European Polymer Journal 64 (2015) 147–156

Contents lists available at ScienceDirect

European Polymer Journal

journal homepage: www.elsevier.com/locate/europolj

Flexible polyurethane foams green production employing lignin

or oxypropylated lignin
Jacopo Bernardini a, Patrizia Cinelli b, Irene Anguillesi a, Maria-Beatrice Coltelli a,
Andrea Lazzeri a,⇑
a
Department of Civil and Industrial Engineering, University of Pisa, Largo L. Lazzarino 1, 56126 Pisa, Italy
b
National Interuniversity Consortium of Materials Science and Technology (INSTM), Via G. Giusti, 9, 50121 Firenze, Italy

a r t i c l e i n f o a b s t r a c t

Article history: An innovative and green chemistry synthetic approach was developed in order to employ
Received 16 July 2014 lignin for the production of flexible polyurethane foams. Soda lignin and oxypropylated
Received in revised form 24 November 2014 soda lignin were tested and compared. Glycerol and PEG 400 were used as polyol fractions
Accepted 26 November 2014
for lignin liquefaction by microwave irradiation, which represents a novel green processing
Available online 3 December 2014
technique. The samples were produced with the ‘‘one-shot’’ technique, using two types of
chain extenders in combination with liquefied lignin: castor oil and polypropylene glycol
Keywords:
triol. Water was used as a single natural blowing agent, and polymeric diphenylmethane
Polyurethane
Flexible foam
diisocyanate (PMDI) was employed as the isocyanate fraction. The work was carried out
Lignin keeping the NCO/OH less than one hundred, thus enhancing the flexibility due to a lower
Oxypropylation crosslinking degree; all the foams were produced in free and controlled expansion. Two of
Microwave the most efficient chain extenders were individuated thus introducing flexible chains into
Renewable sources the macromolecular structure that can reduce the glass transition temperature of the
materials and therefore generate foams with higher flexibility. The properties of the
produced foams were compatible with the technical requirements for applications in
packaging and for the production of the interior part of car seats.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction in many industries such as footwear, construction,

Polyurethanes (PUs) are an important class of polymers
synthesized by means of a polyaddition reaction between
polyols, such as polyhydric alcohols and polyisocyanate
that form urethanic linkages. These polymers are very ver-
satile because they present structures ranging from rigid to
flexible and are commonly used in various forms such as
foams, elastomers, adhesives and many others, therefore
they have a wide range of applications, i.e. automotive
seating, furniture, and packaging and medical devices.
Polyurethanes are also widely used as thermal insulators
⇑ Corresponding author. Tel.: +39 050 2217807; fax: +39 050 2217866.
E-mail address: andrea.lazzeri@unipi.it (A. Lazzeri).
electronics and many others. Actually, common polyols
employed to produce polyurethane foams are petrochemi-
cal in origin and, due to the escalation of crude oil prices,
the cost of polyurethane reactants steadily increased, thus
sparking interest in the quest for renewable sources.
Flexible PU foams are blocks of copolymers whose elas-
tic properties depend on the phase separation between
hard and soft domains. Hard blocks are rigid structures
that are physically cross-linked and give the polymer its
firmness; soft blocks are stretchable chains that give the
polymer its elasticity. Therefore, polyurethane foams can
be customized by adapting the composition and the ratio
of these blocks.
Polyurethanes can be synthesized by using renewable
sources as polyols, such as vegetable oils, by replacing

http://dx.doi.org/10.1016/j.eurpolymj.2014.11.039
0014-3057/Ó 2014 Elsevier Ltd. All rights reserved.
148 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
petrochemical polyols partially or totally. Since the 1960s,
a wide range of vegetable oils have been considered for the
preparation of polyurethane; the most important ones are
highly unsaturated, such as sunflower, palm, rapeseed and
mainly castor and soybean oils [1–3].
Several attempts were made in order to use lignin as a
novel and appropriate renewable feedstock, thus supersed-
ing petrochemical phenols and overcoming the idea that it
is a waste and a low value by-product of wood pulping
[4,5]. The use of lignin as such, or after appropriate treat-
ments, was extensively investigated to prepare polyesters
and polyurethanes exploiting its phenolic and aliphatic
hydroxyl groups [6,7].
Lignocellulosic biomass is composed mainly of three
chemicals: cellulose, hemicellulose and lignin, and its com-
position varies with the source and type of biomass. Lignin
is an aromatic complex polymer that is highly branched
and based on subunits of alcohol 4-hydroxycinnamyl (p-
coumaryl alcohol) having methoxylated derivatives in 3-
(coniferyl alcohol) and 3,5-positions (sinapyl alcohol) [8].
Lignin has a molecular weight of about 3000–7000 Da, an
OH number of about 1000–1500 mg KOH/g, and a methoxy
group content of about 13–14%.
There are many different types of lignin, whose nature
depends on the different preliminary treatments that they
have undergone and on the different types of wood [9]. Lig-
nin is mainly isolated from wood pulping and papermaking
operations, but only a small amount (1–2%) is recovered
and employed as feedstock to produce other chemicals
[10,11]. These pulping treatments provide fragmented lig-
nin, thus enhancing its reactivity due to lower molecular
weights and higher solubility in organic solvents. Soda lig-
nins are sulphur-free and for this reason are considered as
being closer to natural lignins; these lignins are mainly
obtained from non-woody plants.
Many studies were carried out over recent years in
order to use various lignins in polymeric synthesis. In this
sense, Hatakeyama’s group dissolved several lignins, such
as Kraft, organosolv and sulphonate, into oligoether diols
to obtain the polyolic mixture. By using the macrodiol as
both the solvent and a comonomer, they improved the
reactivity of lignin OH groups [12], and the introduction
of flexible oligoether as spacers within the polymer chain
reduced the network stiffness of the isocyanate. Also
semi-rigid polyurethane foams, developed for housing
insulation, were prepared by means of using lignin polyolic
molasses, by controlling the apparent density, and by
changing the mixing ratio between lignin and polyethylene
glycol [13].
Oxypropylation of biomass was widely studied and is
presently accepted as a method for improving the func-
tionality of lignin [14–16]. A schematic pathway is
reported in Fig. 1.
Glasser and his co-workers have modified lignins by
using different conditions of high temperatures and pres-
sures [17], allowing for further investigations on it in
recent years [18].
The oxypropylation process allows to free the phenolic
groups of lignin from steric hindrances and to obtain a lig-
nin-based liquid polyol with lower glass transition temper-
ature compared to the starting powder. It is a green
Fig. 1. Schematic pathway for the oxypropylation process of lignin.
process because it does not need further steps involving
solvents and other by-product removal, nor is there the
need for separation or purification steps in order to recover
the final polyol mixture [19]. In this sense, the purpose of
oxypropylation is to increase the lignin functionality for
synthesizing polyurethane foams [20].
Many studies of oxypropylation were also carried out
on renewable sources in order to produce low-cost natural
polyols. Several biomasses such as cork [21], chitin and
chitosan [19], and sugar beet pulp [22] were considered.
Currently, commercial lignins are not used in the indus-
trial production of polyurethane foams due to the high vis-
cosity of lignin-based polyols combined with a high
temperature and very long reaction times; the extended
structure of the lignin macromolecules must be frag-
mented before the employment in polyurethanes prepara-
tion. This process was defined liquefaction as the solubility
of lignin was modified thanks to the occurring of fragmen-
tation. Usually the reported methods consist in the use of
proper solvent and heating. Microwave treatment can be
an effective alternative because it is able to heat the core
of the system. The microwaves were carefully studied to
degrade hazardous compounds [23] and subsequently this
treatment was tried on lignocellulosic biomasses [24–26]
in order to obtain liquid polyols with shorter reaction
times. This approach was tried on commercial lignin and
the final mixture contained a higher amount of phenolic
OH groups and a lower amount of methoxy groups, thus
improving the system reactivity [27].
Microwaves are able to penetrate and heat the present
reaction medium, thereby reducing reaction times from
hours to minutes; furthermore, when heating with micro-
waves there is a faster increase in temperature and a lower
thermal degradation of reagents than traditional heating
systems, thus being compatible with a green chemistry
approach.
In previous research [28], the authors produced poly-
urethane soft foams from kraft lignin, achieving a promis-
ing quality of foams in terms of density and mechanical
properties. It resulted interesting to investigate lignin pro-
duced using different processes and modified lignin in
order to achieve optimized properties. Thus the present
work was focused on the production of flexible polyure-
thane foams from commercial soda lignin and commercial
oxypropylated soda lignin by adopting a green synthetic
pathway. Two chain extenders were used to produce flex-
ible foams in combination with liquefied lignin: polypro-
pylene glycol triol (PPG triol) and castor oil; polymeric
diphenylmethane diisocyanate (PMDI) was employed as
 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
isocyanate fraction. Cellular structure as well as physical
and thermo-mechanical properties were characterized in
order to compare the properties of the foams produced
with the requirements set for commercial standard poly-
urethane foams. (This research activity was performed in
the EC Project FORBIOPLAST Grant agreement No.
212239, which focused on the use and valorisation of for-
est resources that include lignin as a by-product of the
wood industries and bioethanol production.)
2. Materials and method
The lignin used was Protobind™ 1000, produced as dry
powder by ALM India Pvt. Ltd. using a patented GreenValue
SA technology; this lignin comes from soda pulping of
non-woody biomass. Protobind™ 1000 has high degree of
purity (about 98%) and is free of hemicellulose. Oxypropy-
lation reaction was performed in aqueous alkali at room
temperature at moderate pH (only the phenolic OH groups
are ionized) (Fig. 1).
Polyethylene glycol 400 and glycerol from Aldrich were
used as solvents to liquefy lignin. Polypropylene glycol
triol (PPG triol) and Castor oil [44] (CO) (Fig. 2), were used
as chain extenders because of their flexible molecular
structure, thus suitable for flexible foams. They were both
purchased from Sigma Chemicals and used as received.
PPG triol from Sigma Chemicals has a Molecular weight
of 4800, functionality of 2.6, and viscosity of 850 mPa s at
25 °C. ISO 116/1 was used as an isocyanate, a polymeric
diphenylmethane diisocyanate (PMDI) with 25.7% part by
weight of NCO content, provided by BASF Poliuretani Italia
Spa. The isocyanate was added in excess to the OH groups
on the polyols in order to react with distilled water, thus
forming carbon dioxide, which acted as the foam-blowing
agent. The catalysts used for the production of foams can
be divided into gelling and blowing catalysts. Both of these
must be kept at a proper ratio in order to obtain the desired
products. They were kindly provided by Air Products and
belong to two different series: classic catalysts and new
generation catalysts (NE), with a low emission of amine.
The classic catalysts consisted of non-reactive amines that
can be released from the foams once they are produced.
They were DABCO 33LV (triethylenediamine 67% by
weight, dipropylene glycol 33% by weight, gelling agent)
Fig. 2. Structure of chain extenders.
149
and DABCO BL11 (bisdimethylamminoethylether 70% by
weight, dipropylene glycol 30% by weight, blowing agent).
The NE catalysts used were DABCO NE1070 (3-dimethyl-
aminopropyl urea, gelling agent) and DABCO NE300
(N-[2-[2(dimethylamino) ethoxy]ethyl]-N-methyl-1,3-pro-
panediamine, blowing agent). These catalysts can reduce
emissions because the molecules that act as catalysts can
chemically bind themselves to the polyurethane matrix.
DABCO DC2525, a silicone surfactant made from 70% by
weight of polysiloxane, was used to ensure the uniformity
of structure and enhance cell opening.
2.1. Liquefaction of lignin
Lignin samples were previously dried for 24 h in an
oven with air circulation at 80 °C. Then lignin was mixed
with liquefying polyols (glycerol and PEG 400) at a
predetermined weight ratio, placed in a Teflon vessel,
and subjected to liquefaction treatment in a multimode
microwave oven CEM MDS 2100. A power of about
200 W was fixed to reach 135 °C in 2 min.
2.2. Foam preparation
The liquefied mixture was mixed with the other polyol-
ic components and with PMDI; then the resulting mixture
was deposited into a mould (85  85  55 mm) and
allowed to rise in controlled expansion by closing the
mould thus forcing the foam to take on its shape. Foams
were also produced in free expansion in order to compare
their mechanical properties depending on the type of
expansion. Compositions used to synthesize the foams
are reported in Table 1.
All the foam samples were allowed to cure at ambient
conditions for a minimum of 2 days. Cream times and rise
times are reported in Table 2.
2.3. Methods
The hydroxyl number in the liquefied mixture was
determined by following the ASTM D2849-69 norm: 1 g
mixture and 25 ml of phthalate reagent were heated for
20 min at 110 °C. Then, 50 ml of pure 1.4-dioxane and
25 ml of distilled water was added and the resulting
150 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156

Table 1
Formulations (grams) used for synthesis of PUs foams with polypropylene glycol triol (PPG) or castor oil (CO) as chain extenders.

PPG-A PPG-B PPG-C PPG-D PPG-E PPG-F PPG-G CO-A CO-B

Lignin/glycerol/PEG400 1/0.4/2 1/0.4/2 1/0.2/1 1/0.2/1 1/0.2/1 1/0.2/1 1/0.2/1 1/0.4/2 1/0.4/2
Liquefied lignin 14.2 11.4 10.0 10.0 10.0 10.0 10.0 17.0 17.0
PPG triol 20.8 16.6 14.0 10.0 7.6 14.0 10.0 – –
Castor oil – – – – – – – 27.5 27.5
Water 0.6 0.7 0.66 0.66 0.66 0.65 0.65 0.81 0.5
DABCO33LV (gelling) 0.53 – 0.36 0.34 0.31 – – 0.67 –
DABCOBL11 (blowing) 0.53 – 0.35 0.31 0.31 – – 0.67 –
DABCONE1070 (gelling) – 0.9 – – – 0.63 0.6 – 1.24
DABCONE300 (blowing) – 0.35 – – – 0.2 0.22 – 0.25
Surfactant 0.35 0.56 0.26 0.21 0.17 0.21 0.2 0.45 0.66
Isocyanate 25.0 23.0 20.0 16.8 15.0 20.0 16.9 36.8 32.0
Lignin (% by weight) 6.7 6.7 10 11.9 13.3 10.0 11.8 6.0 6.3

Table 2
Cream time and rise time for foams prepared starting from lignin Protobind (P) and oxypropylated lignin (HP).

Sample Cream time (s) Rise time (s) Sample Cream time (s) Rise time (s)
P-PPG-A 10 15 HP-PPG-A 10 10
P-PPG-B 20 40 HP-PPG-B 20 50
P-PPG-C 30 80 HP-PPG-C 25 40
P-PPG-D 50 170 HP-PPG-D 30 50
P-PPG-E 40 180 HP-PPG-E 30 60
P-PPG-F 40 150 HP-PPG-F 40 110
P-PPG-G 50 180 HP-PPG-G 50 150
P-CO-A 30 45 HP-CO-A 25 40
P-CO-B 70 190 HP-CO-B 60 120

mixture was titrated with a 1 M sodium hydroxide solu-
tion up to the equivalence point. The phthalate reagent
consists of a mixture of 150 g phthalic anhydride, 24.2 g
imidazole and 1000 g dioxane. The hydroxyl number in
mg KOH/g of the sample was calculated by the following
Eq. (1):
ðB  AÞ  N  56:1
Hydroxyl number ¼
W
where A is the volume in ml of the sodium hydroxide solu-
tion required for titration of liquefied lignin sample, B is
the volume of blank solution (ml), N is the normality of
the sodium hydroxide solution, and W is the weight of liq-
uefied lignin (g).
The isocyanate/hydroxyl molar ratio required for the
reaction is calculated using the following Eq. (2):
M NCO W NCO
NCO=OH ¼ h P3 2 i  100
M OH W OH þ i¼1 M Adi W Adi þ 18 W H2 O
where M NCO is the number of isocyanate groups in one
gram of isocyanate, W NCO is the weight of isocyanate (g),
M OH is the number of hydroxyl groups contained in one
gram of polyol (mixture of liquefied lignin and chain
extender, index 1), W OH is the weight of polyols (g), M Adi
is the number of hydroxyl groups in one gram of additives
(blowing catalyst, index 1, gelling catalyst, index 2, and
surfactant, index 3), W Adi is the weight of additives (g)
and W H2 O is the weight of water.
Thermogravimetric analysis (TGA) measurements were
carried out following the ASTM standard procedure D
3850-94. A Rheometric Scientific instrument was used,
and the experiments were performed under a nitrogen
gas atmosphere. The samples were ground into a fine pow-
der prior to measurement and were heated from room
temperature to 900 °C at a rate of 10 °C/min.
For dynamic mechanical analysis (DMTA), foams were
cut into 20  20  20 mm cubes and tested under com-
pression mode between two serrated parallel plates with
ð1Þ a 25 mm diameter (Gabo Eplexor 100 N). Storage modulus
(E0 ) and tan d were recorded at a frequency of 1 Hz over the
temperature range from 150 to 120 °C. The temperature
ramp rate was 2 °C/min.
The density and 50% compression force deflection
(CFDV) of flexible polyurethane foams were measured fol-
lowing the ASTM D 3574-05 norm, which describes the
standard test methods for flexible cellular materials
including slab, bonded and moulded urethane foams. The
sample size was 30 mm  30 mm  30 mm and the result
ð2Þ of ratio between weight (M) and volume (V) of the speci-
mens was the apparent density, which should be in kg/
m3. The compression force deflection test was performed
using Instron 1185 equipment, according to the ASTM D
3574-05 norm: pre-flex the specimen twice up to 75%
and 80% of its original thickness at 240 mm/min; rest for
6 min and measure sample thickness; bring the compres-
sion plates into contact with the specimen and apply a con-
tact load of 140 Pa; compress the specimen up to 50% of its
thickness at 50 mm/min and observe the final load after
60 s. At least three replicates were tested for each sample.
Compression strength was evaluated as the ratio of the
final load at the end of the procedure (N) and the cross
section area of the specimen (mm2).
 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
The FT-IR characterization was performed with a
Thermo Scientific NicoletTM 380; the foams were previ-
ously dried at 105 °C for 4 h. The samples for FT-IR mea-
surement were taken from the centre of the foam buns,
ground and dried at 105 °C for 2 h before analysis to ensure
that no water was adsorbed inside.
A Scanning Electron Microscope (Jeol JSM5600LV) was
used to study the microstructure of the foams and gold
sputtering was applied using an Edwards Sputter Coater
prior to investigation.
3. Results and discussion
The synthesis of polyurethane foams requires polyolic
reagents with low viscosity and high functionality, there-
fore solid powder lignin has been liquefied with the aid
of suitable reagents, such as glycerol and PEG 400, as
reported in literature for Kraft lignin [28]. Liquefied lignin
is rich in free hydroxyl group, but it has high viscosity and
hydroxyl values. For these reasons, it was not suitable to
produce flexible foams. Accordingly, the addition of a chain
extender was necessary in order to decrease the viscosity
of the polyolic mixture and at the same time to increase
the flexibility of the final material. Polypropylene glycol
triol (PPG triol) and castor oil (CO) were used as chain
extenders in this work.
The liquefaction process was carried out by microwave
irradiation and was optimized by increasing the amount of
solvents used to reduce the mixture viscosity.
One of the most important parameters in polyurethane
production is the hydroxyl value, which is generally deter-
mined following the ASTM D 2849-69 norm; the procedure
is based on an esterification between polyol and an excess
of acetic or phthalic anhydride, the hydrolysis of unreacted
anhydride which forms dicarboxylic acid. The resulting
solution is then titrated with sodium hydroxide reducing
the reaction time by using imidazole as an esterification
catalyst [29].
The NCO/OH was fixed once the OH number was deter-
mined, thus setting the weight of isocyanate to be added in
each formulation. A NCO/OH was fixed at less than 100 in
order to maintain a low cross-linking degree in the final
foam, thus giving higher flexibility to the material [30].
The polyurethane foams were produced by adopting the
one-shot technique, which provided a very efficient mixing
among all the materials involved (polyols, chain extender,
silicon emulsifier, blowing agent, catalysts, isocyanate) in a
single step and in a short time. It was possible to use only
two components: one was the polyolic containing all
material except for the isocyanate, which was the second
component. The resulting polyurethane was therefore a
consequence of the very efficient contact between the
two components.
The kinetics of foam formation was monitored by
means of measuring cream and rise times. Cream time is
defined as the time required by the polyurethane mixture
to change from a clear colour to a creamy one (although
liquefied lignin has a very dark colour, it was possible to
recognize a change in colour when cream time was
reached), while the rise one is the time it takes the foam
151
to entirely expand. All data are reported in Table 2. These
two parameters are very important to control the kinetic
process because the reaction between water and isocya-
nate produces carbamic acid, which is unstable and
decomposes to carbon dioxide and primary amine. Carbon
dioxide then diffuses in the mixture thus expanding the
foam structure. Clearly, the nucleation sites in the reacting
mixture are checked by controlling the amount of carbon
dioxide released. Foams produced using classic catalysts
showed reaction times slightly lower than samples pro-
duced using NE catalysts; this was probably due to the
higher reactivity of classic catalysts. Nevertheless, the
properties of the produced foams showed no relevant dif-
ferences regardless of the kind of catalyst employed.
3.1. FT-IR spectroscopy
Polyurethane foams were qualitatively characterized by
the FT-IR technique in order to assess the presence of unre-
acted isocyanate within the final material and to appraise
the formation of typical urethanic bonds. Fig. 3 shows
the FT-IR spectra of a representative sample of each type
of produced foam.
The above spectra qualitatively confirmed the presence
of urethane linkages because they presented approxi-
mately the same trend regardless of the different composi-
tions, type of lignin and type of chain extender. It was
possible to find the typical urethanic absorption bands:
the ANH vibrational stretching (3500–3200 cm1), the
characteristic vibration of AC@O groups (1730–
1640 cm1) [31] and, eventually, the very weak peak of
unreacted ANCO group at about 2270 cm1 [32].
Particularly, two absorption bands can be observed, one
centred at 1710 cm1 and the other one at 1640 cm1. The
first peak at about 1710 cm1 was caused by the presence
of free urea, meanwhile the second one, at about
1640 cm1, was characteristic of bidentate urea that inter-
acted through hydrogen bonds with the surrounding mol-
ecules. These two signals were indicative of a phase
Fig. 3. Comparison among FT-IR spectra on the polyurethane foams
obtained starting from Protobind 1000 lignin (P) or oxypropylated
Protobind 1000 lignin (HP) and by using polypropylene glycol triol
(PPG) or castor oil (CO) as a chain extender.
152 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
separation within the material [33]. The hydrogen-bonded
urea, including both monodentate and bidentate, were an
indication of hard domain inside the foam structure. The
bands in the region below 1600 cm1 represented the fin-
gerprints of the polyurethanes.
The above spectra did not present significant differ-
ences among the several type of foams, thus allowing stat-
ing that the preliminary step of oxypropylation and the
type of chain extender did not influence the chemical
behaviour of the foams. Comparing this work with the pre-
vious one based on Kraft lignin [28], the intensity of unre-
acted isocyanate band was lower, probably due to a lower
steric hindrance and to a simpler structure of the soda lig-
nin with respect to the Kraft one. Despite the fixed NCO/OH
was less than 100, a small amount of unreacted isocyanate
groups was found, which was caused by the steric hin-
drance of the lignin structure that prevented the complete
reaction between the isocyanate and hydroxyl groups.
3.2. Thermogravimetric analysis
Thermogravimetric analysis is a very useful technique
to analyze the degradation steps of polyurethane materials
based on renewable sources [34]. Fig. 4 displays the
Fig. 4. TG (top) and DTG (bottom) curves of the produced polyurethane
foams. All thermograms were acquired at constant heating rate
(10 °C min1) under air atmosphere.
thermogravimetric curves and the derivative curves of
the studied foams.
Foams produced by using PPG triol as a chain extender
presented the beginning of irreversible degradation at
around 240 °C and a first step at 270 °C, which were typi-
cally characteristic of urethane bond; two degradation
steps can also be observed at 389 °C and at 480 °C, which
were correlated to the presence of two phases within the
material [28]. Indeed the first is related to soft segments
(PPGtriol chains), and the second is related to the degrada-
tion of hard segments. Also the foams produced by using
castor oil as a chain extender started thermal degradation
at about 240 °C, thus indicating that the onset of thermal
degradation did not change regardless of the employed
chain extender. However the derivative trend is quite dif-
ferent comparing the systems obtained by using the two
chain extenders. The derivative trend of castor oil contain-
ing foams [35], as reported also by Narine [1], shows three
main typical peaks, whereas the trend observed for PPG
triol is a simpler trend, with a main peak at 389 °C. This dif-
ference can be ascribed to the different structure of the two
chain extenders. In fact in PPG triol the thermal degrada-
tion occurs by the scission of CAO bonds of the ether
chains, whereas the thermal degradation of CO is more
complex, involving both CAO and CAC linkages. Moreover
in the CO molecules the presence of double bonds make
the structure weaker with respect to thermal degradation
[43], thus resulting in a consistent mass loss at 323 °C.
The degradation of pure not modified lignin (not shown
data) occurs in the samples in two main steps at 285 ad
410 °C. Hence, as the content of lignin is, in the different
formulations, below 10% by weight, its mass loss cannot
be easily distinguished in the TGA trends. Foams produced
with PPG triol presented a higher amount of carbon resi-
dues, about 15% by weight, than those produced with cas-
tor oil (less than 10%); this is probably due to the different
cellular structure between the two types of foams. In this
sense, PPG triol-based foams presented more regular cellu-
lar structure, hindering the regular heat transfer and there-
fore resulting in a less efficient pyrolysis; conversely,
castor oil-based foams presented a less regular cellular
structure, and this allowed for a better pyrolysis process.
The results confirm that, as observed in FT-IR analysis,
the oxypropylation step did not change the thermo-chem-
ical behaviour of polyurethane foams produced from
lignin.
3.3. Dynamic mechanical thermal analysis
Fig. 5 shows the storage modulus profile (E0 ) and tan d,
as a function of temperature, of foams produced starting
from Protobind 1000 or oxypropylated Protobind 1000 by
using PPG triol or castor oil as a chain extender. The Tg
value of each sample was determined by the tan d peaks.
Two different transitions are shown in the modulus
profiles: the first is related to the flexible phase in the
foams, and the second to the hard one. Clearly both
domains have different glass transition temperatures and
mechanical stiffness: the soft phases (or polyol-rich
domains), which usually show low Tg ranging between
70 °C and 50 °C, give viscoelastic properties to the
 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156 153

Fig. 5. DMTA results of the produced polyurethane foams starting from

lignin (P) or oxypropylated (HP) lignin by using PPG or castor oil (CO) as
Fig. 6. Compression force deflection value (CFDV) for samples produced
chain extenders.
with PPG triol as a chain extender.

foams thus allowing a higher energy absorption. Con-
versely, the hard phases (or polyurea-rich domains) pres-
ent a glass transition temperature over 100 °C and impart
to the polyurethane material a higher modulus and greater
thermal stability [28]. The first Tg is related to the soft seg-
ments and similar values mean a similar phase separation
between hard and soft domains [36]. Fig. 5 shows no differ-
ences between foams produced by using the same chain
extender, but it shows significant differences when the
chain extender is changed. PPG-based foams presented a
plateau at about 0 °C, which is due to the hard segments,
precisely to the physical crosslinks inside the network;
their breaking caused the end of the plateau, thus decreas-
ing the storage modulus [37].
Tan d curves, as well as modulus profiles, showed no dif-
ferences among foams based on the same chain extender;
samples produced by employing PPG triol presented a Tg at
about 50 °C but a different intensity of the related peak.
In this sense, foams made starting from oxypropylated lig-
nin showed a larger peak, which means a greater amount
of soft domains in the chain network, thus imparting
greater flexibility to the material.
Different considerations were made about castor oil-
based foams in comparison with PPG triol-based foams.
They showed a Tg at about 0 °C and presented peaks with
broader shapes; this greater amplitude signifies a wider
distribution of crosslinking density and, mainly, a lower
structural homogeneity.
A wide peak was observed at about 100 °C for the
foams produced starting from oxypropylated lignin and
castor oil, which was likely due to the motion of dangling
chains [38]. Therefore, this transition was assigned to a
segmental relaxation of a second soft phase, which was
caused by the compatibility between castor oil and oxy-
propylated lignin [39].
3.4. Mechanical properties and morphology
Table 3 reports the apparent density of all the produced
foams.
Foams of series 1 were produced in free expansion,
while those of series 2 were made in controlled expansion.
Table 3 shows that the densities of PPG-based samples
were much lower than those based on castor oil as a chain
extender, which has probably a less regular structure. A
similar difference was observed by using Kraft lignin [28].
The compression force deflection test measures the
force necessary to produce a 50% compression over the
entire top surface area of the foam specimen. It was deter-
mined by following the ASTM norm D 3850-94 and repre-
sents a parameter that increases with the foam stiffness
and depends on the lignin content in the final material.

Table 3
Apparent density of polyurethane foams prepared in free (1) and controlled expansion (2).

Sample Apparent density (kg/m3) Sample Apparent density (kg/m3)

1 2 1 2
P-PPG-A 117.8 113.4 HP-PPG-A 101.6 124.8
P-PPG-B 107.9 147.9 HP-PPG-B 86.6 132
P-PPG-C 87.1 95.2 HP-PPG-C 72.1 99.9
P-PPG-D 80.0 103.7 HP-PPG-D 76.6 100.2
P-PPG-E 83.7 102.5 HP-PPG-E 78.8 93.4
P-PPG-F 69.8 105.0 HP-PPG-F 77.5 105.4
P-PPG-G 81.9 100.5 HP-PPG-G 85.9 96.3
P-CO-A 129.7 147.0 HP-CO-A 127.4 180.9
P-CO-B 145.1 178.7 HP-CO-B 151.1 207.8
154 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
Fig. 7. Compression force deflection value (CFDV) for samples produced
with castor oil as a chain extender.
Higher values of compression strength indicate that foam
is sturdier, which depends on both higher crosslinking
density and greater apparent density. Fig. 6 reports a direct
correlation between CFDV and apparent density for a sam-
ple produced with PPG triol.
This linear increase was in agreement with our above-
cited previous work: with the aim to increase the apparent
density of the foam, it was possible to raise the NCO/OH,
thus giving the foam a higher reticulation degree and
therefore increasing the compression strength value. All
the foams produced with PPG triol as a chain extender,
after the preflex load and recover time of 1 min, showed
a small decrease of thickness, less than 2% of its initial
Fig. 8. SEM micrographs (enlargement 50) of all investigated foams, produced either by changing the lignin or varying the chain extender.
thickness. The value of compression force deflection
increased linearly as the apparent density increased.
Fig. 7 reports the values of compression force versus
apparent density for foams produced by using castor oil
as a chain extender.
As reported above, the castor oil-based foams presented
a higher density when compared to PPG-based ones, and in
this case, as reported in Fig. 7, the density and the com-
pression force values cannot be directly correlated because
the structure of the foam was less regular. Thinner cell
walls and larger foam cells caused the decrease of com-
pressive strength even if the apparent density was higher
than in the PPG-based samples.
Taking into account the nature of lignin, both of the
above figures showed similar behaviour for each kind of
starting material; this means that there was no influence
on the trends of compression strength, whatever the
employed chain extender. Comparing an equal formula-
tion, the foams produced starting from oxypropylated lig-
nin presented a lower compression value than those
based on commercial lignin; this was due to the previous
reaction step that had degraded the structure of lignin
making it more fragmented and therefore lighter.
Low functionality, low hydroxyl number and secondary
hydroxyl groups of castor oil led to a low reactivity when
directly used to produce flexible polyurethane foams. By
mixing CO with polyols such as glycerol and PEG 400, a
polyol mixture with higher hydroxyl number was
obtained, which leads to less flexible foams. In the present
work, replacing PPG triol with CO as a chain extender
resulted in foams with a higher apparent density and,
mainly, apparent density and compression values cannot
be directly correlated, as reported above.
 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
The presence of an open cell structure, which is typical
of flexible foams, was assessed by Scanning Electron
Micrographs (SEM). This aspect is very important since
the mechanical properties of the foams are primarily influ-
enced by the apparent density, thickness and sizes of the
cells [40].
Fig. 8 shows SEM micrographs of foams produced by
using PPG triol or Castor Oil as chain extender and lignin
or oxypropylated lignin as fraction of polyol.
The cell shape of the foam was polyhedral and showed
preferential orientation associated with the polyurethane
rise direction. As reported in literature, the precipitation
of the urea phase and the formation of the urea aggregates
can be directly related to the cell-opening event in flexible
PU foams [41].
Regarding the foams produced by using PPG triol, they
showed more regular and thinner cells than those pro-
duced starting from CO; also the cell shapes were more
regular. To confirm these statements, CO-based foams pre-
sented also a higher number of untouched intercellular
membranes, which were not suitable for flexible behav-
iour. Foams based on castor oil showed coarse cells being
large in size, contributing to a less homogeneous structure;
this heterogeneity led to a lower compressive modulus and
higher apparent densities, compared to the more regular
samples produced with PPG triol.
Considering the effects of lignin treatment, the foams
manufactured starting from oxypropylated lignin exhib-
ited anisotropic cellular structures with elongated cells
along the blow direction, as expected [42,43]. Samples pro-
duced by using oxypropylated lignin and CO presented a
regular structure, despite the cells were less regular than
the samples based on PPG triol.
4. Conclusions
Starting from the our previous work, this article reports
the successful green and innovative synthesis of flexible
polyurethane foams starting from Protobind lignin and
oxypropylated Protobind lignin, by following an environ-
mentally-friendly process based on the use of microwave
irradiation in order to liquefy the lignin, and on the use
of water as a blowing agent. Glycerol and PEG 400 were
used as liquefaction solvents, while polypropylene glycol
triol or castor oil were used as chain extenders in order
to produce flexible foams in controlled expansion, thus tai-
loring the properties of the final foams. All samples were
prepared using a one-shot approach maintaining the
NCO/OH ratio less than one hundred, thus reducing signif-
icantly the degree of crosslinking and giving higher flexi-
bility to the material.
Considering the characterizations carried out, foams
based on PPG triol showed a lower Tg of the soft phases,
clearly separated from the hard ones, while samples based
on castor oil as a chain extender presented both phases
mixed. The amount of lignin in the formulation ranged from
6% to 13% and considering castor oil as renewable source,
the amount of green raw materials can be more than 45%.
The kind of lignin did not greatly influence the structure
and properties of the produced foams, although oxypropy-
lated lignin was dissolved more easily in liquefaction sol-
155
vents than commercial soda lignin, and, in addition, the
miscibility between liquefied lignin and chain extenders
(in particular with castor oil).
All the polyurethane foams showed a high content of
open cells and their thermo-mechanical properties were
suitable for fillings and packaging applications, such as fill-
ing for the interior parts of car seats and packaging of fur-
niture. In this sense, foams produced by using PPG triol as a
chain extender had an apparent density ranging from
about 70 to 150 kg/m3 and a compression force deflection
value from about 2  102 to 7  102 MPa, while foams
based on castor oil had an apparent density ranging from
about 130 to 210 kg/m3, and a compression strength rang-
ing from about 7  103 to 3.5  102 MPa.
Starting from soda lignin or soda oxypropylated lignin
could be considered a good alternative way to produce
flexible polyurethane foams based on renewable sources,
in agreement with previous work based on Kraft lignin.
Acknowledgements
The authors wish to acknowledge the support from FP7
– KBBE project n° 212239 Forbioplast (Forest Resource Sus-
tainability through Bio-Based Composite Development) to
carry out this research.
References
[1] Narine SS, Kong X, Bouzidi L. Physical properties of polyurethanes
produced from polyols from seed oils. II. Foams. J Am Oil Chem Soc
2007;84(1):65–72.
[2] Petrovic ZS. Polyurethanes from vegetable oils. Polym Rev
2008;48(1):109–55.
[3] Sharma C, Kumar S, Raman Unni A, Aswal VD, Rath SK, Harukrishnan
G. Foam stability and polymer phase morphology of flexible
polyurethane foams synthesized from castor oil. J Appl Polym Sci
2014:40668–75.
[4] Stewart D. Lignin as base material for materials applications:
chemistry, application and economics. Ind Crops Prod
2008;27(2):202–7.
[5] Serrano L, Briones R, Melus A, Herseczk Z, Labidi J. Polyols from the
lignocellulosic waste of biodiesel production process. Chem Eng
Trans 2010;21(1):1339–44.
[6] Belgacem MN, Gandini A. Monomers, polymers and composites from
renewable resources. Oxford: Elsevier; 2008.
[7] Li Y, Ragauskas AJ. Kraft lignin-based rigid polyurethane foam. J
Wood Chem Technol 2012;32(3):210–24.
[8] Crestini C, Melone F, Sette M, Saladino R. Milled wood lignin: a linear
oligomer. Biomacromolecules 2011;12(11):3928–35.
[9] Lange H, Decina S, Crestini C. Oxidative upgrade of lignin – recent
routes reviewed. Eur Polym J 2013;49(6).
[10] Bonini C, D’Auria M, Emanuele L, Ferri R, Pucciariello R, Sabia AR.
Polyurethanes and polyesters from lignin. J App Polym Sci
2005;98(3):1451–6.
[11] Cateto CA, Barreiro MF, Rodrigues AE, Brochier-Salon MC,
Thielemans W, Belgacem MN. Lignins as macromonomers for
polyurethane synthesis: a comparative study on hydroxyl group
determination. J App Polym Sci 2008;109(5):3008–17.
[12] Hatakeyama H. Polyurethanes containing lignin, chemical
modification, properties and usage of lignin. New York: Kluver;
2002.
[13] Hatakeyama H, Kosugi R, Hatakeyama T. Thermal properties of
lignin- and molasses-based polyurethane foams. J Therm Anal
Calorim 2008;92(2):419–24.
[14] Glasser WG, Wu LCF, Selin JF. Synthesis, structure, and some
properties of hydroxypropyl lignins. In: Wood and agricultural
residues – research on use for feed, fuels and chemicals. New York:
Academic Press; 1983. p. 149–66.
[15] Crestini C, Crucianelli M, Orlandi M, Saladino R. Oxidative strategies
in lignin chemistry: a new environmental friendly approach for the
156 J. Bernardini et al. / European Polymer Journal 64 (2015) 147–156
functionalisation of lignin and lignocellulosic fibers. Catal Today
2010;156(1–2):8–22.
[16] Aniceto JP, Portugal I, Silva CM. Biomass-based polyols through
oxypropylation reaction. ChemSusChem 2012;5(8):1358–68.
[17] Wu LC, Glasser W. Engineering plastics from lignin. I. Synthesis of
hydroxypropyl lignin. J Appl Polym Sci 1984;29(4):1111–23.
[18] Cateto CA, Barreiro MF, Rodrigues AE, Belgacem MN. Optimization
study of lignin oxypropylation in view of the preparation of
polyurethane rigid foams. Ind Eng Chem Res 2009;48(5):2583–9.
[19] Fernandes S, Freire CS, Pascoal Neto C, Gandini A. The bulk
oxypropylation of chitin and chitosan and the characterization of
the ensuing polyols. Green Chem 2008;10(1):93.
[20] Laurichesse S, Avérous L. Chemical modification of lignins: towards
biobased polymers. Prog Polym Sci 2014;39:1266–90.
[21] Evtiouguina M, Barros AM, Cruz-Pinto JJ, Neto CP, Belgacem N, Pavier
C, et al. The oxypropylation of cork residues: preliminary studies.
Bioresour Technol 2000;73(2):187–9.
[22] Pavier C, Gandini A. Oxypropylation of sugar beet pulp. 1.
Optimization of the reaction. Ind Crops Prod 2000;12(1):1–8.
[23] Zhang L, Guo X, Yan F, Su M, Li Y. Study of degradation behaviour of
dimethoate under microwave irradiation. J Hazard Mater
2007;149(3):675–9.
[24] Krzan A, Kunaver M. Microwave heating in wood liquefaction. J App
Polym Sci 2006;101(2):1051–6.
[25] Zheng Z, Pan H, Huang Y, Chung YH, Zhang X, Feng H. Rapid
liquefaction of wood in polyhydric alcohols under microwave
heating and its liquefied products for preparation of rigid
polyurethane foam. Open Mater Sci J 2011;5(1):1–8.
[26] Xie J, Qi J, Hse CY, Shupe TF. Effect of lignin derivatives in the bio-
polyols from microwave liquefied bamboo on the properties of
polyurethane foams. Bioresources 2014;9(1):578–88.
[27] Ouyang X, Lin Z, Yang D, Qiu X. Chemical modification of lignin
assisted by microwave irradiation. Holzforschung 2011;65(5):
697–701.
[28] Cinelli P, Anguillesi I, Lazzeri A. Green synthesis of flexible
polyurethane foams from liquefied lignin. Eur Polym J
2013;49(6):1174–84.
[29] Kurimoto T, Koizumi A, Doi S, Tamura Y, Ono H. Wood species effects
on the characteristics of liquefied wood and the properties of
polyurethane films prepared from the liquefied wood. Biomass
Bioenergy 2001;21(5):381–90.
[30] Pawlik H, Prociak A. Influence of palm oil-based polyol on
the properties of flexible polyurethane foams. J Polym Environ
2012;20(2):438–45.
[31] Goddard RJ, Cooper SL. Polyurethane cationomers with pendant
trimethylammonium groups. 1. Fourier Transform Infrared
Temperature studies. Macromolecules 1995;28(5):1390–400.
[32] Cole KC, Van Gheluwe P. Flexible polyurethane foam I. FTIR analysis
on residual isocyanate. J App Polym Sci 1987;34(1):395–407.
[33] Zhang XD, Bertsch LM, Makosco CW, Turner RB, House DW, Scott RV.
Effect of amine additives on flexible molded foams properties. Cell
Polym 1998;17(5):327–49.
[34] Javni I, Petrovic Z, Guo A, Fuller R. Thermal stability of polyurethanes
based on vegetable oils. J App Polym Sci 2000;77(8):1723–34.
[35] Satheesh Kumar MN, Siddaramaiah. Thermogravimetric analysis and
morphological behavior of castor oil based polyurethanes polyester
nonwoven fabric composite. J App Polym Sci 2007;106(5):3521–8.
[36] Aneja A, Wilkes GL. Exploring macro- and microlevel connectivity of
the urea phase in slabstock flexible polyurethane foam formulations
using lithium chloride as a probe. Polymer 2002;43(20):5551–61.
[37] Eceiza A, Martin MD, de la Caba K, Kortaberria G, Gabilondo N,
Corcuera MA, et al. Thermoplastic polyurethane elastomers based on
polycarbonate diols with different soft segment molecular weight
and chemical structure: mechanical and thermal properties. Polym
Eng Sci 2008;48(2):297–306.
[38] Danch A, Ilisch S, Sulkowski WW, Moczynski M, Radon A, Radusch
HJ. DMTA study on the urethane network in rubber waste-urethane
composites. J Therm Anal Calorim 2005;79(3):623–30.
[39] Hernandez R, Weksler J, Padsalgikar A, Choi T, Angelo E, Lin JS, et al.
A comparison of phase organization of model segmented
polyurethanes with different intersegment compatibilities.
Macromolecules 2008;41(24):9767–76.
[40] Gibson LJ, Ashby MF. Cellular solids: structure and properties.
Cambridge University Press; 1997.
[41] Rossmy GR, Kollmeier HJ, Lidy W, Schator H, Wiemann M.
Mechanism of the stabilization of flexible polyether polyurethane
foams by silicone-based surfactants. J Cell Plast 1981;17(6):319–27.
[42] Moreland JC, Wilkes GL, Turner RB. Viscoelastic behavior of flexible
slabstock polyurethane foams: dependence on temperature and
relative humidity. I. Tensile and compression stress (load)
relaxation. J App Polym Sci 1994;52(4):549–68.
[43] Gouveia de Souza A, Oliveira Santos JC, Conceição MM, Dantas Silva1
MC, Prasad S. A thermoanalytic and kinetic study of sunflower oil.
Brazil J Chem Eng 2004;21(02):265–73.
[44] Ogunniyi DS. Castor oil: a vital industrial raw material. Biores
Technol 2006;97:1086–91.