, simple ketenes
undergo self-addition to lead to oligomers and polymers. For
example, a polyester (PE) was obtained by polymerizing eth-
Another way to reach the same PE structure lies in the ring-
opening polymerization (ROP) of a b-lactone. In a first step,
the latter can be synthesized by a [2þ2] cycloaddition
between DMK and a carbonyl compound, provided an appro-
1
3
3
6
ylketene, and several studies on the polymerization of
dimethylketene (DMK) showed that a selective reaction of
the double bonds is possible with a suitable choice of both
initiator and solvent to form three different polymer units:
priate catalyst is chosen (Scheme 3). Thus, the use of metal
halides such as AlCl , HgCl , FeCl , and, especially, ZnCl
2
3
2
3
enabled the synthesis of a wide range of substituted propio-
3
7–40
lactone and butyrolactone.
boron trifluoride diethyl etherate (BF -O(Et) ).
Another useful catalyst is the
4
,14–20
41,42
(
1) acetal, (2) ketone, and (3) ester (Scheme 1).
With
More
3
2
anionic initiators such as n-butyllithium (n-BuLi) or sodium
complex catalysts, such as zinc, cobalt, and lead salts of
difluorophosphoric acid were also efficient with other car-
bonyl compounds.
methoxide, and depending on temperature and solvent, the
2
1–25
43,44
PE structure can be obtained predominantly.
DMK can also copolymerize with carbonyl compounds,
namely ketones and aldehydes, such as acetone and deriva-
The second step is the polymerization of the obtained b-lac-
tone. Several publications dealt with the ROP of different b-
V
C
2011 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138
4129

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 3 Polyester obtained by ring opening polymerization
of a b-lactone, previously synthesized by a [2þ2] cycloaddition
between dimethylketene and a carbonyl compound.
SCHEME 1 Dimethylketene polymer units main chain struc-
tures (1: PA, 2: PK, and 3: PE).
dimethylsulfoxide (DMSO; 99.7þ% extra dry), isobutyric an-
hydride (IBAN) (97%), n-butyllithium 1.6 M in hexanes solu-
tion (diluted when needed), zinc chloride 1.0 M in diethyl
ether, sodium acetate (99þ% anhydrous), potassium acetate
4
5–55
51–63
66–69
lactones such as b-propiolactone,
b-butyrolactone,
4
7,64,65
and b-pivalolactone,
and a-disubstituted
as well as b-monosubstituted
ones, mainly with anionic initiators
such as carboxylates and alcoholates, and occasionally with
zwitterionic, cationic, or coordination-insertion initiation.
However, only few publications mentioned the polymerization
of a,a ,b-trisubstituted b-lactones initiated by tetraethylammo-
5
5,70–72
(
99þ% anhydrous), and 18-crown-6 (99%) were purchased
7
3,74
from Acros Organics and used as received. Acetaldehyde
Acros Organics; 99.5%) was dried then distilled over cal-
(
0
cium sulfate, and packaged in sealed bottles before use.
Boron trifluoride diethyl etherate, sodium formate (>99%),
potassium formate (>99%), potassium benzoate (KB; >99%),
and TEAB (>99%) were purchased from Sigma-Aldrich and
used as received. Aluminum bromide (Acros Organics; 98þ%)
was purified by distillation under reduced pressure. Alumi-
num chloride (Acros Organics; 98.5%) was purified by subli-
mation under reduced pressure.
7
5,76
nium benzoate (TEAB).
In this article, we investigated two ways to obtain aliphatic
PEs from DMK and acetaldehyde because (i) to our knowl-
edge, the literature has never mentioned the direct copoly-
merization between DMK and acetaldehyde; (ii) only few
studies mentioned the use of the DMK/acetaldehyde b-lac-
7
7–80
tone (3,3,4-trimethyl-2-oxetanone),
and no literature
was found about its synthesis by [2þ2] cycloaddition; and
Every solution made from the aforementioned initiators was
also (iii) no literature concerning the ROP of 3,3,4-trimethyl-
realized under very low impurity level nitrogen atmosphere
TM
2
-oxetanone could be encountered. Thus, a direct copolymer-
(
Alphagaz
2 from Air Liquide) in a glove box and then
ꢁ
ization initiated by n-BuLi in toluene at ꢀ60 C was carried
out on the one hand, and on the other hand, a two-step pro-
cess involving the ROP, with various initiators and solvents, of
the b-lactone previously synthesized by [2þ2] cycloaddition
between DMK and acetaldehyde was carried out. The structure,
molecular weights, and thermal properties of these polymers
were determined by Fourier transform infrared-attenuated
total reflection spectroscopy (FTIR-ATR), H and C spectros-
copies, steric exclusion chromatography, thermogravimetric
analysis, differential scanning calorimetry, and X-ray diffraction.
stored in sealed bottles. All the following steps were also
TM
carried out under Alphagaz 2 nitrogen atmosphere, to pre-
vent the reaction of DMK with water and the formation of
highly explosive peroxides in the presence of oxygen.
Dimethylketene Synthesis
DMK was synthesized by pyrolysis of IBAN according to
1
13
Scheme 4. The apparatus (Scheme 5), inspired by those
24,81–83
described in the literature,
was optimized and can be
decomposed into three different parts corresponding to the
three steps: DMK synthesis by IBAN pyrolysis, DMK purifica-
tion, and DMK reaction.
EXPERIMENTAL
Materials
IBAN (600 g, 3.793 mol) was introduced, with a flow rate
ꢀ
1
Tetrahydrofuran (THF; Acros Organics; 99.5%) was dried
over potassium hydroxide, distilled over sodium benzophe-
none, and packaged in sealed bottles on molecular sieves
around 240 g h using a dosing pump and under nitrogen
ꢁ
atmosphere, into the pyrolysis oven A brought to 625
C
under a reduced pressure of 40 mbar. In parallel, the result-
ing gaseous mixture was purified by passing through con-
(
4 Å) before use. n-Decane (Acros Organics; 99þ%) was dis-
ꢁ
tilled over sodium under reduced pressure and packaged in
sealed bottles on molecular sieves (4 Å) before use. Toluene
densers B, C, and D, respectively at 60, ꢀ24, and ꢀ30 C:
unreacted IBAN, isobutyric acid, and other by-products were
mainly condensed. Most of DMK, still gaseous at this stage,
(
Acros Organics; 98%) and diethyl ether (VWR; 99%) were
distilled over sodium and packaged in sealed bottles on mo-
lecular sieves (4 Å) before use. Absolute ethanol (99.8%),
SCHEME 2 Polyester obtained by direct copolymerization
SCHEME 4 Synthesis of dimethylketene by pyrolysis of isobu-
between dimethylketene and a carbonyl compound.
tyric anhydride.
4
130
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 5 Dimethylketene synthesis, purification, and reaction apparatus.
ꢁ
ꢀ1
then bubbled through n-decane (200 g) at ꢀ15 C in flask E
834 mmol) were added slowly so that [DMK] ¼ 3.3 mol L
and get caught by liquid nitrogen in flask F. To achieve its pu-
rification, DMK was distilled from F to the reaction reactor G,
where it was trapped by liquid nitrogen. Once the synthesis
and purification steps were over, DMK was gently warmed up
to the desired temperature and brought to atmospheric pres-
sure, giving a yellow solution (58.41 g, 834 mmol). The over-
all synthesis yield, determined by mass balance, was 22%.
and [DMK]/[acetaldehyde] ¼ 1. When the temperature was
stabilized, 3.0 mL of a solution of ZnCl in diethyl ether
2
ꢀ
1
(1.0 mol L ) were added, to obtain [DMK]/[ZnCl ] ¼ 275.
2
The reaction was exothermic; the temperature increased from
ꢁ
ꢀ15 to 10 C in only few seconds. The reactive medium was
ꢁ
kept during 12 h at ꢀ15 C and then allowed to reach room
temperature. Absolute ethanol was poured into the reactor to
react with and neutralize residual DMK and catalyst. The
obtained product was retrieved by evaporating the mixture:
Anionic Copolymerization Between Dimethylketene
and Acetaldehyde
First of all, the authors highlight the danger of this direct po-
lymerization which turned out to be very exothermic and
unpredictable. Freshly distilled DMK collected in reactor G
58.41 g, 834 mmol) was brought to the desired reaction
temperature using an acetone/carbon dioxide ice bath at
60 C. Then toluene (298 mL) and acetaldehyde (45.90 g,
92.30 g of crude liquid product were obtained (yield 97%).
3,3,4-Trimethyl-2-oxetanone Polymerization in Bulk
In a 10-mL round-bottomed flask previously dried in an
oven, 3,3,4-trimethyl-2-oxetanone (5.02 g, 44 mmol) and
TEAB (40.2 mg, 0.44 mmol) were introduced under nitrogen
atmosphere, to obtain [b-lactone]/[TEAB] ¼ 100. The mixture
(
ꢁ
ꢀ
ꢁ
1
L
042 mmol) were added slowly so that [DMK] ¼ 2.8 mol
was brought, under stirring, at 37 C during 15 days. Absolute
ꢀ
1
and [DMK]/[acetaldehyde] ¼ 0.8. When the temperature
ethanol was poured into the reactor to neutralize reactive spe-
cies. Then the mixture was precipitated in large amounts of
absolute ethanol. The obtained polymer was filtered, washed
several times with ethanol, dried under vacuum at room tem-
perature, and retrieved (3.26 g, yield 65%).
was stabilized, 16.7 mL of a solution of n-BuLi in hexanes
ꢀ
1
(
0.5 mol L ) were added dropwise very carefully, to obtain
[
DMK]/[n-BuLi] ¼ 100. The reaction was so exothermic that
ꢁ
the temperature increased from ꢀ60 to 0 C in only few sec-
ꢁ
onds. The reactive medium was kept during 12 h at ꢀ60 C
Measurements
and then, allowed to reach room temperature. Absolute etha-
nol was poured into the reactor to react with and neutralize
residual DMK and initiator. Then the mixture was precipi-
tated in large amounts of absolute ethanol. The obtained
polymer was filtered, washed several times with ethanol,
dried under vacuum at room temperature, and retrieved (8.6 g,
yield 9%).
Fourier Transform Infrared (FTIR) spectra were recorded on
a PerkinElmer Spectrum 2000 FTIR, equipped with a dia-
1
13
mond ATR (Attenuated Total Reflection) device. H and
C
NMR spectra were recorded on a Bruker AC-P 300 MHz ap-
paratus. Chemical shifts are expressed in ppm relative to
CDCl which was used as internal reference and solvent. Gas
3
chromatography was carried out on a PerkinElmer 1020 GC
Plus. Mn and Mw=Mn ratios were determined by size exclu-
sion chromatography (SEC). Samples were solubilized in
CH Cl , filtered (0.45 lm) and analyzed at room temperature
using a Varian PL-GPC50 device equipped with two mixed
packed columns (PL gel mixed type C). The mobile phase
3
,3,4-Trimethyl-2-oxetanone Synthesis
Freshly distilled DMK collected in reactor G (58.41 g,
834 mmol) was brought to the desired reaction temperature
2
2
ꢁ
using an acetone/carbon dioxide ice bath at ꢀ15 C. Then
diethyl ether (253 mL) and acetaldehyde (36.74 g,
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138
4131

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
1
.1 ppm, d), and (d ¼ 1.2 ppm, d) were attributed, respec-
3 4 5
tively, to protons H , H , and H , which corroborated the
copolyester structure (Fig. 2). Other nonattributed signals
were also observed around 1.3, 3.8, 3.9, and 5.9 ppm, which
1
9
were not consistent with the homopolymers of DMK and
were linked to the very exothermic and uncontrolled poly-
merization that could lead to undesired defects in a general
PE structure.
However, it was demonstrated for polydimethylketene homo-
1
polymers that even if their H NMR spectra are quite similar,
polyketone (PK), PE, and polyacetal (PA) structures give dis-
1
3
19
tinct chemical shifts by C NMR. In particular, the PK
structure exhibited a peak at 214 ppm, and the PE structure
exhibited a peak at 177 ppm for their carbonyl, whereas the
PA structure exhibited a peak at 99 ppm attributed to the
1
3
ethylenic carbon of the main chain. According to the
C
NMR spectrum of DMK/acetaldehyde copolyester, the peaks
at d ¼ 174, 74, 46, 21, and 15 ppm were attributed to a PE
structure with a strict alternation of DMK and acetaldehyde
to form ester units (Fig. 3). Moreover, the absence of signal
ꢂ214 ppm attested that no PK units were present in the
FIGURE 1 Dimethylketene/acetaldehyde copolymer FTIR
spectrum.
1
polymer. However, as already mentioned in the H NMR
was CH Cl , and polystyrene standards were used for cali-
2
2
spectrum, several unidentified low-amplitude signals were
also observed in particular at 99 ppm, which could be attrib-
uted to polydimethylketene acetal units.
bration. Thermogravimetric analysis (TGA) measurements
were performed on a TGA Q500 system (TA Instruments)
ꢁ
ꢀ1
under nitrogen atmosphere at a heating rate of 10 C min
.
Starting degradation temperatures were determined at 1%
weight loss. Differential scanning calorimetry (DSC) was per-
formed on a DSC Q2000 apparatus (TA Instruments), under
This polymer exhibited a molecular weight MSEC ¼ 2000 g
n
ꢀ
1
mol , with a polydispersity of 2.0, which corresponds to
MNMR ¼ 7400 g mol by using the relationship M
ꢀ
1
NMR
¼ 3.7
n
n
ꢁ
ꢀ1
nitrogen with a heating rate of 10 C min and a cooling
MSEC to be established further. This molecular weight is quite
n
ꢁ
ꢀ1
rate of 5 C min . T were measured at the midpoint, and
good when compared with the theoretical value of 11,400 g
g
ꢀ
1
Tf at the maximum of the fusion peak. X-ray diffractograms
mol , considering the ratio [DMK]/[BuLi] ¼ 100.
were performed on a Siemens D5000 diffractometer (kCo
.78897 Å) by using Diffract AT Software.
¼
The thermal properties of the obtained copolyester were
investigated by TGA and DSC in nitrogen atmosphere. This
1
RESULTS AND DISCUSSION
Direct Copolymerization Between Dimethylketene and
Acetaldehyde
DMK was prepared as described in the Experimental section.
As there is no information in the literature for the copoly-
merization between DMK and acetaldehyde, the initiator, sol-
vent, and concentrations used for the copolymerization
2
6
between DMK and acetone described by Natta were tried.
This reaction turned out to be very exothermic and so
unpredictable and dangerous that we made, for safety rea-
sons, only one attempt. Nevertheless, full characterization of
this unique polymer was carried out.
The obtained polymer was found soluble in most organic
solvents and insoluble in alcohol and water. Its FTIR-ATR
spectrum (Fig. 1) showed strong absorption bands at 1725,
ꢀ
1
1
265, 1098, and 1049 cm , which indicated a PE structure
even if these bands were slightly shifted compared with the
1
8,26
DMK/acetone copolymer.
1
9
However, as demonstrated by Bunel et al., it is hazardous
to attest the structure of a polyketene only based on its FTIR
analysis; thus, NMR was used to confirm the PE structure. In
the H NMR spectrum, the signals at (d ¼ 5.1 ppm, m), (d ¼
1
FIGURE 2 Dimethylketene/acetaldehyde copolymer
H NMR
1
spectrum.
4
132
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
1
3
FIGURE 4 3,3,4-Trimethyl-2-oxetanone FTIR spectrum.
FIGURE 3 Dimethylketene/acetaldehyde copolymer
C NMR
spectrum.
Thus, BF
3
etherate, used for the [2þ2] cycloaddition
41
ꢁ
between ketene and acetaldehyde by K u¨ ng (Table 1, Run
polymer was found stable until 200 C, before a quick and
complete degradation. Its thermal stability is, therefore, lim-
ited, which is probably due to the presence of impurities
such as acetal units and a low molecular weight.
8
4
1
), ZnCl , previously used by McConnell et al. with DMK
2
and butyraldehyde (Table 1, Run 2), and other Lewis acid
catalysts AlBr3 and AlCl3 (Table 1, Runs 3–4) were tried
with diethyl ether as solvent. For safety reasons, because of
the incontrollable exothermy observed during the copoly-
merization, the temperature advocated by K u¨ ng and
The DSC thermogram did not permit to determine Tg but
showed a broad and complex endothermic peak attributed
ꢁ
ꢀ1
to fusion at 143 C, with DHf1 ¼ 40.2 J g : the obtained
ꢁ
McConnell, ranging, respectively, between 10–15 and 20–28 C,
polymer is, therefore, semicrystalline. The cooling of this
ꢁ
was decreased to ꢀ15 C. Each reaction was exothermic but
ꢁ
polymer revealed a crystallization at 93 C, with DHc
¼
could be controlled, and the maximal temperature reached by
ꢀ
1
ꢀ
30.9 J g , and subsequent heating showed a still broad
ꢁ
the medium was limited to 10 C.
ꢁ
ꢀ1
and complex fusion peak at 139 C with DHf ¼ 31.8 J g
,
2
All catalysts gave satisfactory results in terms of reaction
yield, ranging from 69 to 97%. The best yield was obtained
in agreement with DH . This complex thermogram is prob-
c
ably due to a polymorphism within the polymer, already
1
8,20
with ZnCl
2
catalyst (97%). Hence, the following spectra cor-
observed with DMK homopolymers.
respond to Run 2. Whatever the catalyst used, FTIR-ATR
ꢀ
1
showed a major absorption band at 1815 cm , characteris-
3
,3,4-Trimethyl-2-oxetanone Synthesis
3
6
tic of the b-lactone cyclic ester (Fig. 4). However, another
The DMK/acetaldehyde b-lactone was prepared as described
in the Experimental section. No information could be found
in the literature concerning its synthesis, so different cata-
lysts, previously encountered in the literature for the synthe-
sis of other ketene-based b-lactones, were tested in diethyl
ether to obtain the best yield and purity. All the parameters
for these syntheses are gathered in Table 1.
ꢀ
1
band located at 1740 cm , more or less intense depending
on the catalyst, was also observed. To confirm the synthesis
of the b-lactone and to identify and quantify this by-product,
NMR spectroscopy was used.
1
3
In the C NMR spectrum (Fig. 5), the peaks observed at d ¼
176, 80, 53, 22, and 16 ppm were all attributed to the b-
TABLE 1 3,3,4-Trimethyl-2-oxetanone Syntheses by [212] Cycloaddition Between
Dimethylketene and Acetaldehyde, in Diethyl Ether at 215 8C
[
DMK]/
[DMK]
[DMK]/
Maximum Temperature
ꢀ1
ꢁ
Run Catalyst
[acetaldehyde] (mol L
)
[catalyst] Reached ( C)
Yield (%)
1
2
3
4
BF
3
-O(Et)
2
0.9
1
3.9
3.3
3.3
3.3
30
275
275
275
ꢀ4
10
2
69
97
77
81
ZnCl
2
AlCl
3
1
AlBr
3
1
ꢀ7
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138
4133

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 3,3,4-Trimethyl-2-Oxetanone Purity
Run
Catalyst
b-Lactone Purity (%)
1
2
3
4
BF
3
O(Et)
2
85.1
98.4
93.1
88.8
ZnCl
2
AlCl
3
AlBr
3
1
Taking into account the integrations of the different H NMR
signals corresponding to 3,3,4-trimethyl-2-oxetanone, on the
one hand, and 2,2,4,4-tetramethyl-1,3-cyclobutanedione, on
the other hand, the molar proportion of dimer and conse-
quently the purity of the b-lactone were calculated. The
results were summarized in Table 2. It appeared that the
best purity was obtained with ZnCl2 catalyst, with a very
low proportion of 2,2,4,4-tetramethyl-1,3-cyclobutanedione.
As this catalyst already afforded the best yield, the improved
conditions for the b-lactone synthesis were those retained
for Run 2.
1
3
FIGURE 5 3,3,4-Trimethyl-2-oxetanone C NMR spectrum.
Numerous tests were carried out to purify crude 3,3,4-tri-
methyl-2-oxetanone, but unfortunately, neither distillation
nor chromatography was able to separate efficiently these
two products, which have very similar chemical structure
and hence high affinity. Therefore, for the ROP, the crude
3
lactone structure, especially the characteristic C at 80 ppm
and C at 53 ppm instead of 74 and 46 ppm in the PE.
2
1
Moreover, the b-lactone structure was fully verified by
H
NMR (Fig. 6). Indeed, this spectrum was found very different
from the PE one, and the signals at (d ¼ 1.2 ppm, s), (d ¼ 1.4
ppm, s), (d ¼ 1.4 ppm, d), and (d ¼ 4.4 ppm, q) were all attrib-
uted to the b-lactone structure 3,3,4-trimethyl-2-oxetanone,
3,3,4-trimethyl-2-oxetanone synthesized according to Run 2
was used without further purification, because 2,2,4,4-tetra-
methyl-1,3-cyclobutanedione was expected to be inert.
5 4 6 3
respectively to protons H , H , H , and H . Nevertheless, this
spectrum showed an additional peak at (d ¼ 1.3 ppm, s), in
accordance with the more or less intense by-product signal
observed by FTIR in Figure 4. Actually, this signal complied
with the presence of the DMK symmetric dimer, namely
3,3,4-Trimethyl-2-oxetanone Ring-Opening
Polymerization
Various tests of ROP were conducted to determine the best
conditions (initiator, solvent) in terms of yield and molecular
weights. A first series (Table 3, Runs 5–12) corresponded to
the conditions already widely used with success in literature
for less substituted b-lactones, namely in THF (e ¼ 7.6
2
,2,4,4-tetramethyl-1,3-cyclobutanedione. Indeed, the temper-
ꢁ
ature increased to 10 C during the reaction, and ketenes
are known to be unstable products that can dimerize all the
2
ꢀ1
ꢀ1
more that the temperature increases.
F m ) and DMSO (e ¼ 46.7 F m ) as solvents, with a feed
ratio [b-lactone]/[initiator] ¼ 100 and [b-lactone] ¼ 1.5 mol
ꢀ
1
L
6
, in the presence of a crown ether ([initiator]/[18-crown-
ꢁ
52,53,55,65
] ¼ 1), at 30 C.
A second series corresponded to
some initiators mentioned in the polymerization of more
substituted b-lactones, namely KB and TEAB in THF and
ꢀ
1
ꢁ
DMSO ([b-lactone] ¼ 1.5 mol L ), and in bulk, at 37
C
6
9,72,75,76
(
Table 3, Runs 13–20).
As mentioned in the litera-
ture, the polymerizations were carried out with reaction
times up to 30 days. Nevertheless, in DMSO, the reaction
time was shortened to 10 days to avoid gelation in the reac-
tor. In addition, some tests were stopped earlier when an im-
portant increase of the viscosity was noticed (Table 3, Runs
15, 16, 18–20). In all cases, the reaction was terminated by
adding ethanol.
The obtained polymers were analyzed as previously
described. Their FTIR-ATR spectra were strictly the same as
1
13
for the direct copolymerization (Fig. 1). By H and C NMR,
only the peaks observed and attributed to the PE structure
in the direct copolymerization were obtained (e.g., Figs. 7
and 8, Run 16). Thus, the ROP afforded the expected PE.
1
FIGURE 6 3,3,4-Trimethyl-2-oxetanone H NMR spectrum.
4
134
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 3 3,3,4-Trimethyl-2-Oxetanone Ring-Opening Polymerization
Reaction
ꢁ
a
SEC
n
ꢀ1
NMR
n
ꢀ1
Run
Initiator
T ( C)
Solvent
M/I
Time (days)
Yield (%)
M
(g mol
)
Mw=Mn
M
(g mol
)
b
b
b
b
b
b
b
b
5
6
7
8
9
HCO
HCO
HCO
HCO
2
2
2
2
Na
Na
K
30
30
30
30
30
30
30
30
37
37
37
37
37
37
37
37
THF
100
100
100
100
100
100
100
100
100
100
100
100
100
50
30
10
30
10
30
10
30
10
30
30
3
35
10
25
35
52
52
60
30
–
1,700
500
1.3
1.3
1.4
1.5
1.2
1.2
1.2
1.3
–
6,300
7,000
9,300
10,700
2,600
4,100
8,900
6,300
–
DMSO
THF
2,500
2,900
700
K
DMSO
THF
CH
CH
CH
CH
KB
KB
KB
3
CO
3
CO
3
CO
3
CO
2
2
2
2
Na
Na
K
1
1
1
1
1
1
1
1
1
1
2
a
0
1
2
3
4
5
6
7
8
9
0
DMSO
THF
1,100
2,400
1,700
–
K
DMSO
Bulk
THF
–
–
–
–
DMSO
Bulk
50
65
27
30
55
60
1,300
2,500
1,400
1,000
1,800
1,500
1.1
1.2
1.3
1.3
1.2
1.1
4,800
9,000
5,300
4,100
6,600
5,500
TEAB
TEAB
TEAB
TEAB
TEAB
15
30
20
3
THF
THF
DMSO
DMSO
100
50
1
b
NMR
n
calculated from M nN MR ¼ 3:7 M nS EC
M/I ¼ [b-lactone]/[initiator].
M
.
Additional information on molecular weights was given by
the H NMR spectra of the polymers initiated by benzoates
The yield was always calculated on the amount of precipi-
tated polymer. Thus, this yield cannot be linked to the mono-
mer conversion as it does not include oligomers.
1
(
e.g., Fig. 9, Run 16). Indeed, taking into account the integra-
tions of the different signals corresponding on the one hand
to the benzoate aromatic ring (d ¼ 7.4, 7.5, and 7.9 ppm,
respectively for m-, p-, and o-H) and on the other hand to
the polymer-repeating units and assuming only one benzoate
in each polymer chain, the molecular weights of Runs 16–20
These experiments showed that the ROP was faster in DMSO
whatever the initiator. This could be attributed to a better
ability to dissociate ion pairs due to its higher polarity
(eDMSO >> eTHF). In addition, for equivalent yields (Runs 17–
18 and 19–20), increasing the initiator concentration led, as
(
Table 3) were computed. These latter results showed higher
expected, to an increase in the kinetics and, to a lower
extent, to a decrease in molecular weight. Finally, the system
molecular weights than those measured by SEC, indicating
that dichloromethane was a worse solvent for these poly-
mers than for polystyrene. The relation between these two
molecular weights was linear, with M nN MR ¼ 3:7 Mn (mean
value obtained by linear regression).
SEC
1
3
1
FIGURE 7 C NMR spectrum of 3,3,4-trimethyl-2-oxetanone
FIGURE
8
H
NMR spectrum of 3,3,4-trimethyl-2-oxetanone
ring-opening polymerization product.
ROP product.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138
4135

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 4 DSC Analysis of Polymers Obtained by 3,3,4-
Trimethyl-2-oxetanone ROP
Peak
Temperature
Standard
Deviation
DH
Standard
Deviation
ꢁ
ꢀ1
)
(
C)
(J g
First heating
Cooling
139
107
138
11
10
11
40.1
ꢀ36.6
37.9
8.4
4.4
4.3
Second
heating
Among all the DSC thermograms, either no T
g
could be
ꢁ
ꢁ
determined, or only a weak T around 54 C (stdev ¼ 6 C),
g
could hardly be measured. A broad and complex endother-
mic signal was attributed to fusion (e.g., Fig. 10, Run 19).
These data were collected in Table 4.
All these results were very close to the values obtained with
the direct copolymerization, and their complexity still reveals
a polymorphism. Generally speaking, fusion enthalpies, and,
hence, crystallinity, were higher for the PEs with greater mo-
lecular weights. For example, X-ray diffraction concerning
1
FIGURE
9 H NMR spectrum of 3,3,4-trimethyl-2-oxetanone
ROP product (Table 3, Run 16).
that offered the best compromise in terms of kinetics, molec-
ular weights, and yield was TEAB in DMSO or in bulk.
ꢁ
Run 18 revealed a main peak at 15.5 and smaller peaks at
ꢁ
1
2.7, 14.7 and 17.7 , giving a crystallinity of 0.45 (Fig. 11).
CONCLUSIONS
Polymer Properties
The thermal stability of the obtained PEs was investigated
by TGA in nitrogen atmosphere. All these polymers were
found stable until temperatures ranging from 268 to 274 C,
before they degraded rapidly and totally. Their thermal sta-
bility is therefore better than the PE obtained by direct poly-
merization, surely because these polymers, synthesized by
ROP, cannot contain thermally unstable defects like acetal
units observed in the copolymer.
For the first time, a new DMK–acetaldehyde-based PE was
synthesized, using two different synthetic routes. DMK, syn-
thesized by IBAN pyrolysis, was on the one hand copolymer-
ꢁ
ꢁ
ized with acetaldehyde in toluene at ꢀ60 C, with n-BuLi
initiator. On the other hand, DMK was reacted with acetalde-
hyde in diethyl ether, with Lewis acid catalysts such as
ZnCl , to afford 3,3,4-trimethyl-2-oxetanone in a [2þ2] cyclo-
2
addition mechanism. This b-lactone was subsequently sub-
jected to ROP initiated by various species in THF, DMSO, or
FIGURE 10 DSC thermogram of 3,3,4-trimethyl-2-oxetanone
____
ROP product (Table 3, Run 19) (
and …. second heating)
first heating, - - - - cooling,
FIGURE 11 X-ray diffractogram of 3,3,4-trimethyl-2-oxetanone
ROP product (Table 3, Run 18).
4
136
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138

WWW.POLYMERCHEMISTRY.ORG
ARTICLE
in bulk. In both cases, the obtained polymer presented a
strict alternation between DMK and acetaldehyde to form
ester repeating units even if the direct copolymerization
afforded an uncontrollable and dangerous polymerization,
leading to defects in a globally copolyester scheme. For
safety reasons, the ROP should therefore be preferred to
obtain such a PE structure involving a ketene and an alde-
hyde. This second path gave molecular weights up to 9000 g
28 Natta, G.; Mazzanti, G.; Pregaglia, G.; Binaghi, M. U.S. Pat-
ent 3,337,502, 1967.
2
9 Natta, G.; Pregaglia, G. F.; Binaghi, M. Macromol Syn 1972,
4
, 73–75.
3
0 Natta, G.; Mazzanti, G.; Pregaglia, G. F.; Pozzi, G. J Polym
Sci 1962, 58, 1201–1210.
3
1 Natta, G.; Mazzanti, G.; Pregaglia, G.; Pozzi, G. G.B. Patent
9
50,508, 1964.
ꢀ
1
32 Miller, R. G. J.; Nield, E.; Turner-Jones, A. Chem Ind 1961,
mol , with narrow polydispersity around 1.2. These poly-
4
4–46.
ꢁ
mers were found stable up to 274 C under nitrogen, and
3
3 Natta, G.; Pregaglia, G.; Mazzanti, G.; Binaghi, M.; Pozzi, G.;
ꢁ
exhibited a T near 54 C and a broad and complex endo-
g
Oddo, N.; Zamboni, V. G.B. Patent 970,018, 1964.
4 Pregaglia, G. F.; Binaghi, M. In Encyclopedia of Polymer
Science and Technology, Wiley: New York, 1968; Vol. 8, pp 45–
ꢁ
thermic peak around 139 C, suggesting a complex polymor-
3
phism. The crystallinity, measured by X-ray diffraction, was
close to 0.45.
5
7.
3
5 Natta, G.; Mazzanti, G.; Pregaglia, G.; Binaghi, M.; Peraldo,
M. J Am Chem Soc 1960, 82, 4742–4743.
3
3
3
3
4
4
4
4
4
4
6 Pommier, A.; Pons, J. M. Synthesis 1993, 5, 441–459.
7 Nations, R. G.; Hasek, R. H. U.S. Patent 3,221,028, 1965.
8 Steadman, T. R. U.S. Patent 2,424,589, 1947.
REFERENCES AND NOTES
1
2
3
4
5
Staudinger, H. Ber Dtsch Chem Ges 1905, 38, 1735–1739.
Tidwell, T. T. Ketenes; Wiley Interscience: New York, 1995.
Seikaly, H. R.; Tidwell, T. T. Tetrahedron 1986, 42, 2587–2613.
Zarras, P.; Vogl, O. Prog Polym Sci 1991, 16, 173–201.
Hasek, R. H. Encyclopedia of Polymer Science and Technol-
9 Hagemeyer, H. J., Jr. U.S. Patent 2,478,388, 1949.
0 Steadman, T. R.; Breyfogle, P. L. U.S. Patent 2,424,590, 1947.
1 K u¨ ng, F. E. U.S. Patent 2 356,459, 1944.
2 Hagemeyer, H. J., Jr. U.S. Patent 2,456,503, 1948.
3 Caldwell, J. R. U.S. Patent 2,518,662, 1950.
ogy, 3rd ed.; Wiley: New York, 1980; Vol. 13.
6
Satchell, D. P. N.; Satchell, R. S. Chem Soc Rev 1975, 4,
4 Caldwell, J. R. U.S. Patent 2,739,158, 1956.
231–250.
5 Mathes, N.; Jaacks, V. Makromol Chem 1971, 142, 209–225.
7
Hanford, W. E.; Sauer, J. C. Org React 1946, 108–140.
4
1
6 Yamashita, Y.; Ito, K.; Nakakita, F. Makromol Chem 1969,
27, 292–295.
8
Haener, R.; Laube, T.; Seebach, D. J Am Chem Soc 1985, 107,
5396–5403.
4
7 Kricheldorf, H. R.; Garaleh, M.; Schwarz, G. J Macromol Sci
9
1
Ogata, Y.; Adachi, K. J Org Chem 1982, 47, 1182–1184.
Pure Appl Chem 2005, A42, 139–148.
0 Harpp, D. N.; Bao, L. Q.; Black, C. J.; Gleason, J. G.; Smith,
4
1
8 Kricheldorf, H. R.; Jonte, J. M.; Dunsing, R. Makromol Chem
986, 187, 771–785.
R. A. J Org Chem 1975, 40, 3420–3427.
1
1
1 Clemens, R. J. Chem Rev 1986, 86, 241–318.
2 Brady, W. T. Tetrahedron 1981, 37, 2949.
4
9 Gresham, T. L.; Jansen, J. E.; Shaver, F. W.; Bankert, R. A.;
Fiedorek, F. T. J Am Chem Soc 1951, 73, 3168–3171.
1
1
3 Hayki, N.; Desilles, N.; Burel, F. Macromol Chem Phys 2011,
12, 375–382.
5
3
0 Kricheldorf, H. R.; Scharnagl, N.; Jedlinski, Z. Polymer 1996,
7, 1405–1411.
1
3
4 Staudinger, H.; Felix, F.; Harder, H. Helv Chim Acta 1925, 8,
06–332.
5
1
1 Kricheldorf, H. R.; Berl, M.; Scharnagl, N. Macromolecules
988, 21, 286–293.
1
5 Natta, G.; Mazzanti, G.; Pregaglia, G. U.S. Patent 3,350,357, 1967.
5
2 Kurcok, P.; Kowalczuk, M.; Hennek, K.; Jedlinski, Z. Macro-
1
6 Natta, G.; Mazzanti, G.; Pregaglia, G.; Binaghi, M. Makromol
molecules 1992, 25, 2017–2020.
3 Kurcok, P.; Matuszowicz, A.; Jedlinski, Z.; Kricheldorf, H. R.;
Dubois, P.; Jerome, R. Macromol Rapid Commun 1995, 16,
Chem 1961, 44, 537–549.
5
1
7 Pregaglia, G.; Binaghi, M. Macromol Syn 1969, 3, 152–160.
18 Egret, H. Ph.D. Thesis, University of Rouen, Rouen, France,
5
13–519.
1998.
5
4 Jedlinski, Z.; Kowalczuk, M.; Kurcok, P. Macromolecules
19 Egret, H.; Couvercelle, J.-P.; Belleney, J.; Bunel, C. Eur
1
991, 24, 1218–1219.
Polym J 2002, 38, 1953–1961.
5
5 Jedlinski, Z.; Kurcok, P.; Kowalczuk, M. Macromolecules
20 Bienvenu, A. Ph.D. Thesis, INSA of Rouen, Rouen, France, 2004.
1
985, 18, 2679–2683.
21 Hasek, R. H.; Clark, R. D.; Elam, E. U.; Martin, J. C. J Org
5
6 Moller, M.; Kange, R.; Hedrick, J. L. J Polym Sci Part A:
Chem 1962, 27, 60–64.
Polym Chem 2000, 38, 2067–2074.
22 Yamashita, Y.; Nunomoto, S. Makromol Chem 1962, 58,
5
7 Vivas, M.; Mejias, N.; Contreras, J. Polym Int 2003, 52,
244–246.
1
005–1009.
23 Yamashita, Y.; Miura, S.; Nakamura, M. Makromol Chem
5
8 Jaipuri, F. A.; Bower, B. D.; Pohl, N. L. Tetrahedron: Asym-
1963, 68, 31–47.
metry 2003, 14, 3249–3252.
2
2
2
4 Hasek, R. A.; Elam, E. U. U.S. Patent 3,201,474, 1965.
5 Cash, G. O.; Martin, J. C. U.S. Patent 3,321,441, 1967.
6 Natta, G.; Mazzanti, G.; Pregaglia, G.; Binaghi, M. J Am
5
9 Kricheldorf, H. R.; Kreiser-Saunders, I.; Scharnag, N. Makro-
mol Chem Macromol Symp 1990, 32, 285–298.
60 Kricheldorf, H. R.; Eggerstedt, S. Macromolecules 1997, 30,
Chem Soc 1960, 82, 5511–5512.
7 Natta, G.; Mazzanti, G.; Pregaglia, G.; Binaghi, M. G.B. Pat-
ent 1,015,395, 1965.
5693–5697.
2
61 Kricheldorf, H. R.; Lee, S.-R.; Scharnagl, N. Macromolecules
1994, 27, 3139–3146.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138
4137

ARTICLE
WWW.POLYMERCHEMISTRY.ORG
62 Jedlinski, Z.; Kurcok, P.; Kowalczuk, M.; Kasperczyk, J. Mak-
73 Rieth, L. R.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J
romol Chem 1986, 187, 1651–1656.
Am Chem Soc 2002, 124, 15239–15248.
63 Kricheldorf, H. R.; Scharnagl, N. J Macromol Sci Chem
74 Guillaume, C.; Carpentier, J.-F.; Guillaume, S. M. Polymer
1989, A26, 951–968.
2009, 50, 5909–5917.
64 Wilson, D. R.; Beaman, R. G. J Polym Sci Part A-1: Polym
75 Barbaud, C.; Abdillah, F.; Fabienne, F.; Guerrouache, M.;
Chem 1970, 8, 2161–2170.
5 Jedlinski, Z.; Kurcok, P.; Kowalczuk, M.; Matuszowicz, A.;
Dubois, P.; Jerome, R.; Kricheldorf, H. R. Macromolecules 1995,
Guerin, P. Des Monomers Polym 2003, 6, 353–367.
6
76 Barbaud, C.; Fay, F.; Abdillah, F.; Randriamahefa, S.; Guerin,
P. Macromol Chem Phys 2004, 205, 199–207.
28, 7276–7280.
77 Moyano, A.; Pericas, M. A.; Valenti, E. J Org Chem 1989, 54,
66 Teranishi, K.; Araki, T.; Tani, H. Macromolecules 1972, 5,
573–582.
660–661.
78 Jackson, M. S.; Wild, M. S.; Varley, J. H. G.B. Patent
67 Teranishi, K.; Iida, M.; Araki, T.; Yamashita, S.; Tani, H. Mac-
2,115,432, 1983.
romolecules 1974, 7, 421–427.
79 Van Duuren, B. L.; Langseth, L.; Goldschmidt, B. M.; Orris,
68 Iida, M.; Araki, T.; Teranishi, K.; Tani, H. Macromolecules
L. J Natl Cancer Inst 1967, 39, 1217–1228.
1977, 10, 275–284.
80 Salkowski, H., Jr. J Prakt Chem 1923, 106, 253–265.
69 Ramiandrasoa, P.; Guerin, P.; Girault, J. P.; Bascou, P.;
81 Pregaglia, G. F.; Binaghi, M. Macromol Syn 1968, 3,
Hammouda, A.; Cammas, S.; Vert, M. Polym Bull 1993, 30,
152–160.
5
01–508.
8
8
2 Stanin, T. E.; Brown, V. L., Jr. Fr Patent 1,381,831, 1964.
3 Mugno, M.; Bornengo, M. Chimica e l’Industria 1964, 46,
70 Etienne, Y.; Fischer, N. Fr Patent 1,231,163, 1960.
71 Ames, W. A.; Ward, J. J.; Carter, E. H. J. U.S. Patent
5–9.
3,879,352, 1975.
8
4 McConnell, W. V.; Moore, W. H. J Org Chem 1963, 28,
72 Bigdeli, E.; Lenz, R. W. Macromolecules 1978, 11, 493–496.
822–827.
4
138
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2011, 49, 4129–4138