Acetalization of poly(vinyl alcohol) by a fatty aldehyde in water medium: Model study, kinetics, and structure analysis

Article abstract of DOI:10.1002/pola.28942

Acid catalyzed poly(vinyl alcohol) (PVA) acetalization was investigated in aqueous medium at 80?°C for a PVA concentration of 8 wt?%. The reactant, 10-undecenal, was composed of a long alkyl chain with a vinyl end group, and the functionalization reaction was studied in heterogeneous media for low reactant concentrations (from 0.33 to 2.0 mol?% compared with PVA hydroxyl groups concentration). First, the reaction was scrutinized with pentane-2,4-diol, as a model compound of PVA. Besides the expected reaction, the oxidation of the aldehyde into 10-undecenoic acid in the presence of water was evidenced. This carboxylic acid appeared unreactive toward esterification of pentane-2,4-diol and PVA in water. Characterization of acetal stereochemical structure formed on the PVA backbone was performed by NMR spectroscopy in accordance to the model approach. A protocol based on ¹H NMR analysis was developed to quantify grafted aldehyde, residual aldehyde, and created carboxylic acid through direct sampling of the reaction medium. Conversions and reaction rate constants were calculated for pH ranging from 1 to 3. Finally, the acetalization yield was found to be enhanced at low pH and, in such conditions, the oxidation reaction contribution was limited.

Full text of DOI:10.1002/pola.28942

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Acetalization of Poly(vinyl alcohol) by a Fatty Aldehyde in Water Medium:
Model Study, Kinetics, and Structure Analysis
1
2
Pierre Piluso, Fernande Da-Cruz Boisson, Veronique Bounor-Legare ,1 Eliane Espuche1
ꢀ
ꢀ
1
ꢀ
ꢀ
ꢀ
Univ Lyon, Universite Claude Bernard Lyon1, UMR CNRS5223, Ingenierie des Materiaux Polyme`res, Lyon, F-69622, France
2
ꢀ
ꢀ
Univ Lyon, INSA Lyon, UMR CNRS5223, Ingenierie des Materiaux Polyme`res, Villeurbanne, F-69621, France
ꢀ
Correspondence to: B.-L. Veronique (E-mail: bounor@univ-lyon1.fr)
Received 7 November 2017; accepted 11 December 2017; published online 00 Month 2017
DOI: 10.1002/pola.28942
ABSTRACT: Acid catalyzed poly(vinyl alcohol) (PVA) acetaliza-
tion was investigated in aqueous medium at 80 8C for a PVA
concentration of 8 wt %. The reactant, 10-undecenal, was com-
posed of a long alkyl chain with a vinyl end group, and the
functionalization reaction was studied in heterogeneous media
for low reactant concentrations (from 0.33 to 2.0 mol % com-
pared with PVA hydroxyl groups concentration). First, the reac-
the PVA backbone was performed by NMR spectroscopy in
accordance to the model approach. A protocol based on 1H
NMR analysis was developed to quantify grafted aldehyde,
residual aldehyde, and created carboxylic acid through direct
sampling of the reaction medium. Conversions and reaction
rate constants were calculated for pH ranging from 1 to 3.
Finally, the acetalization yield was found to be enhanced at
low pH and, in such conditions, the oxidation reaction contri-
tion was scrutinized with pentane-2,4-diol, as
a model
C
V
bution was limited.
2017 Wiley Periodicals, Inc. J. Polym.
compound of PVA. Besides the expected reaction, the oxida-
tion of the aldehyde into 10-undecenoic acid in the presence of
water was evidenced. This carboxylic acid appeared unreactive
toward esterification of pentane-2,4-diol and PVA in water.
Characterization of acetal stereochemical structure formed on
Sci., Part A: Polym. Chem. 2017, 00, 000–000
KEYWORDS: aldehydes; kinetics; NMR spectroscopy; polymers;
water chemistry
acetaldehyde in water to control the concentration and tem-
perature domain in which modified PVA remained water sol-
uble.16 Similarly, Macho et al. used various aldehydes to
acetalize PVA at low degrees (from 1.5 to 17.9 mol %) to
tune its lipophilicity.17 PVA acetalization reaction was also
explored in more details by several authors, first using
model compounds. Nakamura and Suzuki used butane-1,3-
diol as a model compound of PVA to understand the reaction
mechanisms and kinetics and then transfer these knowledge
to study PVA acetalization and its reverse reaction, namely,
hydrolysis.18 They showed that, for 6 wt % diol in water, an
increase in acid concentration (0.01 to 0.1 mol L21) or tem-
perature (25 to 40 8C) led to an increase in acetalization
rate, the phenomenon being the same for PVA acetalization.
The increase in temperature in the studied range also drove
the reaction equilibrium toward acetal formation. Fujii et al.
acetalized pentane-2,4-diol and PVA with acetaldehyde and
studied the hydrolysis reaction of the acetals formed in a
mixture of 87.5% (v/v) dioxane/water at 60 8C using
0.02 mol L21 hydrochloric acid. Results showed that acetals
of isotactic diads from PVA were less prone to cleavage com-
pared with acetals of syndiotactic diads.19–21 Shibatani et al.
obtained similar results by comparing the reactivity of
INTRODUCTION Poly(vinyl alcohol) (PVA) is a synthetic poly-
mer that finds numerous applications, notably due to its
high water solubility, easy process, low cost, and commercial
availability.1–4 PVA is used extensively in packaging, textile,
and paper industry. Its ability to form hydrogen bonds
through the hydroxyl groups along the polymer chain leads
to materials with a high cohesive energy density and a sig-
nificant hydrophilicity. Moreover, the properties of PVA can
be easily tuned through chemical modifications involving
these hydroxyl groups. Esterification, acetalization, urethana-
tion, etherification, or click chemistry reactions have been
applied successfully to this polymer.5–9
Among all these possible reactions, acetalization has been
one of the most widely used. Its main application is the pro-
duction of polyvinyl acetals such as poly(vinyl formal) and
poly(vinyl butyral).10–14 In these cases, up to 80 mol % of
PVA hydroxyl groups are acetalized and the resulting mate-
rial properties (solubility, glass transition temperature,
mechanical, and adhesive properties) differ drastically from
native PVA.15 Besides, some authors focused on lower acetal-
ization degrees to tune some of the properties of the poly-
mer while keeping its main native ones. Christova et al.
performed from 10 to 25 mol % PVA acetalization by
C
V
2017 Wiley Periodicals, Inc.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Description of the Reaction Media Used to Study the
formaldehyde with the meso- and racemic diol units of
heptane-2,4,6-triol and with the meso- and racemic pentane-
2,4-diol. In that case, the reaction was performed from 30 to
80 8C, in aqueous medium using 0.1 mol L21 HCl.22 Accord-
ing to our knowledge, all the reactions studied were carried
out in homogeneous water medium.
Model Reaction
Reaction Medium
Bulk
Concentrated
Diluted
Water (mmol)
2.5
13.3
2.30
2.30
161.7
2.30
2.30
Pentane-2,4-diol (mmol)
2.30
2.30
10-Undecenal or
10-undecenoic
acid (mmol)
Concerning PVA acetalization, a wide range of acetalization
degrees were studied using various solvents. Gousse and
Gandini studied the acetalization at room temperature of
fully hydrolyzed PVA with a bulky aldehyde, furfural, in
water, and dimethylsulfoxide (DMSO). Such modified poly-
mers precipitated during the reaction when carried out in
water. 50 mol % of the initial furfural was grafted onto the
polymer, and the reaction yield did not depend on the sol-
vent used.8 Delattre et al. reported the use of a wide range
of water insoluble aldehydes such as benzaldehyde, naphtyl,
or aliphatic derivatives, notably undecanal, to modify from
80 to 99 mol% hydrolyzed PVA with Mw ranging from 9,000
to 175,000 g mol21. Several organic solvents (tetrahydrofu-
ran, acetonitrile, and DMSO) were used.23 The acetalization
degrees ranged from 18 to 80 mol % using 1.2 equivalent of
aldehyde per two hydroxyl groups of the PVA and the acetal-
ization yields did not depend on the PVA chain length. More-
over, they used 4-nitrobenzaldehyde to modify a 80 mol %
hydrolyzed PVA at 60 8C in THF using p-toluenesulfonic acid
as catalyst. However, they did not succeed in acetalizing
PVAs with degree of hydrolysis (DH) equal or higher than 92
mol % in the same conditions. Smets and Petit also observed
such tendency and thus the difficulty to acetalize PVA of
high DH.24 Moreover, Rumyantsev et al. showed that acetali-
zation kinetics was slowed down by the addition of acetal
groups onto PVA, especially in water.25 While the reaction
duration of previous works dealing with PVA acetalization
mostly exceeded 4 h, Toncheva et al. managed to acetalize
fully hydrolyzed PVA in aqueous medium using benzaldehyde
derivatives with low polarities in relatively low reaction
times.26 For example, the use of 4-methoxybenzaldehyde at
60 8C with 0.22 mol L21 sulfuric acid allowed to obtain 50
mol % acetalization degree in 2 h. It is noticeable that none
of the studies reported the use of a fatty aldehyde to acetal-
ize PVA in water as solvent.
solvent and a fast (less than 90 min) one-pot process. The
amount of reactant will be maintained below 2 mol % to limit
the modification of bulk material properties. The vinyl end
group brought out by the 10-undecenal can be useful for sub-
sequent reactions. A model reaction will be first performed in
different media to identify the reaction products and, to evi-
dence the reaction mechanism and kinetics. Thanks to this
model study, a quantitative NMR method will be introduced
allowing the quantification of PVA acetalization reaction yields
in the reactant concentration range studied. Finally, PVA acetali-
zation kinetics will be followed in water as a function of
medium acidity using the developed method.
EXPERIMENTAL
PVA (DH: 98 mol %, viscosity of a 4 wt % solution in water:
28–31 mPa s21) was purchased from Wego Chemical Group.
NMR analysis (not shown here) revealed that this PVA was
atactic (48 mol % isotactic diads and 52 mol % syndiotactic
diads). 10-undecenal (95% purity) was purchased from Kal-
psutra Chemicals. Pentane-2,4-diol (98% purity, 50.3 mol %
of racemic diol and 49.7 mol % of meso-diol, as determined
by 1H NMR, not shown here) was purchased from Fluka
Chemika.47 10-undecenoic acid (98% purity), sulfuric acid
(96% purity), perdeuterated dimethyl sulfoxide (DMSO-d6),
and perdeuterated chloroform (CDCl3) were provided by
Sigma-Aldrich. All materials, reagents, and solvents were
used as received without further purification.
Equimolar amount of pentane-2,4-diol (253 mL, 2.30 mmol)
and 10-undecenal (462 mL, 2.30 mmol) or 10-undecenoic
acid (464 mL, 2.30 mmol) were added into a 3-mL vial. For
bulk conditions studies, 50 mL of a 10 wt % sulfuric acid
solution were added as catalyst to obtain a final sulfuric acid
concentration of 6.6.1022 mol L21. For solution studies, dis-
tilled water was previously acidified to pH 5 2 with a 30
wt % sulfuric acid solution. Then, 240 mg or 2.91 g of acidi-
fied water were added to the reactional medium for concen-
trated or diluted medium studies, respectively. The final diol
concentrations were 50 wt % for the concentrated medium
and 8 wt % for the diluted medium. The vial was then
heated at 80 8C for 90 min in an oil bath under magnetic
stirring at 800 rpm. To withdraw samples for NMR analyses,
the vial was removed from the oil bath and, after a quick
decantation, a small quantity of the organic phase was trans-
ferred in a 2-mL vial and diluted with the NMR solvent. The
main conditions studied are summarized in Table 1.
Among the previous literature on PVA acetalization, reaction
yields and kinetics were often followed by chemical titration
of the aldehyde or remaining OH-groups, implying several
reaction, titration, and purification methods.17,27,28 These
procedures, although reliable, present the disadvantage to be
quite laborious and lack sensitivity when quantifying low
reactant concentrations. NMR was also used extensively for
poly(vinyl acetal) characterization, including identification
and quantification of minor substructures and thus fulfills
the prerequisites to follow acetalization kinetics, even for
very low reactant concentrations.8,16,25,29–46
This study aims at grafting low amounts (less than 2 mol %)
of a fatty aldehyde onto PVA through acetalization. The reac-
tion will be carried out in a green fashion using water as
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 1 Reaction scheme of the model acetalization with the description of the different chair conformation of acetal stereo-
isomers that have been observed.
PVA (22 g) was added slowly into a 250-mL three-neck
round-bottom flask filled with 178 mL of distilled water. The
flask was heated at 90 8C under mechanical stirring at
400 rpm for 60 min to dissolve the PVA. Then, 75 mL dis-
tilled water were added and the temperature of the solution
was stabilized at 80 8C. The solution was acidified to pH val-
ues 1, 2, or 3 with a solution of 10 wt % sulfuric acid. 10-
undecenal 308 mL (1.54 mmol) for expected 0.33 mol %
grafted aldehyde and 1867 mL (9.33 mmol) for expected 2.0
mol % grafted aldehyde were added and the acetalization
reaction was studied as a function of time and pH.
conditions (concentrated or diluted medium). Previous studies
used model alcohol compounds such as butane-1,3-diol, pen-
tane-2,4-diol, or heptane-2,4,6-triol.8,18–22,25 Among those com-
pounds, pentane-2,4-diol, with its two secondary alcohol
functions, appears to be the most representative of a PVA unit.
Thus, a 50/50 mixture of meso- and racemic pentane-2,4-diol
(abbreviated as diol in the document) was chosen as an effi-
cient model reagent to perform the acid-catalyzed acetalization
with 10-undecenal as depicted in Figure 1. For the sake of sim-
plicity, the (R;R) diastereoisomer is noted trans-acetal and the
(R;S) diastereoisomer is noted cis-acetal.
NMR Identification of the Reaction Products
Purification of modified PVA was achieved through precipita-
tion in acetone from hot PVA solutions (T ꢀ 60 8C). After fil-
tration, the solid was recovered and dried at room
temperature during 48 h.
The model acetalization reaction was first carried out in
bulk medium to identify the reaction products. Acetal’s
structures were determined by analyzing the 90 min reaction
1
time crude medium by H, 13C (including DEPT135), Correla-
1
1D H-{13C} spectra, 13C, and Distortionless Enhanced Polari-
tion SpectroscopY (COSY), Heteronuclear Single Quantum
Coherence spectroscopy (HSQC), and Heteronuclear Multiple-
Bond Correlation spectroscopy (HMBC) (DEPT135 and 2D
zation Transfer (DEPT) 135 NMR spectra of the model com-
pound, and 2D 1H/1H and 1H/13
C correlations were
NMR analyses not shown here).47 Attribution of H spectrum
1
recorded with a 5-mm BBFO1 probe on a Bruker AVANCE
III spectrometer operating at 400.1 MHz for 1H and 100.6
MHz for 13C. Experiments were carried out at 298 K for the
model compound and at 313 K for PVA study using CDCl3
and DMSO-d6 as solvent, respectively.
is detailed on Figure 2. 1H and 13C NMR chemical shifts are
displayed in Table 2 and compared with those of similar
aldehydes from the literature.48–52
Two proton signals were observed at 4.50 and 4.82 ppm,
indicating the formation of two acetals diastereoisomers as
depicted in Figure 1. Chemical shifts and multiplicities were
in good agreement with data reported in the literature for
similar structures. The formation of two acetal diaster-
eoisomers is explained by the presence of two isomers in
the initial pentane-2,4-diol. A global 95 mol % diol conver-
sion was calculated from NMR spectra and resulted in fact,
from the conversion of 90 mol % of the racemic diol into its
corresponding trans-acetal and to the total conversion of the
meso-diol into the cis-acetal form. From these results, it
could be concluded that the reactivity of the meso-diol
toward acetalization in our experimental conditions was
slightly higher than the reactivity of the racemic diol. These
results were in agreement with previous studies evidencing
that reaction with the meso-diol is kinetically and thermody-
namically favored.22,47 Indeed, Fujii et al. calculated the equi-
librium constant for the acetalization of meso- and racemic
1D 13C and DEPT135 spectra of native and modified PVA
were recorded at 313 K with a 10 mm 13C selective probe
on a Bruker AVANCE II spectrometer operating at 100.6 MHz
for 13C. DMSO-d6 was used as solvent.
For modified PVA analysis, 100 mg aliquots of PVA solutions
were withdrawn from the reaction medium at different reac-
tion times and diluted without further treatment in 700 mg
of DMSO-d6.
RESULTS AND DISCUSSION
Study of the Model Compound Acetalization
Our study investigated a very low functionalization degree
(less than 2 mol %) of PVA chains with hydrophobic moieties
through acid catalyzed acetalization in water medium. A model
reaction was investigated to facilitate further analysis of acetal-
ized PVA NMR spectra and to observe the effect of the reaction
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 2 1H NMR spectrum (CDCl3 at 298 K) of the acetalization model reaction medium after 90 min at 80 8C (reaction in bulk
medium with H2SO4 as catalyst).
pentane-2,4-diol at 60 8C in bulk conditions. The equilibrium
constant of meso-diol acetalization was 8.70 3 103 versus
3.70 3 102 for the racemic diol acetalization, showing the
higher stability of the cis-acetal ring.19 The same observa-
tions were made by Shibatani et al. who calculated equilib-
rium and reaction rate constants for the acetalization of
TABLE 2 Experimental 1H and 13C NMR Chemical Shifts for Model Acetal Compared with Literature Data
1H Chemical Shifts (ppm) 13C Chemical Shifts (ppm)
Literature
Trans
Experimental
Trans Cis
Literature
Trans
Experimental
Trans Cis
H/Ca
Cis
Cis
1-Axial
1-Equatorial
2-Axial
2-Equatorial
3-Axial
3-Equatorial
4
1.34b21.35c
1.20b21.15c
3.95c
4.28c
1.86c
1.20–1.35c
4.82b,c
1.63g
1.30g
1.37g
–
1.32
1.20
3.93
4.28
1.82
1.33
4.82
1.54
1.25-1.45
1.35
2.03
5.80
4.92
4.98
–
18c
22c
67c
68c
37c
–
17
22
67
68
37
–
1.21d
4.00e
–
1.11e
1.48e
4.72f
1.20
3.69
–
22e
65e
22
72
1.23
1.51
4.50
1.60
41e
99e
41
94c
94
102
5
33g
29g
22g
34g
139g
114g
35
6–10
29
11
24
12
2.03g
5.80g
4.92g
4.98g
34
13
139
114
Cis-14
Trans-14
a
e
f
Atoms references are displayed in Figure 1.
From Ishihara et al.48
From Matsuo et al.51
From Lorente et al.52
b
c
g
From Denmark et al.49
Experimental data from our 1H and 13C NMR spectra of 10-undecenal
d
From Denmark et al.50
in CDCl3 at 298 K.
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
proton signal (13) at 5.80 ppm was used as a reference for
quantification.
From the data obtained for different reaction times, the acetali-
zation yields versus time were calculated for aqueous media
(Fig. 3). The conversions were quite low and highly dependent
on dilution. Indeed, for 90 min reaction time, acetalization
yields decreased from 40 mol % to 7 mol % when the diol
concentration decreased from 50 wt % to 8 wt %. This phe-
nomenon is likely to be due to the dilution of reactive species
and to poor contact between organic and aqueous phases while
carrying out the reaction in biphasic medium. Both phenomena
are emphasized as the amount of water increases.
FIGURE 3 Evolution of the amount of acetal formed (mol %)
versus time in two aqueous media at 80 8C.
Besides, to evidence the influence of the reaction media on
acetalization stereo-orientation, the ratios between cis-acetal
and trans-acetal were also calculated according to 1H NMR
analyses and compared with the initial meso- over racemic
diol ratio (0.99). The ratios obtained for 90 min reaction
time were compared in Table 3. The formation of cis-acetal
was favored in all cases. In bulk medium, the reaction
reached its equilibrium. A weak orientation of the reaction
to the formation of the cis-isomer was observed (ratio cis-
over trans-acetal: 1.1). In aqueous media, the reaction was
only observed in its early stages before the equilibrium
could be attained.
pentane-2,4-diol with formaldehyde in acidic aqueous
medium at several temperatures.22
Signals from protons (1) to (5) (Table 2 and Fig. 2) are
shown to be dependent on the stereochemistry of the acetal
which chair conformation structures are shown in Figure
1.53,54 The cis-acetal (obtained through acetalization of the
meso-diol) could reach a stable chair configuration with the
methyl groups and the alkyl chain all in equatorial position,
leading to a symmetrical molecule. On the contrary, the ace-
tal ring of the trans-acetal (obtained through acetalization of
the racemic diol) adopted a configuration with one of the
two methyl groups in axial position. It was previously shown
that those acetals did not adopt boat conformations. More-
over, no exchange between two chair conformations was evi-
denced by NMR for temperatures ranging from 277 to
150 8C. Indeed, for the trans-isomer, two different signals
were observed for the methyl (1) and (10) and the methine
(2) and (20) protons of the 1,3-dioxane ring. Similarly, for
both acetals, the axial and equatorial protons (3) were not
equivalent. Protons (3) signals were observed at 1.82 ppm
and 1.33 ppm for the trans-acetal and 1.51 ppm and 1.23
ppm for the cis-acetal. Chemical shifts obtained through 13C
analysis (not shown here) were also in accordance with the
formation of both cis-acetal and trans-acetal forms.
The orientation was strong and equivalent in both aqueous
media studied (1.8 and 1.9 for 50 wt % and 8 wt % concen-
trated diol, respectively).19,31 Significant reaction orientation
differences could only be observed when the reaction was
not complete and indicated that the kinetics of formation of
the trans-acetal was lower than the cis-acetal one. Observa-
tion of the reaction at early stages in bulk medium and
kinetics on larger reaction times in aqueous media could
confirm this hypothesis but were not investigated.
1
As described previously, H NMR spectra were used to calcu-
late 10-undecenal, 10-undecenoic acid, and acetal percen-
tages in the organic phases (Fig. 4). In both cases, formation
of 10-undecenoic acid was observed and kinetics relied on
the diol concentration. In concentrated medium, acetal for-
mation rate remained stable for the duration of the reaction.
After 90 min, the aldehyde conversion reached 66 mol %,
and 27 mol % of the aldehyde was converted into carboxylic
acid. In diluted medium, the acetalization reaction was only
Influence of the Reaction Medium on Kinetics
The model reaction was also carried out in diluted and con-
centrated aqueous media. In these cases, the medium was
heterogeneous with an aqueous and an organic phase. The
latter phase, which contained all the compounds with the
vinyl alkyl group, was analyzed by 1H NMR. Actually, it was
not possible to calculate a diol conversion from the resulting
spectra because the diol partially remained in the aqueous
phase. It is also noteworthy that formation of 10-undecenoic
acid by oxidation of 10-undecenal was observed in these
experimental conditions. Integration values of proton signals
from the aldehyde group at 9.76 ppm, cis- and trans-acetal
methine (4) at, respectively, 4.50 and 4.82 ppm, and methy-
lene next to carboxylic acid group of 10-undecenoic acid at
2.34 ppm were used to calculate the amounts of unreacted
aldehyde, acetal, and carboxylic acid. The area of vinyl
TABLE 3 Comparison of the Ratios cis-Acetal Over trans-Acetal
Depending on the Reaction Medium for 90 Min Reaction Time
Ratio cis-Acetal/
trans-Acetal at 90 Min
Reaction Time
Global
acetalisation
Yield (mol %)
Reaction Medium
Pentane-
2,4-diol
0,99
-
(Meso-/racemic)
Bulk
1.1
1.8
1.9
95
39
7
Concentrated
Diluted
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
The aldehyde oxidation being a side reaction in aqueous sol-
utions, we specifically studied the reactivity of the acid with
the diol. The esterification reaction was first carried out in
bulk medium with analogous conditions than for acetaliza-
tion model studies (Fig. 5).
Formation of mono-ester and di-ester was confirmed by GC-
MS and NMR analyses (not shown here). Conversion was
1
determined according to H NMR spectra for several reaction
times up to 180 min (Fig. 6). The monoester was formed
rapidly in the early stages of the reaction and its amount
reached a plateau at around 50 mol % conversion after 90
min, indicating that the monoesterification equilibrium was
reached. The kinetics of formation of the diester was
observed to be much slower than in the case of monoester
formation.
When the esterification reaction was carried out in water
solutions, no ester formation was observed whatever our
conditions. When compared with model compound acetaliza-
tion, it could be concluded that 10-undecenoic acid exhibited
a much lower reactivity toward the diol compound com-
pared with 10-undecenal. The absence of reactivity in aque-
ous medium could be explained by a decrease in reaction
kinetics due to the dilution of the reactants and by a shift of
the esterification reactions equilibria toward reactants in
presence of large excess of water. As a conclusion of these
model studies, the acetalization reaction studied appeared to
be selective in the conditions investigated. The presence of
water appeared to both reduce the acetalization kinetics by
dilution effect and to prevent the reaction equilibrium to be
shifted toward the acetal product. Oxidation of 10-undecenal
as a side reaction was a major issue to obtain efficient graft-
ing of aldehyde onto pentane-2,4-diol. Depending on the
dilution conditions, this phenomenon could consume up to
27 mol % of the initial aldehyde in 90 min at 80 8C.
FIGURE 4 Evolution of the species percentage (mol %) versus
time for the acetalization model reaction carried out at 80 8C in
several media: (a) at pH 5 2 in 50 wt % concentrated diol aque-
ous solution; (b) at pH 5 2 in 8 wt % concentrated diol aqueous
solution.
observed after 75 min of reaction. Aldehyde conversion
attained 7 mol % after 90 min. A constant value of 3 mol %
carboxylic acid was observed, corresponding to its native
amount in 10-undecenal.
PVA Acetalization
Reactivity Toward Esterification
These observations indicated that the more diluted medium
prevented the formation of acetalization products and
decreased acetalization reaction kinetics. The presence of
water in such acid conditions also allowed 10-undecenoic
acid to be formed through oxidation of 10-undecenal.
Aldehyde oxidation observed in aqueous media for model
study led us to consider the possible esterification of PVA by
10-undecenoic acid. Thus, an esterification reaction was per-
formed at a pH value of 1 by substituting 10-undecenal with
10-undecenoic acid using the protocol described in the
FIGURE 5 Reaction scheme of the model esterification reaction and molecular structure of its products.
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
4.50 ppm or 4.82 ppm, where methine protons signals of the
acetal ring (11) should appear, partially due to superposition
with hydroxyl protons’ signals.
The main proof of efficient grafting of 10-undecenal onto
PVA through acetalization came from 13C NMR and DEPT135
spectra of acetalized PVA, which depicted a single signal at
101 ppm corresponding to the methine carbons resonance
(7 and 9) of the cis-acetal ring (Fig. 9). Formation of trans-
acetal ring was not evidenced by 13C NMR analyses, indicat-
ing that the acetalization reaction selectively proceeded on
PVA isotactic diads. 1H NMR chemical shifts of Table 4 were
mainly determined thanks to HSQC correlations.
FIGURE 6 Compounds proportion (mol %) versus time from
the esterification reaction (80 8C, in bulk, with H2SO4 as
catalyst).
The higher reactivity of isotactic diads has already been
observed by several authors. Fujii et al. investigated acetaliza-
tion reaction by acetaldehyde of PVAs with different tacticities
(2, 42, and 57 mol % syndiotacticity).19 At the acetalization
reaction equilibrium, grafted aldehyde percentages were
higher for the most isotactic PVAs: 85.4, 82.7, and 77.5
mol %, respectively. Shibatani et al. studied the acetalization
kinetics of several PVAs with formaldehyde in dioxane-water
solution (50–50) at 60 8C with 0.1 mol L21 hydrochloric acid
as catalyst.22 They showed that a 72 mol % isotactic PVA
reacted faster than an atactic PVA or a 40 mol % isotactic
PVA; 15.5 mol % acetalization degree was reached after 28 h
at 60 8C for the 72 mol % isotactic PVA versus 12.5 mol % for
the 40 mol % isotactic one. They also revealed that, for all
kinds of PVAs, formation of cis-acetal was favored in great
extents: for the 72 mol % isotactic PVA, a ratio cis-formal
over trans-formal of 5 was obtained. By following the kinetics
of formation of cis- and trans-formal, they also evidenced an
exchange from trans-formal rings to cis-formal rings at long
reaction times. In their conditions, formation of cis-formal
was kinetically, and thermodynamically favored. Pollers et al.
studied acetalization of PVA with formaldehyde in water at
50 8C using 2 vol% H2SO4 as catalyst during 24 h and
revealed similar orientation of the reaction: a 27.6 mol % ace-
talized PVA revealed only 8 mol % trans-acetal rings in its
structure.32 These studies dealt with high acetalized PVA
degrees, whereas in our study, grafted aldehyde percentages
did not exceed 2 mol %. This difference could explain that
the orientation toward the formation of cis-acetal was total in
our conditions.
experimental section. No reaction between the carboxylic
acid and PVA was observed. As expected from the results of
the esterification model study in aqueous media, the large
excess of water was sufficient to prevent the formation of
the ester. As these conditions were the most favorable to
observe carboxylic acid reactivity, it was assumed in the fol-
lowing work that PVA esterification did not occur during
PVA acetalization.
Synthesis, NMR Characterization of Acetalized PVA, and
Stereo Orientation of the Reaction
Hydrophobic moieties with vinyl end groups were grafted
onto PVA through acetalization with 10-undecenal at 80 8C,
at acidic pH ranging from 1 to 3 (Fig. 7), aiming for grafted
aldehyde percentages ranging from 0.33 to 2 mol %. As 10-
undecenal is not soluble in water, the reaction medium was
heterogeneous.
Synthesis was first carried out at a pH value of 2, at 80 8C
during 90 min to obtain a highly acetalized PVA (expected
modification degree: 2 mol %). The modified PVA was char-
1
acterized by H NMR after purification (Fig. 8).
1
Characterization of initial PVA by H and 13C NMR was based
on previous work and will not be detailed here. Signals cor-
responding to the modified PVA chain were attributed thanks
1
to our model study (Table 4). The H spectrum exhibited the
alkyl chain methylene (13–18) signals from 1.25 to 1.45
ppm. Vinyl protons resonances were well separated from
other signals at 5.79 ppm (20) and at 4.96 ppm (21). How-
ever, no additional signal could directly be observed around
PVA Acetalization Kinetics
Quantification Method
Among all the acetal signals of the 1H spectrum, only the
methine vinyl proton signal (20) was not overlapping
FIGURE 7 Scheme of PVA acetalization reaction and molecular structure of the acetalized polymer.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 8 1H NMR spectrum of acetalized PVA after purification (313 K in DMSO-d6).
another signal from PVA. By comparing the integrals from
this proton signal to an unmodified signal from PVA such as
the methylene groups (2, 5, 8, and 10) multiplet, it is theo-
retically possible to quantify the percentage of acetalized
OH-groups. However, as the vinyl proton signal intensity was
very weak, the comparison with strong signals intensity
from unmodified PVA, could induce an important error. Thus,
final analysis of pure modified PVA was not used to quantify
the percentage of acetalized OH-groups.
aliquots withdrawn during acetalization. This method pro-
vided a spectrum where several signals could be used for
quantification: methine proton resonance of vinyl groups of
residual 10-undecenal, 10-undecenoic acid, and grafted alde-
hyde (20) superimposed at 5.79 ppm, resonances of methy-
lene protons next to carboxylic acid group of 10-undecenoic
acid at 2.16 ppm (23), and aldehyde proton resonance of 10-
undecenal at 9.64 ppm (22) (Fig. 10).
Integration of those signals allows determining the molar
percentage of grafted aldehyde, residual aldehyde, and car-
boxylic acid formed according to the following formula:
Another method to quantify the acetalization yield was devel-
oped which consisted in analyzing by 1H NMR PVA solution
TABLE 4 Comparison of the 1H NMR and 13C NMR Chemical Shifts Observed for Acetalized PVA in DMSO-d6 and Model Acetal in
CDCl3
1H Chemical Shifts (ppm)
Model Acetal
13C Chemical Shifts
Model Acetal Acetalized PVA
Acetalized PVA
H/Ca
trans
cis
3.69
trans
cis
7
3.93/4.28
1.33/1.82
3.93/4.28
4.82
3.75b
67–68
37
72
73b
8
1.23/1.51
3.69
Superimposed
3.75b
4.47b
41
Superimposed
73b
9
67–68
94
72
11
12
13–18
19
20
21
4.50
102
101
1.54
1.60
Superimposed
35
Superimposed
1.25–1.45
2.03
1.25
2.00b
29
29
34
33b
138
114
5.80
5.79
139
114
4.92/4.98
4.96
a
b
Atoms references are displayed in Figure 8.
Determined thanks to HSQC analysis.
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
the three species of interest versus reaction time were
reported in Figure 11.
The amount of aldehyde decreased rapidly whatever the pH
value in the considered range. As expected, a more acidic pH
led to an increase in the aldehyde consumption: at pH 5 1,
80 mol % of aldehyde was consumed in less than 10 min,
whereas a same amount needed 60 min to be consumed at
pH 5 3. The formation of the acetal was fast for low pH: the
final yield was reached in 15 min at pH 5 1 (76 mol %), 30
min at pH 5 2 (64 mol %) versus 60 min at pH 5 3 (39
mol %), corresponding to final grafted aldehyde percentages
of, respectively, 0.25, 0.21, and 0.13 mol %.
FIGURE 9 DEPT135 NMR spectra of the model acetal synthe-
tized in bulk medium (top spectrum) and the highly acetalized
PVA (bottom spectrum) at 313 K in DMSO-d6.
It was also observed that the formation of 10-undecenoic
acid was slower than acetal formation and was not favored
at low pH. Indeed, this reaction reached the final yield in 60
Ið20Þ2Ið22Þ2 Ið23Þ
2
% Grafted aldehyde5
3100
(1)
(2)
(3)
Ið20Þ
Ið22Þ
Ið20Þ
% Remaining aldehyde5
3100
Ið23Þ
% Acid formed5
3100
23Ið20Þ
I(x) is the integral value of the proton(s) signal x.
The expected grafted aldehyde percentage of the reaction
proceeded at pH 5 2 was 2.0 mol % (Fig. 10) but the effec-
tive grafted aldehyde percentage was found to be 0.72
mol %.
Influence of the pH
As it is well known, catalyst concentration significantly
impacts acetalization kinetics.18 In our case, pH values of 1,
2, and 3 were compared, all other parameters remaining
constant (80 8C, 8 wt % PVA in water, 0.33 mol % 10-
undecenal compared with OH-groups). The concentrations of
FIGURE 11 Kinetics of PVA acetalization with 10-undecenal at
different pH. (a) remaining aldehyde; (b) 10-undecenoic acid
formed; (c) Grafted aldehyde (80 8C, 8 wt % PVA in water, 0.33
mol % 10-undecenal compared with OH-groups).
FIGURE 10 1H NMR spectrum (313 K in DMSO-d6) of an aliquot
of 2.0 mol % expected acetalized PVA solution after 90 min
reaction time at 80 8C, at pH 5 2.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
9

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
CONCLUSIONS
TABLE 5 Rate Constants Calculated for PVA Acetalization at
Different pH Values
A model reaction was developed to study the acetalization
reaction of PVA with 10-undecenal in water by using pentane-
2,4-diol in several dilution conditions (50 wt % diol, 8 wt %
Rate Constant k
(L mol21 h21
)
Reaction pH
1
diol). The acetal formed was characterized by H, 13C, and 2D
1
2
3
4.6
1.8
0.8
NMR analysis. Kinetic studies revealed the preferential forma-
tion of cis-acetal over trans-acetal in our experimental condi-
tions. Oxidation reaction of the 10-undecenal into 10-
undecenoic acid was observed to compete with diol acetaliza-
tion reaction in water medium. This carboxylic acid was
unreactive toward pentane-2,4-diol esterification as well as
toward PVA esterification in water medium. Thus, oxidation
reaction was considered as a side reaction in our study.
min at pH 5 1 and pH 5 2 (20 and 31 mol % 10-undecenoic
acid, respectively). No reaction equilibrium was reached in
90 min at pH 5 3, the yield being 49 mol %.
Reaction rate constant k was calculated according to the
equation used by Kormanovskaya and Vlodavets. When p0 is
different from 2a0, the equation can be written as:55
Based on this model approach, modified PVA analysis
revealed that only the cis-acetal form was obtained in our
synthesis conditions. A reproducible method to calculate the
1
reaction yield by H NMR for initial reactant contents as low
1
12ap
k3t5
ln
(4)
as 0.33 mol % compared with PVA hydroxyl groups was
developed. Thanks to this method, acetalization kinetics
were followed for pH values ranging from 1 to 3, and
revealed that acetalization rate is higher at low pH values.
Besides, the increase in the pH enhanced aldehyde oxidation
reaction in competition to PVA acetalization reaction. Acetali-
zation rate constants were calculated and compared with
those reported in the literature for other aldehydes and reac-
tion conditions. The longer aldehyde chain and lower solubil-
ity of 10-undecenal are likely to explain its slightly lower
reactivity compared with more usual aldehydes such as
formaldehyde or acetaldehyde.
p022a0 12aa
p02p
and aa5
p0
a02a
where : ap5
a0
and with p and a being the monomer units and aldehyde
concentrations at the time t of the reaction, respectively, and
p0 and a0 being the monomer units and aldehyde initial con-
centrations, respectively.
The results obtained are gathered in Table 5. Compared with
the data obtained from other studies dealing with PVA ace-
talization reactions, our values are in the same range of con-
stant rates. For PVA acetalization with formaldehyde, the
rate constants ranged from 0.02 to 880 L mol21 h21, and for
PVA acetalization with acetaldehyde, the rate constants
Finally, this study evidenced the efficiency to functionalize
PVA with hydrophobic moieties bearing subsequent reactive
functions such as vinyl end-groups through acid catalyzed
acetalization reaction in water medium to reach low-
modification degrees ranging from 0.13 to 0.72 mol %.
ranged from 0.004 to 68 L mol21 h21 18,22,24,56
For the most
.
similar conditions (70 8C, 0.5 mol L21 sulfuric acid as cata-
lyst, PVA concentration of 0.30 mol L21 and aldehyde con-
centration of 0.15 mol L21), Ogata et al. obtained a rate
constant value of 2.45 L mol21 h21 using formaldehyde.
Despite the harsher conditions used in this study, the rate
constants remained relatively low probably due to the lack
of solubility of 10-undecenal in water medium.
ACKNOWLEDGMENTS
The authors thank the NMR Polymer center of Institut de Chi-
mie de Lyon (FR5223) for assistance and access to the NMR
facilities.
In our reaction conditions, a competition between PVA ace-
talization and oxidation reactions of 10-undecenal was
revealed. It was observed that a lower pH favored acetaliza-
tion reaction instead of oxidation phenomenon. When the pH
value increased, oxidation reaction became the predominant
reaction: at pH 5 3, 49 mol % of the aldehyde was oxidized
into its carboxylic acid and 39 mol % was grafted onto PVA.
It is noteworthy that oxidation of the aldehyde has not been
previously observed for aldehydes such as formaldehyde or
acetaldehyde.
REFERENCES AND NOTES
ꢀ
1 S. Clemenson, L. David, E. Espuche, J. Polym. Sci. Part A:
Polym. Chem. 2007, 45, 2657.
ꢀ
ꢀ
2 S. Clemenson, D. Leonard, D. Sage, L. David, E. Espuche, J.
Polym. Sci. Part A: Polym. Chem. 2008, 46, 2062.
3 J. G. Pritchard, Poly(Vinyl Alcohol) Basic Properties and
Uses; H. Morawetz, Ed.; Gordon & Breach, Science Publishers:
London, England, 1970; Vol. 4.
4 T. Okaya, In Polyvinyl Alcohol Developments; C. A. Finch,
Ed.; John Wiley & Sons: Aylesbury, England, 1992; pp 1–30.
Finally, it was possible to acetalize PVA in water medium by
acidic catalysis at effective grafted aldehyde percentages
ranging between 0.13 and 0.72 mol % of PVA hydroxyl
groups, and with a selective formation of cis-formal rings.
5 B. N. Gacal, B. Koz, B. Gacal, B. Kiskan, M. Erdogan, Y.
Yagci, J. Polym. Sci. Part A: Polym. Chem. 2009, 47, 1317.
ꢀ
6 S. J. Carlotti, O. Giani-Beaune, F. Schue, J. Appl. Polym. Sci.
2001, 81, 2868.
10
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
7 I. Sakurada, A. Nakajima, K. Shibatani, J. Polym. Sci. A. 1964,
2, 3545.
31 T. Uhlich, G. Tomaschewski, H. Komber, React. Funct.
Polym. 1995, 28, 55.
8 C. Gousse, A. Gandini, Eur. Polym. J. 1997, 33, 667.
32 I. Pollers, P. Adriaensens, R. Carleer, D. Vanderzande, J.
Gelan, Macromolecules 1996, 29, 5875.
ꢀ
ꢀ
ꢀ
9 V. Gimenez, J. A. Reina, A. Mantecon, V. Cadiz, Polymer.
1999, 40, 2759.
33 S. Amiya, M. Uetsuki, Macromolecules 1982, 15, 166.
10 Y. Pan, K. Shi, C. Peng, W. Wang, Z. Liu, X. Ji, ACS Appl.
Mater. Interfaces 2014, 6, 8651.
34 B. M. Budhlall, K. Landfester, E. D. Sudol, V. L. Dimonie, A.
Klein, M. S. El-Aasser, Macromolecules 2003, 36, 9477.
11 Y. Pan, W. Wang, C. Peng, K. Shi, Y. Luo, X. Ji, RSC Adv.
2013, 4, 660.
35 B. M. Budhlall, K. Landfester, D. Nagy, E. D. Sudol, V. L.
Dimonie, D. Sagl, A. Klein, M. S. El-Aasser, Macromol. Symp.
2000, 155, 63.
12 P. Chetri, N. N. Dass, Polymer 1997, 38, 3951.
36 D. C. Bugada, A. Rudin, Polymer 1984, 25, 1759.
13 Z. M. Zhou, D. J. David, W. J. Macknight, F. E. Karasz, Turk.
J. Chem. 1997, 21, 229.
^ ^
37 Y. Inoue, A. Nishioka, R. Chujo, J. Polym. Sci. Polym. Phys.
Ed. 1973, 11, 2237.
14 B. Wang, F. Zhang, M. Jiang, G. Ye, J. Xu, J. Appl. Polym.
Sci. 2012, 125, 2956.
38 Y. Morishima, H. Iimuro, Y. Irie, S. Nozakura, J. Polym. Sci.
Polym. Lett. 1975, 7, 157. Ed.,
15 Asahina, K. In Polyvinyl Alcohol Developments; C. A. Finch,
Ed.; John Wiley & Sons: Aylesbury, England, 1992; pp. 673–
688.
39 T. Moritani, I. Kuruma, K. Shibatani, Y. Fujiwara, Macromo-
lecules 1972, 5, 577.
16 D. Christova, S. Ivanova, G. Ivanova, Polym. Bull. 2003, 50,
367.
40 D. W. Ovenall, Macromolecules 1984, 17, 1458.
41 T. Moritani, Y. Fujiwara, Macromolecules 1977, 10, 532.
42 T. K. Wu, D. W. Ovenall, Macromolecules 1973, 6, 582.
43 T. K. Wu, M. L. Sheer, Macromolecules 1977, 10, 529.
44 A. E. Tonelli, Macromolecules 1985, 18, 1086.
ꢀ
17 V. Macho, M. Fabıni, M. Rusina, S. Bobula, M. Harustiak,
Polymer. 1994, 35, 5773.
18 N. Nakamura, K. Suzuki, J. Polym. Sci. Part A: Polym.
Chem. 1997, 35, 1719.
45 G. Van der Velden, J. Beulen, Macromolecules 1982, 15,
1071.
19 K. Fujii, J. Ukida, M. Matsumoto, Makromol. Chem. 1963,
65, 86.
46 F. F. Vercauteren, W. A. B. Donners, Polymer 1986, 27, 993.
20 K. Fujii, J. Ukida, M. Matsumoto, J. Polym. Sci. [B]. 1963, 1,
693.
47 T. Fukuroi, Y. Fujiwara, S. Fujiwara, K. Fujii, Anal. Chem.
1968, 40, 879.
21 K. Fujii, K. Shibatani, Y. Fujiwara, Y. Ohyanagi, J. Ukida, M.
Matsumoto, J. Polym. Sci. [B]. 1966, 4, 787.
48 K. Ishihara, N. Hanaki, H. Yamamoto, J. Am. Chem. Soc.
1993, 115, 10695.
ꢁ
22 K. Shibatani, K. Fujii, Y. Oyanagi, J. Ukida, M. Matsumoto,
49 S. E. Denmark, T. M. Willson, N. G. Almstead, J. Am. Chem.
Soc. 1989, 111, 9258.
J. Polym. Sci. Part C Polym. Symp. 2007, 23, 647.
23 E. Delattre, G. Lemie`re, J.-R. Desmurs, B. Boulay, E.
50 S. E. Denmark, N. G. Almstead, J. Am. Chem. Soc. 1991,
113, 8089.
~
Dunach, J. Appl. Polym. Sci. 2014, 131, 49677.
24 G. Smets, B. Petit, Makromol. Chem. 1959, 33, 41.
51 M. Matsuo, M. Kawase, Y. Sugiura, S. Takematsu, C. Hara,
Macromolecules 1993, 26, 4461.
25 M. Rumyantsev, S. V. Zelentsov, A. V. Gushchin, Eur.
Polym. J. 2013, 49, 1698.
ꢀ
52 A. Lorente, M. Pellicena, P. Romea, F. Urpı, Tetrahedron
26 V. D. Toncheva, S. D. Ivanova, R. S. Velichkova, Eur. Polym.
J. 1992, 28, 191.
2015, 71, 1023.
53 E. L. Eliel, S. M. Knoeber, J. Am. Chem. Soc. 1968, 90, 3444.
27 T. Shiomi, K. Imai, C. Watanabe, M. Miya, J. Polym. Sci.
Polym. Phys. Ed. 1984, 22, 1305.
54 M. Watanabe, Y. Kosaka, K. Oguchi, K. Sanui, N. Ogata,
Macromolecules 1988, 21, 2997.
28 K. F. Beal, C. J. B. Thor, J. Polym. Sci. 1946, 1, 540.
55 G. N. Kormanovskaya, I. N. Vlodavets, Bull. Acad. Sci. USSR
Div. Chem. Sci. 1964, 13, 1661.
29 P. A. Berger, E. E. Remsen, G. C. Leo, D. J. David, Macromo-
lecules 1991, 24, 2189.
56 Y. Ogata, M. Okano, T. Ganke, J. Am. Chem. Soc. 1956, 78,
2962.
30 M. D. Bruch, J. A. K. Bonesteel, Macromolecules 1986, 19,
1622.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 00, 000–000
11