Biomacromolecules 2001, 2, 1029-1037 1029

Composite Films Based on Biorelated Agro-Industrial Waste

and Poly(vinyl alcohol). Preparation and Mechanical Properties
Characterization†
Emo Chiellini,*,‡ Patrizia Cinelli,‡,§ Syed H. Imam,|,⊥ and Lijun Mao
Department of Chemistry & Industrial Chemistry, University of Pisa, Via Risorgimento 35, 56126 Pisa, Italy;
and Plant Polymer Research Unit, National Center for Agricultural Utilization Research,
Agricultural Research Service, USDA, 1815 North University Street, Peoria, Illinois 61604
Received May 4, 2001

As a part of an ongoing project on the production of composite materials based on poly(vinyl alcohol)
(PVA) and polymeric materials from renewable resources, the present paper reports on the incorporation of
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

agricultural waste materials as organic fillers in a film matrix based on PVA as continuous phase. In this
study lignocellulosic fibers byproducts, derived from sugar cane (SC) and apple (AP) and orange (OR) fruit
Downloaded via UNIV DE CONCEPCION UDEC on May 15, 2021 at 17:19:18 (UTC).

juice extraction, were cast from PVA aqueous solutions. The effect of fiber type and composition on the
relative properties of cast films was evaluated and compared. OR resulted to be suitable for blending in
higher amounts by weight than SC and AP. Glycerol and urea were added as plasticizing agents and were
observed to be effective in giving flexible films. Additionally, cornstarch was added to further increase the
composition of polymers from renewable resources in cost-effective and ecoefficient composite film
formulations. The prepared films resulted sensitive to moisture and water. To reduce water sensitivity,
hexamethoxymethylmelamine (HMMM) was tested as a cross-linking agent for the present composite
formulations. Cross-linked films exhibited significant improvement in water-resistance that can be taken as
a tuneable structural feature for customized applications. The mechanical properties of the prepared composite
films (elongation at break, tensile strength, Young modulus) were found to be dependent upon the nature
and content of the filler and on environmental conditions.

Introduction Waste”. According to such a directive a progressive reduction

is planned of the amount of the organic biodegradable
More than 150 million metric tons of synthetic plastics fraction in the municipal solid waste to a level of 75% in
are produced globally from petroleum-based raw materials.1 the year 2006, 50% in the year 2009 and 35% in the year
Much of these plastics are used to manufacture single-use 2016 of the total organic fraction present in the municipal
consumer products. This poses a genuine environmental solid waste in the year 1995.4 Once these figures are
concern, as most of these plastic items are recalcitrant to accomplished, a minimization of the organic components in
environmental attack and do not degrade when disposed in the waste to be disposed in landfill can be envisaged, with
the environment at the completion of their life cycle. In landfilling practice applying only to the disposal of inerts
industrialized countries the annual output of municipal solid by the year 2030.
waste (MSW) is in the range of 0.5 metric tons per capita
Currently, any systematic collection of plastic waste for
with a plastic waste share that ranges from 5 to 10 wt %,
recycling and/or disposal is expensive and is limited only to
that is 15-25 vol %.2 Most of this plastic was recognized
certain communities. In particular, when plastics are con-
as packaging material intended for disposal after one-time
taminated with soil, foods, or other chemicals, their recycling
use.3
is rather difficult,5 and for this reason, only 2% of the plastic
At the same time landfilling practice is going to be banned
waste is nowadays recycled in the United States.
in the near future for environmental and public health rea-
sons.4 In Europe the European Commission has passed Much effort has been focused in recent years to develop
recently a Directive (1999/31/EC) on “The Landfill of environmentally compatible plastic products by incorporating
renewable materials as an alternative to petroleum-based
* To whom all correspondence should be addressed. Telephone: +39- synthetic polymers.6 As an added advantage, renewables are
50-918299. Fax: +39-50-28438. E-mail: chlmeo@dcci.unipi.it. comparatively less expensive, environmentally friendly, and
† Names are necessary to report factually on available data; however
naturally biodegradable. These bioplastics cannot replace
the USDA neither guarantees nor warrant the standards of the product, and
the use of the name USDA implies no approval of the product to the synthetic plastics in every application but they can result in
exclusion of others that may also be suitable. appropriate specific products, especially for those applica-
‡ University of Pisa.
§ E-mail: cinellip@dcci.unipi.it.
tions in which recovery of plastics is not economically
⊥ E-mail: imamsh@ncaur.usda.gov. feasible, viable, and controllable such as plastic items for
| USDA. one-time use.3
10.1021/bm010084j CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/06/2001
1030 Biomacromolecules, Vol. 2, No. 3, 2001 Chiellini et al.

Biodegradability of biopolymers is the base for their The purpose of the present work was 2-fold: (a) to produce
applications, but their processing, performances, and, in cast films by blending PVA with fibers of sugar cane
particular, price are by far more important issues since bagasse, apple, and orange byproducts and compare the
biopolymers should compete with synthetic low cost poly- performances of the different fibers; (b) to investigate the
mers.3 In these regards, materials such as renewable crops, effect on mechanical properties of starch, plasticizer, and
agricultural waste, and/or byproducts are a good source of cross-linking agent addition in order to optimize the relative
low cost natural polymeric materials.7 Among them there amounts to be introduced in the formulations.
are considerable quantities of agro-based fibers available on For this purpose, PVA was first blended with lignocellu-
a worldwide basis for a variety of applications.8 The poor losic fibers in increasing amounts in absence of any plasticiz-
dimensional stability and lack of thermoplasticity of ligno- ers. Subsequently a 50/50 PVA/fibers blend was chosen as
cellulosic fibers have limited the use of these materials to reference material and glycerol and urea were added as
produce single use articles. Thus, to produce items of plasticizers and processing aids. Urea is a well-known
practical use fibrous materials have been blended with fertilizing agent and was selected in view of utlizing PVA/
thermoplastic matrices in composites containing various lignocellulosic-based items for agriculture applications.
percentages of the fibers.9 Cornstarch was added to further increase the composition
In this regard, poly(vinyl alcohol) (PVA), a hydrolysis of polymers from renewable resources in cost-effective and
product of polyvinyl acetate, is well suited for blends with ecoefficient composite films formulations.
natural polymeric materials since it is highly polar, can be Films based on PVA and natural fillers are known to be
manipulated in water solutions, and is also biodegradable.10 moisture and water sensitive; thus, cast films are usually
PVA films and molded items have good physical properties coated with water resistance layers or submitted to a cross-
such as tensile strength and elongation at break. PVA and linking treatment. In a previous study hexamethoxymeth-
starch blends have been prepared by casting of water ylmelamine (HMMM) exhibited good performance as a
solutions11-13 and subsequently proposed on the market for cross-linker agent in PVA/starch formulations.21 In the
use as agricultural mulch films and as water-soluble laundry present paper, HMMM has been tested on PVA/starch-based
bags.14-16 Other applications in agriculture, civil construction, composite materials enriched also with a consistent amount
one-time-use containers and utilities are documented on the of lignocellulosic fibers. HMMM is a low iminomelamine-
web sites of the following companies: Environmental formaldehyde cross-linker approved by the Food and Drug
Polymers,17 Aquafilm,18 Idroplast,19 and Millenium.20 Administration for its low formaldehyde release effect. The
More recently studies have been performed on cast films effect of HMMM addition on mechanical properties and
made from PVA and other natural polymers. Thus, cast films water resistance was assessed.
based on PVA have been prepared with chitosan and
pectin.21-23
Experimental Section
PVA and cellulose cast films prepared in N,N-dimethy-
lacetamide-lithium chloride showed good miscibility due Materials. Poly(vinyl alcohol) (PVA Airvol 425) was
to the mutual ability of PVA and cellulose to form intra- purchased from Air Products & Chemicals Inc., Allentown,
intermolecular hydrogen bonds between hydroxyl groups.24 PA. It was characterized by 95.5-96.5% degree of hydrolysis
As a part of a research program aimed at the preparation with an average molecular weight ranging between 100 000
and evaluation of environmentally degradable composite and 146 000. Hexamethoxymethylmelamine (Cymel 303) is
materials based on PVA and natural polymers for various a low imino melamine-formaldehyde cross-linking agent with
applications in agricultural and packaging practices, we a DP of 1.75 and an average degree of methylation of 97%
started to consider the possibility of utilizing lignocellulosic and was purchased from Cytec Industries. Inc., Wallingford,
fibers generated in agricultural industry. A previous study, CT. Citric acid was obtained from Aldrich Chemical Co.,
with specific reference to the in situ formulation of self- Milwaukee, WI. Glycerol and urea were purchased from
fertilizing mulching films, has shown that formulations Fisher Chemicals, St. Louis, MO. Unmodified commercial-
containing PVA, gelatin, and sugar cane bagasse are suitable grade corn/starch (Buffalo 3401) with approximately 30%
both for preparing films by casting or for simply spraying amylose and 70% amylopectin was obtained from CPC
water suspensions of the polymeric components.25-28 International Inc., Argo, IL. Lignocellulosic materials were
The present contribution reports on the incorporation of from three different sources. Sugar cane bagasse (SC) was
agricultural waste materials as organic fillers in a film matrix supplied by US Sugar Corporation, Clewiston, FL, orange
composed of PVA. The lignocellulosic organic fillers, (OR) and apple pomace (AP) were the remains of fruit
derived from three different sources, were sugar cane bagasse residue after juice extraction. They were supplied by the
(SC) and waste from fruits such as apple (AP) and orange Sunflo Cit-Russ Limited, Lahore, Pakistan, and Tanner’s
(OR) after their juices have been extracted. Orchard, Speer, IL, respectively. All lignocellulosic materials
Casting films production is a well-established technology were milled and sieved to obtain 0.188 mm size particles.
for films based on polymeric materials from renewable Sieved materials were analyzed for their composition and
resources,16,29 but it represents also an opportunity to study moisture content (Table 1).
the effect of different components on film properties and Formulations and Sample Preparation. A predetermined
hence to provide cost-effective ecoefficient formulations amount of PVA was added to water to achieve a solution of
customized to specific needs.30,31 10% solid content. The mixture was slowly heated to about
Composite Films Biomacromolecules, Vol. 2, No. 3, 2001 1031

Table 1. Composition of Lignocellulosic Fibers Utilized in the stubs with graphite-filled tape, vacuum coated with gold-
Formulation of Composite Mixtures palladium, and examined under the scanning electron mi-
crude crude croscope (JEOL JSM 6400V, JEOL Inc., Peabody, MA).
protein crude fiber ash cellulose lignin moisture Fractures surfaces were prepared under liquid nitrogen by
fibersa (%) fat (%) (%) (%) (%) (%) (%) breaking the strips and specimens were examined under the
SC 4.22 1.97 32.70 11.41 35.64 10.56 7.71 SEM.
OR 12.88 6.60 14.74 11.30 17.70 4.76 9.88
AP 4.66 11.48 27.33 1.27 21.58 21.06 7.40
a SC ) sugar cane bagasse, OR ) orange byproducts, AP ) apple
Results and Discussion
byproducts.
Film Morphology and Mechanical Testing. In the first
90 °C with stirring until a homogeneous solution was formed. instance, cast films were prepared by blending PVA and
Water was added to compensate for any moisture loss that fibers alone without any plasticizers present in the formula-
may have occurred during the heating process. In a 250 mL tion. Respectively 10, 30, 50, and 70 wt % of fibers were
beaker, about 62 g of PVA solution was introduced, and the introduced in the composite films (Table 2). Blending of
desired weight amount of glycerol, urea, starch, and water fibers with PVA conferred a peculiar coloration to the films.
was added to bring a final concentration of solids equivalent PVA films produced with addition of SC, AP, and OR
to 10 wt %. The resulting mixture was first heated at 80 °C presented respectively a brown, an orange, and a reddish-
for 30 min under stirring, and 6.2 g of fibers was added and brown coloration. The amount of fiber content in the film
the mixture was stirred for additional 10 min. For cross- influenced both the appearance and integrity of the films.
linked samples, the desired amount of hexamethoxymeth- At 10% fiber concentration, films were translucent; above
ylmelamine and a catalytic amount of citric acid were also 30%, they became opaque. When the fiber contents were
added into the formulations, and the resulting mixture was allowed to rise higher than 50%, films became hard and
stirred at 70 °C for 45 min. After cooling at room temper- fragile. At 70% fiber content, films were too brittle and not
ature, 3 drops of BYK-019 aqueous defoamer was added suitable for Instron testing.
and the mixture was further stirred for 5 min. Table 2 Shape and composition data of fibers also influenced film
provides the description of blends and their compositions. properties. SC was lower in apparent density than AP and
To prepare films, about 17 g of aqueous suspension (as OR. Previous reports have indicated that the SC was not a
described above) was poured into a polypropylene plate (8 homogeneous material but rather a mixture of large cylindri-
× 8 cm) and left to dry overnight at ambient temperature cal (about 60%) and small spherical (about 40%) shaped
(23-24 °C) and finally for 3 h in a oven at 50 °C. Films fibers.32 As a result, the surfaces of the SC-based film
were cooled to room temperature before being peeled off appeared to be extremely corrugated (Figure 1a). A freeze-
the plates. Dog-bone-shaped specimens were stamp-cut and fractured surface of PSC50 indicated many empty spaces
stored under 50% controlled relative humidity (RH) for 10 due to the formation of aggregates of fibers as shown in
days at 23 °C before testing. Specimens were also condi- Figure 2a. Moreover microphotographs outlined a random
tioned at different relative humidity at 23 °C for 28 days in distribution of the fibers in the PVA matrix attributed to the
a saturated LiCl-water solution atmosphere (15% RH) and casting procedure.
in a saturated Na2SO4-water solution atmosphere (95% RH) Cast films based on AP fibers presented similar corrugation
before testing. The specimens, type IV ASTM D638, had a on the upper surface even if less pronounced than in films
4 mm width in the test neck, a length of 80 mm, and a neck based on SC (Figure 1b). Pictures of the freeze-fractured
length of 30 mm. The film thickness was determined by surface revealed a uniform and close-packed distribution of
averaging of three measurements along the test length using fibers of about 10 µm in diameter embedded in a homoge-
a Mini Test 3000 (Elektro-Physik, Cologne, Germany). neous smooth polymeric matrix (Figure 2b).
Mechanical Testing. Percent elongation (EL), ultimate Cast films based on OR presented a smoother upper
tensile strength (UTS), and Young modulus (YM) for each surface than those observed in films based on SC and AP
film were measured using an Instron Universal Testing (Figure 1c). Freeze fracture showed a compact structure with
System (Instron Corp., Canton, MA). The gauge length was a uniform distribution of the filler in the PVA matrix (Figure
25 mm, the grip distance was 45 mm, and the crosshead 2c). Composition analysis of fibers from orange byproducts
speed of the Instron was 10 mm/min. Data were collected reported indeed a lower content of lignin and crude fiber
at a rate of 20 points/s. At least five specimens for each and a higher content of protein and starch than AP and SC
sample were tested, and data were averaged. (Table 1).
Sample Analysis. To measure water sensitivity of the Consequently the effect on mechanical properties of PVA
films, sample strips (10 × 40 mm) were soaked in a test based cast films was different for the addition of, respec-
tube containing 20 mL of deionized water at 25 °C, and time tively, SC, AP, and OR. When conditioned at 50% RH, a
was recorded until total disintegration occurred. Moisture PVA cast film had an ultimate tensile strength (UTS) of 44
content of samples was evaluated with a Sartorius MA 30 MPa, the elongation at break (EL) was 224%, and the Young
moisture analyzer by heating about 1 g of substance at 130 modulus (YM) was 1040 MPa. Introduction of the ligno-
°C up to constant weight. Percent weight loss was recorded cellulosic fibers in the PVA matrix sharply reduced EL even
and reported as moisture uptake. For scanning electron at relatively small concentrations (Figure 3a). EL changed
microscopy (SEM), samples were mounted on aluminum from 224% of pure PVA to values lower than 30% just for
1032 Biomacromolecules, Vol. 2, No. 3, 2001 Chiellini et al.

Table 2. Composition of the Composite Films Prepared by Casting from Water Suspensions of PVA
samplea PVA (%) natural fibers (%) glycerol (%) urea (%) starch (%) HMMMb (%) citric acid (%)
PSC10, PAP10, POR10 90.0 10.0
PSC30, PAP30, POR30 70.0 30.0
PSC50, PAP50, POR50 50.0 50.0
PSC70, PAP70, POR70 30.0 70.0
PSCG1, PAPG1, PORG1 40.0 40.0 10.0 10.0
PSCG2, PAPG2, PORG2 33.3 33.3 16.6 16.6
PSCG3, PAPG3, PORG3 25.0 25.0 25.0 25.0
PSCSt1, PAPSt1, PORSt1 31.0 31.0 15.0 15.0 8.0
PSCSt2, PAPSt2, PORSt2 28.6 28.6 14.3 14.3 14.3
PSCSt3, PAPSt3, PORSt3 25.0 25.0 12.5 12.5 25.0
PSCX1, PAPX1, PORX1 27.7 27.7 13.8 13.8 13.8 2.7 0.3
PSCX2, PAPX2, PORX2 27.3 27.3 13.7 13.7 13.7 4.0 0.4
PSCX3, PAPX3, PORX3 26.3 26.3 13.1 13.1 13.1 7.7 0.8
a PSC series based on sugar cane bagasse (SC) as filler, POR series based on orange (OR) as filler, PAP series based on apple (AP) as filler.
b HMMM ) hexamethoxymethylmelamine.

Figure 1. SEM micrographs (100×) of (a) PSC50, (b) POR50, and

(c) PAP50.

10% of fibers in the composite films. The dramatic reduction

Figure 2. SEM fracture (750×) of (a) PSC50, (b) POR50, and (c)
in EL was observed for the addition of all the fibers, and no PAP50.
significant differentiation among SC, AP and OR was
recognized. in the composite film in comparison with the effect derived
Interestingly, for the values of UTS and YM a significant from OR addition. In cast films based on PVA and SC or
difference was observed when SC and AP were introduced AP, UTS and YM decreased with fiber content increasing
Composite Films Biomacromolecules, Vol. 2, No. 3, 2001 1033

observed also in preliminary studies.30 In films prepared with

50% OR, UTS and YM dropped to lower values comp-
arable with what was recorded for films prepared with 50%
SC or AP.
Since the purpose of this study was that of investigating
the possibility to produce composite materials with high
content of natural fibers in the synthetic matrix, blends with
50% of fibers (PSC50, PAP50, POR50) were selected as
starting formulations for further investigation. At this PVA/
fiber ratio, to obtain flexible films and to improve elongation
at break, the addition of plasticizers is required. Increasing
amounts of plasticizers were introduced in the formulation
in order to define the optimal composition for films produc-
tion. Thus, glycerol and urea were introduced in formulations
in overall amounts corresponding respectively to 20%
(PSCG1, PAPG1, PORG1), 34% (PSCG2, PAPG2, PORG2),
and 50% (PSCG3, PAPG3, PORG3) by weight of the total
composition (Table 2).
Addition of glycerol and urea softened the films even if a
different behavior was recorded for the different fiber types.
Increasing amounts of plasticizer are expected to increase
EL and film flexibility but to reduce UTS and YM. Thus,
EL changed from 3% recorded for PSC50 to a maximum of
18% for PSCG3 (Figure 4a). Anyway the increase in EL
recorded for films based on PVA and SC due to the
plasticizers addition was modest and lower than what was
expected. Also, PSCG1, PSCG2, and PSCG3 presented
similar values for EL even if the amounts of plasticizer were
significantly different in the three compositions. This suggests
that, for SC, the 50% fiber content by weight is representing
a too high volume for an effective intimate blending with
the continuous polymeric matrix even in the presence of a
consistent amount of plasticizing agents. Also, in cast films
based on OR and AP, an increase in the plasticizer contents
corresponded to a sharp increase in EL values. In fact EL
changed from 3% recorded for PAP50 and POR50 to a
maximum of 320% for PAPG3 and 370% for PORG3.
In all the blends based on PVA/SC, PVA/AP, and, in
particular, PVA/OR, UTS and YM decreased sharply when
plasticizer contents in the mixtures increased from 0 to 20%
(Figure 4b,c). For plasticizer contents increasing from 20%
Figure 3. (a) Effect of increasing lignocellulosic fibers content on to 50%, a decrease of UTS and YM was also observed.
the elongation at break (El) in PVA/fiber films. (b) Effect of increasing Blends based on PVA and SC presented the lowest values
lignocellulosic fibers content on the ultimate tensile strength (UTS) for UTS in accordance with previous considerations. In all
of PVA/fiber films. (c) Effect of increasing lignocellulosic fibers content
on the Young’s modulus (YM) of PVA/fiber films. the PVA/fiber blends, the 50% plasticizers concentration
resulted to be excessive thus in PSCG3, in PAPG3, and,
from 0 to 50% (Figure 3b, 3c). This behavior was attributed especially, in PORG3 leaching of glycerol out of the film
to the presence of voids and to the random distribution of surfaces was observed with aging.
the fibrous fillers in films produced by casting. Cast films The formulation with 34% of plasticizers (PSCG2, PA-
based on PVA and OR presented a different behavior thus PG2, PORG2) was thus selected as the reference blend for
for a 10% content of fiber in the PVA matrix an increase of the investigation of the effect of starch addition on mechan-
UTS and YM was recorded in comparison with a PVA-cast ical properties. For this purpose, starch was introduced in
film (Figure 3b,c). This behavior suggests that in this amount increasing amounts corresponding to, respectively, 8%
(10%) OR acted as particulate reinforcement for the PVA (PSCSt1, PAPSt1, PORSt1), 14% (PSCSt2, PAPSt2, PORSt2),
matrix. In PVA/OR film with 30% OR, UTS and YM were and 25% (PSCSt3, PAPSt3, PORSt3) by weight on the final
still comparable with PVA values. This effect of OR addition mixture. Addition of starch in the blends resulted in a lowered
in comparison with SC and AP was attributed to the PVA and plasticizer percentages (Table 2). Starch further
difference in fiber composition together with the smaller improved the mechanical properties in PVA/SC films by
particle size and lower density of OR than SC and AP as causing increased cohesiveness between the polymers. This
1034 Biomacromolecules, Vol. 2, No. 3, 2001 Chiellini et al.

Figure 5. (a) Effect of increasing starch content on the elongation

Figure 4. (a) Effect of increasing plasticizer content on the elongation at break (EL) in PVA/fiber films. (b) Effect of increasing starch content
at break (EL) in PVA/fiber films. (b) Effect of increasing plasticizer on the ultimate tensile strength (UTS) of PVA/fiber films. (c) Effect of
content on the ultimate tensile strength (UTS) in PVA/fiber films. (c) increasing starch content on the Young modulus (YM) of PVA/fiber
Effect of increasing plasticizer content on the Young modulus (YM) films.
in PVA/fiber films.
prepared with increasing starch content. Similarly YM was
resulted in a more stabilized EL for PSCSt1, PSCSt2, and slightly increased by increasing starch content in the blends
PSCSt3 that was comparable with EL of the blend prepared and also in this case PORSt2 and PORSt3 presented similar
without starch (PSCG2) as shown in Figure 5 a. Beside by values. Thus, for PVA/SC and PVA/OR blends mechanical
increasing the amount of starch a higher UTS was recorded properties were almost the same from 8% to 25% of starch
for PVA/SC blends (Figure 5b). Also YM significantly content. In particular PORSt3 appeared to be a promising
increased from 10 MPa for PSCG2 to 23 MPa for PSCSt3. formulation for future developments of PVA/OR blends in
A sharp decrease in EL was observed in blends based on consideration of the fact that only 25% of PVA was used
PVA/AP and, in particular, on PVA/OR for addition of 8% for this film production. Also, in PVA/AP blends the highest
of starch. An increasing amount of starch decreased EL in amount of starch (25%) induced several defects such as holes
PVA/AP blends while interestingly for PVA/OR blends EL and phase separation among the component with consequent
presented similar values when starch was increased from 8% loss of cohesiveness and mechanical properties, thus indicat-
to 25% of the total formulation. Thus, UTS turned out to be ing that 8% and 14% of starch contents are to be preferred
almost the same for PAPG2, PORG2 (0% starch), and blends for PVA/AP blends.
Composite Films Biomacromolecules, Vol. 2, No. 3, 2001 1035

Table 3. Time of Disintegration in Water at 25 ˚C of Various

Composite Cast Films
sample td (h) sample td (h) sample td (h)
PSCG2 1.0 PORG2 1.0 PAPG2 2.0
PSCSt1 1.5 PORSt1 2.0 PAPSt1 4.0
PSCSt2 4.0 PORSt2 6.0 PAPSt2 4.0
PSCSt3 6.0 PORSt3 24 PAPSt3 24
PSCX1 24 PORX1 48 PAPX1 48
PSCX2 48 PORX2 96 PAPX2 72
PSCX3 240 PORX3 312 PAPX3 240
a td ) time of permanence in water before disintegration.

Sensitivity to Moisture and Water. Composites based

on PVA and fillers from renewable resources are known to
be extremely water sensitive. Results presented in Table 3 Figure 6. Moisture uptake in PVA, natural fibers, and starch as a
indicated time of permanence of the films when immersed function of RH.
in water before disintegration. The behavior of cast films
plasticized with 34% of urea and glycerol (PSCG2, PAPG2,
PORG2) was compared with that of the corresponding
composites enriched with increasing amounts of starch (from
8% to 25%). Samples based on PVA/fiber mixtures (PSCG2,
PAPG2, PORG2) swelled immediately when immersed in
water and disintegrated into small pieces (about 2 × 2 mm)
in a relatively short time (1-2 h). In composite films
prepared by increasing the starch content, from 8% to 25 wt
% of the total mixture, an increased time of permanence in
water was observed. This behavior is in accordance with the
improvement in composites cohesiveness observed by the
introduction of a limited amount of starch in the formulation,
as previously discussed. Also, these samples were observed
to be at the same swelling and loss of cohesiveness in a Figure 7. Moisture uptake in PVA/natural fibers un-cross-linked and
relatively short time, which corresponded to 1.5-6 h for the cross-linked composite films at, respectively, 15%, 50%, and 95%
series of composites based on PVA and SC and to about 24 RH.
h for the composites based on PVA/AP and on PVA/OR as uptake in PVA/fiber films at different RH lightly differed
reported in Table 3. for fibers type and films composition.
To improve resistance to water, HMMM was tested as Figure 7 compares moisture uptake in blends prepared with
cross-linking agent for the PVA/fibers composites. Samples 14% of starch (PSCSt2, PAPSt2, PORSt2) and the corre-
prepared with the same amounts of glycerol, urea, and starch sponding blends cross-linked with 8% of HMMM (PSCX3,
(PSCSt2, PAPSt2, PORSt2) were chosen as starting formula- PAPX3, PORX3). As expected, blends based on PVA/SC
tions to study the effect of addition of increasing amounts presented a lower moisture uptake in comparison with similar
of HMMM, respectively 3% (PSCX1, PAPX1, PORX1), 4% blends based on PVA/AP and PVA/OR. All samples stored
(PSCX2, PAPX2, PORX2), and 8 wt % of the total at 95% RH atmosphere for 28 days became wet and started
composition (PSCX3, PAPX3, PORX3). The indicate samples to lose cohesiveness. In particular, in samples containing
showed increased cohesiveness, and the degree of swelling starch-OR or starch-AP, growth of fungi on the film
in water decreased with increasing cross-linking. Time of surface was observed, and only samples containing SC as
permanence in water improved in a significant way by filler were cohesive enough to allow tensile tests. A lower
increasing HMMM over 3%. Indeed, only samples containing moisture uptake at 95% RH and resistance to fungal growth
8% of cross-linking agent were cohesive enough to maintain was observed in cross-linked samples (PSCX3, PAPX3,
their integrity when lifted off water. Thus, this amount of PORX3) most likely due to the steric hindrance exerted by
HMMM is recommended to be used in durable and relatively tighter entanglement. Also, cross-linked samples presented
water resistant formulations. a lower degradation rate in preliminary soil burial tests.33
Results presented in Figure 6 show moisture uptake in raw The different moisture uptake, after conditioning at dif-
materials when conditioned for 28 days at relative humidity ferent RH, influenced mechanical properties of the blends.
(RH) of respectively of 15%, 50%, and 95%. Moisture Equilibration of films at 15% RH severely reduced EL
absorption in PVA and pure starch has been shown previ- (Figure 8a) while increasing UTS (Figure 8b) and YM
ously to be directly proportional to relative humidity.13 (Figure 8c), thus confirming the importance of the plasticiz-
Results presented here show that moisture uptake in ligno- ing effect of water in composites of PVA and polymeric
cellulosic was enhanced as the relative humidity increased materials from renewable resources.13 The reduction in EL
especially above 50%. SC had the least moisture absorption was slightly higher in cross-linked blends while UTS was
capacity compared to AP and OR. Consequently moisture almost the same for cross-linked and un-cross-linked samples
1036 Biomacromolecules, Vol. 2, No. 3, 2001 Chiellini et al.

with 8% of HMMM significantly improved mechanical

properties maintenance after storing in atmosphere at rela-
tively high humidity (over 50%).

Conclusions

Composite films based on PVA and lignocellulosic waste

from renewable resources were cast from aqueous suspen-
sions. The PVA/fiber ratio and type of fibers influenced
mechanical properties of the resulting films. The presence
of voids and random distribution of the fibers lead to fragile
films for blends based on PVA/SC and PVA/AP also at a
relatively low amount of fibers in the composition. The low
content of hard fibers and the small size of OR fibers allows
for the maintaining of consistent UTS in blends containing
up to 30% of fibers. This indicates that OR can be blended
with PVA in a relative higher weight amount than that of
AP and SC.
The use of glycerol and urea or other suitable plasticizers,
in blends based on a 50/50 PVA/fiber ratio, yielded flexible
films. Their flexibility was depending upon the content and
nature of the plasticized component. Results suggest that a
50% amount of SC in the PVA matrix represents a too high
volume of lignocellulosic fibers for permitting efficient and
intimate blends with PVA also in the presence of plasticizers.
While for PVA/AP and PVA/OR films, high elongation
values and flexibility can be obtained with the appropriate
plasticizers addition.
Addition of cornstarch resulted in improved cohesiveness
in PVA/SC blends and in modest decreases in the mechanical
properties of PVA/OR blends. For these blends interesting
film forming properties and flexibility were maintained even
in the presence of 25% starch in the formulation. In PVA/
AP blends a 14% amount of starch has to be preferred due
to the formation of small defects in the films at higher starch
Figure 8. (a) Elongation at break (EL) of PVA/fiber un-cross-linked contents. The fact that starch could be successfully incor-
and cross-linked composite films at, respectively, 15%, 50%, and 95% porated in the blends offers a valuable potential for a
RH. (b) Ultimate tensile strength (UTS) of PVA/fiber un-cross-linked substantial reduction of final film cost.
and cross-linked composite films at, respectively, 15%, 50%, and 95%
All the blends turned out to be sensitive to water and
RH. (c) Young modulus (YM) of PVA/fiber un-cross-linked and cross-
linked composite films at, respectively, 15%, 50%, and 95% RH. moisture; thus hexamethoxymethylmelamine was observed
to be effective in enhancing resistance to disintegration in
(Figure 8b). At this relative humidity a significantly higher water and in reducing moisture uptake at high RH (95%)
value for YM (Figure 8c) was also recorded for cross-linked with a consequent improvement in the maintenance of
specimen indicating the formation of a more rigid structure mechanical properties.
in the presence of cross-linking agent. The introduction of The prepared films were environmentally friendly as the
HMMM in the formulation did not affect mechanical materials used in film preparation were nontoxic and
properties of blends conditioned at 50% RH. PSCX3, naturally biodegradable. Moreover, it is expected that if the
PORX3, and PAPX3 presented values of El, UTS, and YM films were to be used in agricultural applications, such as in
comparable to the values recorded for PSCSt2, PORSt2, and mulching practice, the presence of fibers would contribute
PAPSt2, respectively (Figure 8). As previously reported, organic matter to the soil upon degradation, whereas urea
samples stored at 95% RH atmosphere for 28 days underwent would provide an added nitrogen source to the soil.
a strong deterioration, and only PSCSt2 was cohesive enough By manipulating the type and amount of additives in the
to perform mechanical tests. Anyway, also for PVA/SC formulation, it is possible to obtain films with predictable
blends, cross-linked specimen (PSCX3) maintained higher properties. Beside the good compatibility showed by PVA
UTS and YM values. For blends based on PVA/AP and and SC, AP and OR fibers have induced interest for ongoing
PVA/OR, only cross-linked samples were tested (PAPX3, studies about the use of PVA and lignocellulosic fibers for
PORX3) showing values of EL and YM comparable to the the production of compression molded, extruded, and foamed
values recorded for the specimen conditioned at 50% RH, items. The choice of the fiber type as well as their relative
even if a reduction in UTS was observed. Thus, cross-linking contents in the formulations with PVA as the synthetic
Composite Films Biomacromolecules, Vol. 2, No. 3, 2001 1037

polymeric matrix can be taken as a parameter for tuning the (11) Otey, F. H.; Mark, A. M.; Mehltretter, C. L.; Russell, C. R. Ind.
ultimate properties of the relevant composites. Eng. Chem. Res. 1974, 13, 90-92.
(12) Lahalih, S. M.; Akashah, S. A.; Al-Hajjar, F. H. Ind. Eng. Chem.
Res. 1987, 26, 2366-2372.
Acknowledgment. Research was carried out as col- (13) Chen, L.; Imam, S. H.; Gordon, S. H.; Greene, R. V. J. EnViron.
laboration between the University of Pisa, Italy and the Plant Polym. Degrad. 1997, 5, 111-117.
Polymer Research Unit of the USDA-ARS-National Center (14) Otey, F. H.; Mark, A. M. U.S. Patent 3,949,145, 1976.
for Agricultural Utilization Research, Peoria, IL. The authors (15) Otey, F. H.; Doane, W. M. Starch Chemistry and Technology, 2nd
ed., Whistler, R. L., Bemiller, J. N., Paschall, E. F., Eds.; Academic
thank Ms. Paulette Smith, Ms. Jan. Lawton, and Mr. Gary Press: New York, 1984, pp 389-416.
Groose for technical assistance. The financial support for (16) Akashah, S. A.; Lahalih, S. M.; Al-Hajjar, F. H. E.P. 0,224,990, 1986.
Ms. Patrizia Cinelli’s Ph.D. thesis research was provided by (17) http://www.epgplc.com/company.html.
the Ministry of University and Technology of Italy and in (18) http://www.aquafilmuk.com/index.html.
(19) http://www.idroplast.com.
part by the NCAUR/ARS/USDA. We are grateful to US (20) http://www.milleniumpolymers.com.
Sugar Corporation, Clewiston, FL; Sunflo Citrus Limited, (21) Coffin, R.; Fishman, M. L.; Ly, T. V. J. Appl. Polym. Sci. 1996, 57,
Lahore, Pakistan, and Tanner’s Orchard, Speer, IL for 71-79.
(22) Mucha, M.; Piekielna, J.; Wieczorek, A Macromol. Symp. 1999, 144,
supplying the respective byproduct materials used in this 391-412.
study. Assistance of the USDA International Program Office (23) Kim, J. H.; Kim, J. Y.; Lee, Y. M.; Kim, K. Y. J. Appl. Polym. Sci.
is also greatly appreciated. 1992, 45, 1711.
(24) Nishio, Y.; Hratani, T.; Takahashi, T.; Manley, R. S. J. Macromol-
References and Notes ecules 1989, 22, 2547-2549.
(25) Cinelli, P. Ph.D. Thesis, Formulation and Characterization of
(1) Vogl, O. J. Macromol. Sci.sPure Appl. Chem. 1996, A33, 963- Environmentally Compatible Polymeric Materials for Agricultural
966 Applications, University of Pisa, 1999.
(2) Ress, B. B.; Calvert, P. P.; Pettigrew, C. A.; Barlaza, M. A. EnViron. (26) Kenawy E. R.; Cinelli, P.; Corti, A.; Miertus, S.; Chiellini, E.
Sci. Technol. 1998, 32, 821-827.
Macromol. Symp. 1999, 144, 351-364.
(3) Doane, W. M. J. Polym. Mater. 1994, 1, 229-237.
(27) Bargiacchi, E.; Chiellini, E.; Cinelli, P.; Corti, A.; Magni, S.; Miele,
(4) Document 399L0031, Council Directive 1999/31/EC of 26th April
1999 on the Landifll of waste, Official Journal L 182, 16/07/1999; S. US Pat. Appl. 09/450,016, 1999.
pp 0001-0019. (28) Chiellini, E.; Cinelli, P.; Corti, A.; Kenawy, E. R.; Grillo Fernandes,
(5) Llop. C.; Perez. A. Makromol. Chem., Macromol. Symp. 1992, 57, E.; Solaro, R. Macromol. Symp. 2000, 152, 83-94.
115-121. (29) Fishman, M. L.; Coffin, D. R. U.S. Patent 5,451,673, 1995.
(6) Steinbüchel, A. J. Macromol. Sci.sPure Appl. Chem. 1995, A32, (30) Westhoff, R. P.; Kwolek W. F.; Otey, F. H. Starch/starke 1979, 31,
635-660. 163-165.
(7) Narayan, R. In Polymers from Agricultural Coproducts; Fishman, (31) Lawton, J. W.; Fanta, G. F. Carbohydrate Polym. 1994, 23, 275-
M. L., Friedman, R. B., Huang, S. J., Eds.; American Chemical 280.
Society: Washington, DC, 1994, pp 2-28. (32) Nassar, M. M.; Ashour, E. A.; Wahid, S. S. J. Appl. Polym. Sci.
(8) Young, R.A. Vegetable Fibers, Chapter In Kirk Othmer Encyclopedia 1996, 61, 885-890.
of Chemical Technology, 4th ed., John Wiley & Sons: New York, (33) Chiellini, E.; Cinelli, P.; Imam, S. H. In Biorelated Polymers-
1994; Vol. 10, pp 727-744. Sustainable Polymer Science and Technology; Chiellini, E., Gil, H.,
(9) Young, R. A. In First International Lignocellulosics-Plastics Com- Buchert, J., Braunegg, G., Van der Zee, M., Gatenholm, P., Eds.;
posites, Leao, A. L.; Carvalho, F. X.; Frollini, E., Eds., Sao Paolo, Kluwer Plenum Press Publ.: Dordrecht, The Netherlands, 2001 (in
Brazil, March 13-15, 1996; pp 1-22. press.).
(10) Matsumura, S.; Tomizawa, N.; Toki, A.; Nishikawa, K.; Toshima,
K. Macromolecules 1999, 32, 7753-7761. BM010084J