molecules

Article
Potential of Cellulose Microfibers for PHA and PLA
Biopolymers Reinforcement
Gonzalo Mármol 1, * , Christian Gauss 2 and Raul Fangueiro 1,3
1 Centre for Textile Science and Technology (2C2T), University of Minho, 4800-058 Guimarães, Portugal;
rfangueiro@dem.uminho.pt
2 School of Science and Engineering, University of Waikato, Hamilton 3216, New Zealand;
cgauss@waikato.ac.nz
3 Department of Mechanical Engineering, University of Minho, 4800-058 Guimarães, Portugal
* Correspondence: gonzalomarmol@fibrenamics.com; Tel.: +351-917-798-754

Academic Editors: Pietro Russo and Sylvain Caillol




Received: 14 August 2020; Accepted: 7 October 2020; Published: 13 October 2020

Abstract: Cellulose nanocrystals (CNC) have attracted the attention of many engineering fields
and offered excellent mechanical and physical properties as polymer reinforcement. However,
their application in composite products with high material demand is complex due to the
current production costs. This work explores the use of cellulose microfibers (MF) obtained by
a straightforward water dispersion of kraft paper to reinforce polyhydroxyalkanoate (PHA) and
polylactic acid (PLA) films. To assess the influence of this type of filler material on the properties of
biopolymers, films were cast and reinforced at different scales, with both CNC and MF separately,
to compare their effectiveness. Regarding mechanical properties, CNC has a better reinforcing effect
on the tensile strength of PLA samples, though up to 20 wt.% of MF may also lead to stronger
PLA films. Moreover, PHA films reinforced with MF are 23% stronger than neat PHA samples.
This gain in strength is accompanied by an increment of the stiffness of the material. Additionally,
the addition of MF leads to an increase in the crystallinity of PHA that can be controlled by heat
treatment followed by quenching. This change in the crystallinity of PHA affects the hygroscopicity
of PHA samples, allowing the modification of the water barrier properties according to the required
features. The addition of MF to both types of polymers also increases the surface roughness of
the films, which may contribute to obtaining better interlaminar bonding in multi-layer composite
applications. Due to the partial lignin content in MF from kraft paper, samples reinforced with MF
present a UV blocking effect. Therefore, MF from kraft paper may be explored as a way to introduce
high fiber concentrations (up to 20 wt.%) from other sources of recycled paper into biocomposite
manufacturing with economic and technical benefits.

Keywords: cellulose; biopolymers; multis-scale; microfibers; cellulose nanocrystals

1. Introduction
Composites are a dominant tool from a material science point of view as their properties facilitate
the shape and form design while offering economic advantages for a wide variety of applications.
Nevertheless, most of the synthetic petrol-derived composites introduce environmental disadvantages,
of which their lack of biodegradability stands out. Plastic-based materials become worthless once
they reach the end of their lifespan and, most of the time, these products end up in landfills with a
subsequent environmental effect. Using a biobased or biodegradable polymer matrix to manufacture
composite materials allows the obtainment of biodegradable products: green composites [1,2].
This technology leads to manufactured goods with a closed lifecycle, which may promote a circular
economy. After green composites are discarded from their initial purpose, biogas production may be

Molecules 2020, 25, 4653; doi:10.3390/molecules25204653 www.mdpi.com/journal/molecules

Molecules 2020, 25, x FOR PEER REVIEW 2 of 17
Molecules 2020, 25, 4653 2 of 17
production may be expected as fuel or feedstock for more biopolymer manufacture [3]. From the
different biodegradable bioplastics, polylactic acid (PLA), polyhydroxyalkanoates (PHA),
expected as fuel or feedstock for more biopolymer manufacture [3]. From the different biodegradable
polycaprolactone (PCL), chitosan, starch, and cellulose can be highlighted. However, up to now PLA
bioplastics, polylactic acid (PLA), polyhydroxyalkanoates (PHA), polycaprolactone (PCL), chitosan,
and PHA are the ones with a wider range of applications given their physicochemical and mechanical
starch, and cellulose can be highlighted. However, up to now PLA and PHA are the ones with a wider
properties and reasonable price.
range of applications given their physicochemical and mechanical properties and reasonable price.
PLA is known as the utmost effective polymer for biomedical and packing composites, and it is
PLA ischoice
the likely knownfor as the
theutmost effectiveof
replacement polymer for biomedical
conventional plasticsandinpacking
a morecomposites,
sustainableand wayit is [4].
the
likely choice for
Nonetheless, otherthefeatures
replacementsuch of asconventional plastics
fragility, reduced in a more
impact sustainable
strength, and low way [4]. Nonetheless,
thermal stability of
other features such as fragility, reduced impact strength, and
PLA become a hurdle for the application of PLA in engineering purposes [5]. Likewise, low thermal stability of PLAPHAsbecome area
hurdle for attention
attracting the application of PLA in engineering
as biodegradable alternativespurposes [5]. Likewise,
for conventional PHAsand
polymers, are attracting
a significant attention
effort
as biodegradable alternatives for conventional polymers, and a significant
has been made in this regard. On the other hand, PHAs are polyesters synthesized by numerous effort has been made in this
regard. On
bacteria withtheremarkable
other hand,properties
PHAs arefor polyesters
the nextsynthesized
generation by of numerous bacteriafriendly
environmentally with remarkable
materials
properties
[6,7]. Medical for the
andnext generation
healthcare of environmentally
applications friendly materials
as cardiovascular [6,7]. Medical
tissue engineering, and healthcare
cartilage repair,
applications as cardiovascular
ophthalmological transplantations,tissuenerveengineering, cartilage
regeneration, skinrepair, ophthalmological
treatment, drug delivery transplantations,
systems, cell
nerve regeneration,
anchorage, amongstskin treatment,
others are some drugof delivery
the mostsystems,
encouraging cell anchorage,
uses [6]. Itamongst others are
is also possible tosome
find
of the mostapplications
alternative encouraging of uses
PHA [6]. It is also possible
for commercial nonmedicalto find alternative
purposes applications
like packaging, of material,
fiber PHA for
commercial
biofuels, nonmedical
a precursor purposes
of carbon like packaging,
material, fiber material,
paper finishing biofuels, astabilization
and nanoparticle precursor ofamongstcarbon
material,
others [6].paper finishing
However, theand nanoparticle
application stabilization
of PHA amongst
for composite others [6].isHowever,
production challenging the application
due to the
of PHA for composite production
associated high cost of its production [8]. is challenging due to the associated high cost of its production [8].
Another approach to reducing the composite materials’ carbon footprint
Another approach to reducing the composite materials’ carbon footprint is the addition of other is the addition of other
biodegradable components
biodegradable components in in their
their composition.
composition. In In this
this regard,
regard, discrete
discrete natural
natural fiber
fiber reinforcement
reinforcement
(NFR) improves the sensitivity to temperature and mechanical
(NFR) improves the sensitivity to temperature and mechanical performance of PLA, though performance of PLA, though it it is
is
difficult to
difficult to obtain
obtain aa homogenous
homogenous mixture mixture and and good
good adhesion between matrix
adhesion between matrix andand fiber. Given the
fiber. Given the
hydroxyl functional groups on the ends of natural fibers that confer
hydroxyl functional groups on the ends of natural fibers that confer them a hydrophilic behavior, them a hydrophilic behavior,
there is
there is aa polarity
polarity variation
variation between
between the the fiber–matrix
fiber–matrix interface
interface (Figure
(Figure 1),
1), contributing
contributing to to their
their poor
poor
adhesion [9].
adhesion [9]. In
In the
the case
case ofof PHA
PHA matrices,
matrices, limited
limited mechanical
mechanical improvement
improvement has has been
been reported
reported in in the
the
literature when cellulose-based fibers were used as reinforcement
literature when cellulose-based fibers were used as reinforcement [10–12]. [10–12].

Figure 1. Schematic connection of polylactic acid (PLA) and polyhydroxyalkanoate (PHA) with cellulose.
Figure 1. Schematic connection of polylactic acid (PLA) and polyhydroxyalkanoate (PHA) with
cellulose.
Since the mechanical properties of composites are essential in material selection, a possible addition
with reinforcing properties for both PLA and PHA matrices are cellulose nanoparticles. Several studies
Since the mechanical properties of composites are essential in material selection, a possible
reported that the use of cellulose nanocrystals (CNC) in PLA matrices increased the tensile
addition with reinforcing properties for both PLA and PHA matrices are cellulose nanoparticles.
strength [13,14], whereas the addition of CNC in PHA leads to limited reinforcement [15,16]. Besides the
Several studies reported that the use of cellulose nanocrystals (CNC) in PLA matrices increased the
improved mechanical properties of the PLA matrices, CNC exhibit singular structural features
tensile strength [13,14], whereas the addition of CNC in PHA leads to limited reinforcement [15,16].
and excellent physicochemical properties such as biocompatibility, biodegradability, renewability,
Besides the improved mechanical properties of the PLA matrices, CNC exhibit singular structural
low density, adaptable surface chemistry, and optical transparency.
Molecules 2020, 25, 4653 3 of 17

However, an aspect that may hinder the addition of CNC in composite commodities is the current
production process of this type of filler, which is energy-intensive [17]. Despite all the scientific
advances to obtain CNC in a more sustainable way, currently, this type of filler is not suitable for
large-scale industrial processes with a high demand for raw materials.
As an effortless solution to CNC addition, this work explores an innovative processing method to
obtain natural fiber reinforcement (NFR) from kraft paper microfibers and their addition in polymeric
matrices. The obtainment of this type of fibers is less energy demanding and requires low-tech
equipment. Thus, a higher replacement of polymer by natural fibers of up to 30% by mass in composite
matrices appears to be possible, bringing both economic and environmental advantages. On one
hand, the use of microfibers (MF) from kraft pulp creates a path to incorporate recycled paper in
composite manufacture, which increases sustainability and reduces production costs. On the other
hand, microfibers (MF) may help to improve the adherence between the polymeric matrix and the
natural fiber reinforcement used in later stages for the manufacture of multi-layer green composites,
combining several layers of staked biopolymers and natural mats. For that, the production of thin
films made out of PLA and PHA will incorporate CNC and MF separately as fillers to assess their
comparative tensile strength and microstructural properties.

2. Materials and Methods

2.1. Materials
Polylactic acid (PLA) Ingeo 4043D was kindly provided by NatureWorks® with an Melt Flow
Rate = 6 g/10 min at 10 ◦ C, specific gravity of 1.24 g/cm3 , and peak melt temperature between 145 ◦ C
and 160 ◦ C. Polyhydroxyalkanoate (PHA) was supplied by Goodfellow Cambridge, MFR = 3 g/10 min
at 170 ◦ C, specific gravity of 1.24 g/cm3 and peak melt temperature between 140 ◦ C and 160 ◦ C.
CNC with an average particle size of 75 nm was purchased from Celluforce (Montreal, QC, Canada),
in spray-dried form. The paper kraft used in this study is made from commercial Tunisian Alfa
bleached Stipa tenacissasima. The morphological parameters determined by Morfi apparatus are an
average length weighted in length = 841 µm, average width = 16.7 µm, coarseness = 0.073 mg/m,
average kink number = 1.253, average angle = 130.8◦ , kinked fibers = 29.3% and average curl = 8.15%.
Reagent grade chloroform (CHCl3 ) with 99% purity was purchase from Honeywell, CAS: 67-66-3,
with a density value of 1.48 g/cm3 (20 ◦ C) and vapor pressure of 210 hPa (25 ◦ C).

2.2. Sample Production

The method to obtain MF from kraft paper started with a shredding process of the kraft
until particles of around 20 cm2 were obtained. Then paper particles were immersed in water for
48 h. After long water immersion, mechanical stirring was applied for 1 h for paper disintegration.
The paper suspension in water was drained and oven-dried at 105 ◦ C until it reached constant weight.
To deagglomerate MF clusters, polymer pellets were added to the correspondent amount of MF
according to the different formulations (Table 1) and placed in a blender machine for mechanical
abrasion between MF and polymer particles.

2.2.1. CNC-Filled Samples

For CNC samples, first 10 mL of CNC suspensions in CHCl3 were placed in centrifuge tubes
for CNC deagglomeration by mechanical vibration for 10 min using a vortex shaker IKA VORTEX 3.
Then, suspensions were diluted by adding CHCl3 to obtain a total volume of 50 mL and suspension
stabilization was carried out by ultrasonic treatment for 5 min, so proper dispersion and distribution
of the fibers in CHCl3 was assured.
Molecules 2020, 25, 4653 4 of 17

2.2.2. MF-Filled Samples

For MF samples, the right fiber dispersion was obtained by high shear mixing using a co-rotating
twin-screw extruder Microlab, Rondol® . PLA and MF were blended using a temperature profile of
145 ◦ C–200 ◦ C–200 ◦ C–195 ◦ C–190 ◦ C (feed to die) and a rotation speed of 55 rpm, while PHA and MF
were blended with a temperature profile of 130 ◦ C–145 ◦ C–150 ◦ C–150 ◦ C–140 ◦ C at a rotation speed of
65 rpm. Later MF-polymer blends were chopped until pellet homogenization.
For every film production (both CNC-filled and MF-filled samples), 2.84 g of polymer was added
to 50 mL of CHCl3 (suspension in the case of CNC samples) for its dissolution with the aid of magnetic
stirring for 1 h at 60 ◦ C and 1 h cooling down at room temperature. Film casting took place in ceramic
trays to promote solvent evaporation at ventilated room conditions. In all the cases, solvent-cast films
with a thickness between 0.15 and 0.2 mm were obtained.

Table 1. Formulations of the different films assessed.

Polymer Filler Type Content (wt.%)

PHA PHA None 0
PHA + 1% CNC PHA CNC 1
PHA + 2% CNC PHA CNC 2
PHA + 3% CNC PHA CNC 3
PHA + 10% MF PHA MF 10
PHA + 20% MF PHA MF 20
PHA + 30% MF PHA MF 30
PLA PLA None 0
PLA + 1% CNC PLA CNC 1
PLA + 2% CNC PLA CNC 2
PLA + 3% CNC PLA CNC 3
PLA + 10% MF PLA MF 10
PLA + 20% MF PLA MF 20
PLA + 30% MF PLA MF 30
Sample preparation for each type of filler used, either cellulose nanocrystals (CNC) or microfibers (MF),
followed different routes.

2.3. Sample Testing

Tensile testing was conducted following the ASTM D882-02 standard. A total of 5 specimens for
each formulation were cut into 100 × 10 mm strips and tested at a crosshead displacement of 5 mm/min
in a universal testing machine (Hounsfield Tinius Olsen, model H100 KPS) equipped with a load cell of
2.5 kN. The averages of tensile strength, elongation at break and Young’s modulus for each composite
formulation are presented with the corresponding coefficient of variation (COV). The differences among
the treatment conditions on the evaluated properties were checked by a Tukey test and analysis of
variance (ANOVA) for significant (p < 0.05) differences. All analyses were performed using MINITAB
Release 18 Statistical Software.
Differential scanning calorimetry (DSC) analyses were carried out in a covered aluminum crucible
under a nitrogen flow of 100 mL/min in a Metler Toledo, (822e ) differential scanning calorimeter
following the ASTM D3418 standard. Samples were heated from room temperature to 200 ◦ C at a
heating rate of 10 ◦ C/min and cooled down at a rate of 10 ◦ C/min down to 30 ◦ C to determine the
transition temperature (Tg ), melting temperature (Tm ), cold crystallization temperature (Tcc ) and,
therefore, calculate the heat of fusion and degree of crystallinity of the samples (Xc ). Degree of
crystallinity (Xc ) was calculated according to [18]:

∆Hm
Xc ( % ) = × 100 (1)
W × ∆Hpolymer
Molecules 2020, 25, 4653 5 of 17

where ∆Hm is the melting enthalpy of the samples. ∆Hpolymer is the enthalpy for 100% crystalline PLA
and PHA, which is approximately 93.6 J/g and 146 J/g, respectively, [18,19] and W is the net weight
fraction of PLA or PHA in the composites. The enthalpy values were evaluated as the integral of their
corresponding peak, i.e., the area under the endothermic peak to estimate the melting enthalpy.
Thermogravimetric analysis was performed using a thermal analyzer (HITACHI STA 7200).
Dried samples were introduced in alumina crucibles and were heated from 20 ◦ C to 400 ◦ C at 10 ◦ C/min
under nitrogen flow at 40 mL/min.
Infrared spectra of the films were performed on a Shimadzu, IRAffinity-1S Bruker Fourier
Transform Infrared Spectrophotometer using an attenuated total reflectance (ATR) module.
Transmittance spectra were registered between 400 and 4000 cm−1 , with a resolution of 4 cm−1
and 45 scans.
Microscopic analysis was conducted by optical microscopy, scanning electron microscopy (SEM)
and atomic force microscopy (AFM). The surface of the films was assessed by an optical microscope
using a Leica DM750M at 50× and 100× magnification. Microfiber length and width, as well
as the apparent pore size of the films, were determined for every film formulation by averaging
100 image measurements with the aid of ImageJ software. Morphological analyses were realized in an
Ultra-high-resolution Field Emission Gun Scanning Electron Microscopy (FEG-SEM), NOVA 200 Nano
SEM, FEI Company. Topographic images were obtained with a secondary electron detector at an
acceleration voltage of 10 kV. Before morphological analyses, samples were coated with a thin film
(8 nm) of Au-Pd (80–20 weight %), in a high-resolution sputter coater, 208HR Cressington Company,
coupled to an MTM-20 Cressington high-resolution Thickness Controller.
The morphological analyses were performed by Scanning Atomic Force Microscopy, using a
Nanoscope III Multimode Atomic Force Microscope, from Digital Instruments, where images of
5 × 5 µm2 and 10 × 10 µm2 were acquired on the surface of all samples. The scanning mode used was
intermittent contact or tapping mode, in air, with a cantilever whose spring constant is 42 N/m and the
resonance frequency was approximately 310 kHz, determining the surface roughness of the films.
The contact angle was determined after 15 s of the contact of the drop of water with the surface of
the film using a Dataphysics OCA Contact Angle System.
The light transmittance of the film samples was determined using a UV–visible spectrophotometer
(UV/Vis spectrophotometer), UV-2600 Shimadzu in the wavelength of 200–600 nm. The transmittance
at 600 nm (T600 ) and 300 nm (T300 ) was used to evaluate the transparency and UV barrier property of
the films, respectively.

3. Results and Discussion

The results obtained by the mechanical, physical, and thermal analyses are summarized in Tables 2
and 3 and discussed in the next sections.

3.1. Tensile Tests

Samples were tested under tensile configuration to obtain the mechanical properties of the different
composite formulations. Figure 2 displays the values of tensile strength and elongation at ultimate
strength. Comparing the tensile strength values of both neat PLA and PHA samples, PLA exhibits
higher values (52.3 MPa) compared to PHA (20.2 MPa). These values are in accordance with the results
reported in the literature [20,21]. When CNCs are added to both types of matrices, tensile strength
increases for every formulation. The greater improvement in tensile strength is most noticeable in
PLA samples, where a maximum increase of up to 38% was achieved for 1% CNC PLA samples,
with no evident effect on the elongation at break. In fact, all the PLA formulations reinforced with
CNC presented statistically equivalent elongation at break values (Table 2). Regarding PHA samples,
for every CNC addition content, a lower tensile strength enhancement (approximately 10%) is observed
compared to PLA films. However, the elongation at break is considerably improved, especially the 1%
CNC PHA formulation with an increase of approximately 130%.
Molecules 2020, 25, 4653 6 of 17

Table 2. Mechanical properties PLA and PHA composites with different reinforcing fillers (CNC and
MF). Same letters (a, b, or c) mean there is no statistical difference among the composites with the
same matrix.

Tensile Strength (MPa) Elongation at Break (%) Young’s Modulus (GPa)

n=5
Avr. COV Avr. COV Avr. COV
PHA 20.2 c 0.07 4.47 b 0.06 0.54 c 0.12
PHA + 1% CNC 22.5 a,b,c 0.02 10.43 a 0.23 0.72 b 0.02
PHA + 2% CNC 22.6 a,b,c 0.06 5.85 b 0.13 0.81 a,b 0.06
PHA + 3% CNC 22.2 b,c 0.03 5.95 b 0.25 0.81 a,b 0.05
PHA + 10% MF 23.5 a,b 0.03 4.33 b 0.10 0.81 a,b 0.03
PHA + 20% MF 24.9 a 0.05 3.78 b 0.11 0.94 a 0.14
PHA + 30% MF 23.0 a,b 0.06 3.93 b 0.04 0.93 a 0.06
PLA 52.3 b,c 0.02 3.86 a 0.13 1.95 b,c 0.05
PLA + 1% CNC 72.9 a 0.05 4.02 a 0.12 2.12 a,b,c 0.07
PLA + 2% CNC 68.6 a 0.06 3.51 a 0.10 2.05 a,b,c 0.03
PLA + 3% CNC 65.5 a 0.06 3.54 a 0.13 1.92 c 0.03
PLA + 10% MF 55.2 b 0.02 2.00 b 0.16 2.28 a,b 0.10
PLA + 20% MF 65.2 a 0.05 3.38 a 0.04 2.39 a 0.09
PLA + 30% MF 47.0 c 0.12 2.37 b 0.12 2.36 a 0.07
MF 841 ± 71 µm
Fiber length
CNC ≈75 nm

Table 3. Summary of results obtained by physical characterization of PLA and PHA composites with
different reinforcing fillers (CNC and MF). Different letter between treatments means a significant
statistical difference (p > 0.05).

Maximum Degradation Rate First Heating Second Heating UV-VIS Transmittance

Temperature at Mass Loss at Contact
Xc Tg Tcc Xc Tg Tcc T300 T600
Peak Center Peak Center Angle
(◦ C) (%) (%) (◦ C) (%) (◦ C) (◦ ) (%)
PHA pellets 12.7 60.4
PLA pellets 28.5 60.8
PHA 269 342 39.81 87.86 8.1 ** 96.8 11.6 48.6 108.5 55.7 31.26 71.6
PHA + 1% CNC 270 330 35.16 88.38 7.2 ** 94.4 12.8 61.8 110.7 73.8 26.82 72.5
PHA + 2% CNC 273 330 32.45 87.93 8.5 ** 94.2 12.2 48.6 110.2 95.7 21.72 70.1
PHA + 3% CNC 275 317 29.93 87.15 8.3 ** 94.1 11.8 55.4 109.8 78.8 21.22 69.0
PHA + 10% MF 270 341 35.11 91.21 9.1 ** 96.8 13.9 49.2 112.3 78.6 15.7 71.0
PHA + 20% MF 277 341 19.08 80.55 10.4 ** 95.7 12 51.4 112.3 78.9 3.59 55.6
PHA + 30% MF 278 341 18.14 88.03 7.8 ** 99.6 13.5 52.8 113.4 67.5 0.39 30.1
PLA 368 57.46 42.7 55.6 ** 26.1 62.7 121.6 67.5 64.66 86.3
PLA + 1% CNC 368 54.19 33.1 57.7 ** 25.2 63 123.8 43.6 68.07 84.2
PLA + 2% CNC 368 54.13 34.4 57 ** 19.5 62 120.5 57.2 63.76 83.8
PLA + 3% CNC 368 54.06 37.2 57.7 ** 37.6 62.7 124.3 60.6 66.71 83.8
PLA + 10% MF 360 74.88 45.6 57 ** 50.7 62.5 118.2 67.6 27.81 79.1
PLA + 20% MF 357 77.6 38.9 59.4 ** 59.9 62.5 118 61.1 7.01 70.3
PLA + 30% MF 357 77.24 30.7 59.5 ** 31.2 62.4 114.2 60.9 2.04 40.5
MF 357 8.14 841 ± 71 µm
Fiber length
CNC 305 4.67 ≈75 nm
** Stand for non-determined values.

Regarding the use of MF as filler, PLA samples have lower tensile strength compared to PLA
containing CNC. Samples with 30% by mass of MF show a decrease in tensile strength. However,
tensile strength improves by around 25% with preservation of the elongation at ultimate strength in
the 20% MF-PLA samples. Therefore, the specific energy absorbed by 20% MF-PLA samples increases
under tensile loading. In comparison with samples with CNC, tensile strength improves even more for
MF-PHA composites, with an increase between 14 and 23%, where the highest value is obtained by the
20% MF samples. Nevertheless, this increase in tensile strength is accompanied by a reduction of the
elongation at ultimate strength.
 Molecules 2020, 25, x FOR PEER REVIEW 6 of 17

by the 20%
Molecules 2020, MF samples. Nevertheless, this increase in tensile strength is accompanied by a reduction
25, 4653 7 of 17
of the elongation at ultimate strength.

Neat Polymer 10% MF 20% MF 30% MF Neat Polymer 10% MF 20% MF 30% MF
45 80 14 14
75
70 12 12
40
65

Elongation at break (%)

Tensile strength (MPa)

Elongation at break (%)

Tensile strength (MPa)

10 10
60
35
55
8 8
50
30
45
6 6
40
25
35
4 4
30
20 25 2 2
20
CNC-PHA MF-PHA CNC-PLA MF-PLA CNC-PHA MF-PHA CNC-PLA MF-PLA
15 15 0 0
Neat Polymer 1% CNC 2% CNC 3% CNC Neat Polymer 1% CNC 2% CNC 3% CNC
45 80

(a) (b)
Figure2.2. Tensile
Figure Tensile strength
strength (a) and elongation
elongation at
at break
break(b)
(b)of
ofPHA
PHAand
andPLA
PLAfilms
filmsreinforced
reinforced with
with CNC
CNC
and
andMF
MFwith
withvarying
varying the
the filler
filler ratios. Error
Error bars
bars indicate
indicateone
onestandard
standarddeviation
deviationininallallcases.
cases.

Some
Somestudies
studies indicate
indicate that the addition
addition of of fibers
fibersto tonon-polar
non-polarthermoplastics
thermoplasticsreduces reducesthe the tensile
tensile
strength due to the poor fiber–matrix adhesion at the interface, which is related
strength due to the poor fiber–matrix adhesion at the interface, which is related to low interfacial to low interfacial stress
transfer [22]. In[22].
stress transfer orderIn to improve
order the reinforcing
to improve effect of
the reinforcing the of
effect fibers in these
the fibers polymers,
in these coupling
polymers, agents
coupling
are usedare
agents to used
increase the compatibility
to increase between
the compatibility the fibers
between and the
the fibers andmatrix [3]. However,
the matrix [3]. However,in this
in work,
this
work,
MF MF addition
addition (without (without
a coupling a coupling
agent) ledagent) led to enhancements
to enhancements in both in both modulus
tensile tensile modulus and
and strength,
strength,affinity
showing showing affinitythe
between between
MF and thethe
MF and the polymers
polymers used hereused (PLAhereand(PLAPHA). and PHA). Consistent
Consistent with these
with these
findings, findings,studies
previous previous studies
using PHAusing PHA composites
composites reinforced reinforced
with agave with agave
fibers fibers presented
presented identical
identical results, where the tensile modulus of the PHAs increased
results, where the tensile modulus of the PHAs increased with fiber addition without the use with fiber addition without theof
use of coupling agents [23]. Similarly, PLA-based composites reinforced with
coupling agents [23]. Similarly, PLA-based composites reinforced with regenerated cellulose (Lyocell), regenerated cellulose
(Lyocell),
hemp, kenaf,hemp, kenaf,
or cotton alsoorpresent
cotton an
also present an improvement
improvement in tensile strength in tensile strength
and Young’s and Young’s
modulus without
modulus without the addition
the addition of coupling agents [24]. of coupling agents [24].
Forall
For allthe
thereinforced
reinforcedsamples
samplesproduced
producedininthis this study,
study, Young’s
Young’s modulus
modulus increases
increases compared
compared to
to neat
neat polymer
polymer films (Table
films (Table 2), though
2), though in some
in some casescases
without without a statistical
a statistical difference,
difference, withwith the exception
the exception of 3%
of 3% CNC-PLA samples that exhibit slightly lower stiffness values compared
CNC-PLA samples that exhibit slightly lower stiffness values compared to neat PLA samples. The same to neat PLA samples.
The same
positive positive is
behavior behavior
observed is observed on the tensile
on the tensile strength strength
of all of allreinforced
the the reinforced composites,
composites, except
except the
the samples of PLA reinforced with 30% of MF. Thus, it is proven that the
samples of PLA reinforced with 30% of MF. Thus, it is proven that the addition of MF in both types of addition of MF in both
types of biopolymer may present a reinforcing effect when adequately formulated.
biopolymer may present a reinforcing effect when adequately formulated.
In tensile tests that used non-woven flax mat PHB composites produced by the film-stacking
In tensile tests that used non-woven flax mat PHB composites produced by the film-stacking
method with different fiber contents, the stiffness of the composite materials increased with the fiber
method with different fiber contents, the stiffness of the composite materials increased with the fiber
content [25]. For high fiber contents (>30 vol.%), flax/PHA composites presented similar elastic
content [25]. For high fiber contents (>30 vol.%), flax/PHA composites presented similar elastic modulus
modulus to short fiber flax/polypropylene and glass-mat-reinforced thermoplastic composites.
to short fiber flax/polypropylene and glass-mat-reinforced thermoplastic composites. Regarding tensile
Regarding tensile strength, the effect of fiber addition to PHA was not significant, whereas the
strength, the effect of fiber addition to PHA was not significant, whereas the addition of flax resulted
addition of flax resulted in a lower elongation at ultimate strength of the composites, i.e., approx.
in a lower elongation at ultimate strength of the composites, i.e., approx. 1.5% for all fiber volume
1.5% for all fiber volume fractions. Thus, the results brought here suggest a favorable application of
fractions. Thus, the results brought here suggest a favorable application of MF to reinforced PHA
MF to reinforced PHA matrices in composite products.
matrices in composite products.

3.2. Differential Scanning Calorimetry

Differential scanning calorimetry tests were performed to evaluate and compare the effect of the
reinforcement addition on the crystallinity of each polymer. Since annealing, i.e., slow cooling after
molding, relieves the internal stresses introduced during the film fabrication [26], a thermal treatment
consisting of a heating at 80 ◦ C for 24 h and later cooling at room temperature was applied before
DSC testing. The temperature chosen lies in between the glass transition temperature, Tg (~60 ◦ C)
and the cold crystallization temperature, Tcc (>100 ◦ C). It has been found that the polymer crystallite
morphology can be also modified upon annealing at a temperature higher than the Tg , thus changing
Molecules 2020, 25, 4653 8 of 17
Molecules 2020, 25, x FOR PEER REVIEW 8 of 17

morphology can be also modified upon annealing at a temperature higher than the Tg, thus changing
the physical properties of the polymers [26]. Thermal annealing is a simple route for stabilizing glassy
the physical properties of the polymers [26]. Thermal annealing is a simple route for stabilizing glassy
polymers via the densification of their polymer chains. The effects of annealing are depicted on the
polymers via the densification of their polymer chains. The effects of annealing are depicted on the
first DSC heating cycle of the samples (Figure 3a). Since no exothermic peak was observed during
first DSC heating cycle of the samples (Figure 3a). Since no exothermic peak was observed during
cooling,
cooling,the
the∆HΔHmmvalue
valuewas
wasused
used to
to calculate theXXc cofofthe
calculate the thePLA/PHA–CNC/MF
PLA/PHA–CNC/MF composites
composites (Table
(Table 3). 3).
Compared
Compared to the neat PHA pellets, pure PHA films after solvent casting undergo a reduction in in
to the neat PHA pellets, pure PHA films after solvent casting undergo a reduction
crystallinity, while
crystallinity, whilethe reinforcement
the reinforcementofofthe thefilms
filmsleads
leadstotohigher
highercrystallinity
crystallinity regardless
regardless ofof the
the type
type of
reinforcement. For low
of reinforcement. For low X values (Table 3) as is the case of PHA samples (8.1%), the
c Xc values (Table 3) as is the case of PHA samples (8.1%), the crystalline crystalline nature
of nature
the reinforcements may leadmay
of the reinforcements to anlead
increase
to an in the crystallinity
increase nucleation
in the crystallinity of the matrix
nucleation of thedue to adue
matrix better
arrangement of the polymer
to a better arrangement of around the surface
the polymer aroundofthe thesurface
fillers [27].
of theTherefore,
fillers [27].higher reinforcement
Therefore, higher
contents have a more
reinforcement pronounced
contents effectpronounced
have a more on polymereffect crystallinity,
on polymersincecrystallinity,
MF additionsincehas aMFhigher impact
addition
onhas
the acrystallinity
higher impact on the
of PHA crystallinity
compared withofCNC
PHAsamples.
compared Onwith
the CNC
other samples.
hand, PLA Onfilms
the other hand,
increase their
PLA films with
crystallinity increase their crystallinity
annealing in comparisonwith with
annealing
samples in comparison
without thermalwith samples
treatment.without thermal of
The addition
thetreatment. The addition
reinforcement reduces the of the reinforcement
crystallinity of the reduces
films forthe mostcrystallinity
formulations,of the films for
although most
the highest
formulations, although the highest crystallinity is achieved by samples
crystallinity is achieved by samples with 10% MF. A single peak in the melting region of neatwith 10% MF. A single peakPLA
in the3a)
(Figure melting
indicatesregion of neat PLA distribution
a homogeneous (Figure 3a) indicates
of ordereda crystals.
homogeneousHowever,distribution of ordered
the addition of MF in
PLAcrystals. However,
matrices leads to the addition of MF in
a heterogeneous PLA matrices
crystal leadsevidenced
distribution, to a heterogeneous crystalmelting
by a two-step distribution,
process.
evidenced by a two-step melting process.

Figure 3. Differential Scanning Calorimetry curves of the PHA and PLA films reinforced with CNC
and MF during the first heating stage (a) and during the second heating stage (b).
Molecules 2020, 25, 4653 9 of 17

Since a second heating step at a rate of 10 ◦ C/min was applied after a fast cooling (quenching),
the effect of quenching on the crystallinity of the samples is expected to be observed (Figure 3b).
The quenched PLA samples exhibit a strong cold crystallization peak, which is particularly sharp
for MF reinforced PLA films. In contrast, PHA samples exhibit a two-step broader low intensity
cold crystallization peak compared to the first heating. Thus, quenching leads to a recovery of
the crystallinity of PHA samples, to values similar to the initial values before processing (Table 3),
and increases Tcc , though a heterogeneous crystal distribution is promoted [28]. These exothermic
peaks reveal the presence of a large number of active nuclei. For neat PLA films, quenching reduces
crystallinity, but for PLA samples reinforced with MF, particularly for higher concentrations (20 and
30%), this fast cooling is translated into a higher crystallinity.

3.3. Fourier Transformed Infrared Spectroscopy

In order to assess any possible chemical interaction between the different polymers and the
reinforcing filler, FTIR analysis was performed. Regarding PHA samples, the bands in the range
3015–2955 cm−1 are assigned to –CH3 asymmetric stretching vibrations and those in the range
2940–2915 cm−1 to –CH2 asymmetric stretching vibrations [29]. The range from 2885 to 2845 cm−1 is
distinctive of symmetric stretching modes of –CH3 and –CH2 [29,30]. The peaks at 2976 cm−1 , 2934 cm−1
and 2874 cm−1 and the shoulder at 2923 cm−1 arise from the crystalline state and that at 2997 cm−1 from
the amorphous phase [31,32]. The shoulder at 3007 cm−1 , associated to –CH3 asymmetric stretching,
points to the presence of intermolecular CH–O hydrogen bonds in PHB crystals [29]. Regarding the
composites’ spectra, little differences may be noticed when composites are reinforced with CNC due
to the low concentration of the fillers and, therefore, no chemical modification is perceived in the
functional groups. In the case of MF reinforced composites, at 1163 cm−1 there is an increase in
intensity that represents the –C–O–C– stretching of the cellulose of the reinforcement [33]. However,
additional functional groups are not observed, evidencing a lack of chemical interference.
From the FTIR spectrum of PLA samples (Figure 4b), distinct characteristic FTIR peaks were
observed at 866 cm−1 , 1073 cm−1 , 1454 cm−1 , 1742 cm−1 and 2926 cm−1 corresponding to –C–O–C–
bond stretching, –CH3 asymmetric vibrations, –CH bending vibrations, –C=O vibrations and –CH3
symmetric vibrations, respectively, for neat PLA films [34]. All these peaks are also present in every
PLA-based sample, which indicates that PLA functional groups are not altered with the addition of
reinforcing fillers. For the case of CNCs, the peak at 2808 cm−1 represents the –CH stretching band,
peaks at 1163 cm−1 represent the –C–O–C– stretching in β-1,4-d glycosidic linkage present in CNCs
and peaks at 1427 cm−1 represent the symmetric –CH2 bending peak [33]. The presence of peaks
near 1750 cm−1 in all the composites indicates the presence of free –CO groups. Similarly, the peaks
around 1450 cm−1 and 2940 cm−1 in all the nanocomposites represent the symmetric –CH2 bending
and stretching vibrations, respectively [33]. It is worth mentioning that the spectrum of MF is similar
to CNC, revealing a high cellulose content in the microfibers after their bleaching.
Molecules 2020, 25, 4653 10 of 17
Molecules 2020, 25, x FOR PEER REVIEW 10 of 17

Figure Fourier
4. 4.
Figure FourierTransform
TransformInfrared
Infraredspectrograms of the
spectrograms of the PHA
PHA(a)
(a)and
andPLA
PLA(b)
(b)films
films reinforced
reinforced with
with
CNC
CNCandandMF.
MF.

3.4. Thermogravimetric Analysis

3.4. Thermogravimetric Analysis
Figure
Figure5 shows
5 shows the the
thermal
thermaldegradation
degradation withwith the increase in temperature
the increase of eachofanalyzed
in temperature polymer,
each analyzed
reinforcement filler, and composite to assess and compare the influence
polymer, reinforcement filler, and composite to assess and compare the influence of the addition of of the addition of cellulose
particles in these
cellulose polymeric
particles in these films. Regarding
polymeric films.theRegarding
different types of reinforcement,
the different MF presents a higher
types of reinforcement, MF
peak degradation temperature (357 ◦ C) compared to CNC (305 ◦ C). This difference may be attributed
presents a higher peak degradation temperature (357 °C) compared to CNC (305 °C). This difference
to may
a partial lignin content
be attributed in MFlignin
to a partial even after
contenttheir
in bleaching
MF even after sincetheir
lignin undergoes
bleaching sincethermal degradation
lignin undergoes
upthermal ◦
to 600 degradation
C [35]. PHAup (Figure
to 600 5a) exhibits
°C [35]. PHAa(Figure
two-step 5a)degradation process,
exhibits a two-step with two process,
degradation clearly defined
with
peaks ◦ C and 345 ◦ C. The thermal decomposition between 170 ◦ C and 250 ◦ C is related
twocentered at 275 peaks
clearly defined centered at 275 °C and 345 °C. The thermal decomposition between 170 °C
to and
the loss
250 of°C low molecular
is related to theweight
loss ofcompounds
low molecular of the biopolymer
weight compounds [36].ofThe
the maximum
biopolymermass [36]. loss
The of
PHAmaximum
occurs at mass ◦
275 loss
C andof PHA occurs atwith
is associated 275 °Ctheand is cleavage
ester associatedofwith the ester
the PHA cleavageby
component of elimination
the PHA
component
reaction by elimination
[36]. The addition ofreactionboth MF [36].
andTheCNC addition of both MF
reinforcement and CNC
increases thereinforcement
thermal stability increases
of PHA
the thermal stability of PHA since the mass loss is reduced
◦ in the composites
since the mass loss is reduced in the composites at 275 C. The difference in behavior between MF and at 275 °C. The difference
CNCin behavior
reinforced between MF and
composites atCNC
345 ◦reinforced
C is related composites
to the mass at 345 °C is of
fraction related to the samples,
MF-based mass fraction
which of is
MF-based samples, which is 10 times higher than the filler content in CNC-based
10 times higher than the filler content in CNC-based samples. However, the overall thermal stability samples. However,
the overall
is similar thermal reinforced
for samples stability is with similarlowforMFsamples
contentreinforced with low MF
(10 wt.%) compared to CNCcontent (10 wt.%)
samples. In the
compared to CNC samples. In the case of PLA films, pure PLA samples present a single-step
case of PLA films, pure PLA samples present a single-step decomposition process with a maximum
decomposition process with a maximum decomposition rate at 375 °C, which takes place at higher
decomposition rate at 375 ◦ C, which takes place at higher temperatures than in the PHA samples
temperatures than in the PHA samples (275 °C) and both pure CNC (305 °C) and MF (357 °C) (Table
(275 ◦ C) and both pure CNC (305 ◦ C) and MF (357 ◦ C) (Table 3). The PLA degradation reaction is
3). The PLA degradation reaction is based on a hydroxyl end-initiated ester interchange process and
based on a hydroxyl end-initiated ester interchange process and chain homolysis [37]. Contrary to
chain homolysis [37]. Contrary to what is noted for PHA samples, the addition of MF in PLA samples
what is noted for PHA samples, the addition of MF in PLA samples (Figure 5b) reduces the thermal
(Figure 5b) reduces the thermal stability of the films by 37%. However, the addition of CNC preserves
stability of thestability
the thermal films byof37%. However,
the PLA the addition
films, since it has littleofeffect
CNCin preserves
this regardthegiven
thermal stability
the low CNCof the PLA
content.
films, since it has little effect in this regard given the low CNC content.

3.5. Microscopy Analysis

From Figure 6a,d it can be noted that both PHA and PLA films present a smooth surface with small
apparent pores, 22 and 27 µm, respectively. This apparent porosity is caused by the solvent casting
method since during solvent evaporation some air bubbles are entrapped and remain as voids within
the solid phase. Despite the porosity of the samples, the neat polymer matrices exhibit conventional
mechanical performance compared to the results reported in the literature [20,21]. A clear effect of the
addition of CNC in both PHA and PLA is the increase in apparent porosity since for PHA samples
the number of superficial pores is increased (Figure 6b) and for PLA samples the size of the pores is
enlarged (Figure 6e). This rise in apparent porosity leads to higher specific surfaces, which translate
into modifications of the hygroscopy of the materials.
Molecules 2020, 25, 4653 11 of 17
Molecules 2020, 25, x FOR PEER REVIEW 11 of 17

Molecules 2020, 25, x FOR PEER REVIEW 12 of 17

Figure
MF, the 5. Thermograms
number and theirat
of pores remains derivatives
the sameof oflevel
the PHA (a)unreinforced
as for and PLA (b) films reinforced
samples, with CNC
while
Figure 5. Thermograms and their derivatives the PHA (a) and PLA (b) films reinforced with the
CNCaverage
and
pore sizeMF.
increases up to 3.8 μm.
and MF.

3.5. Microscopy Analysis

From Figure 6a,d it can be noted that both PHA and PLA films present a smooth surface with
small apparent pores, 22 and 27 μm, respectively. This apparent porosity is caused by the solvent
casting method since during solvent evaporation some air bubbles are entrapped and remain as voids
within the solid phase. Despite the porosity of the samples, the neat polymer matrices exhibit
conventional mechanical performance compared to the results reported in the literature [20,21]. A
clear effect of the addition of CNC in both PHA and PLA is the increase in apparent porosity since
for PHA samples the number of superficial pores is increased (Figure 6b) and for PLA samples the
size of the pores is enlarged (Figure 6e). This rise in apparent porosity leads to higher specific
surfaces, which translate into modifications of the hygroscopy of the materials.
Contact angle measurement was conducted to evaluate the wettability of the pristine PHA and
PLA film and the nanocomposite films after their combination with different concentrations of CNC
and MF. Contact angle measurements and surface roughness may be related to each other [27].
Contact angle results (Table 3) reveal that the hydrophobicity, and therefore the water barrier
properties of the nanocomposite film, are somewhat related to the crystallinity degree of the films,
rather than the concentration of fillers. When the Xc is compared with the contact angle for every
Figure
sample afterSurface
6. Surface optical
optical
the first micrographs
micrographs
heating of neat
of
stage, a higher neatcrystallinity
PHA films
PHA films (a),
(a), PHA leads
PHA
degree films reinforced with 2 wt.% CNCfor
to higher hydrophobicity
(b),
both typesreinforced with
of polymer.
reinforced 20
with 20 wt.%
The MF (c),
increase
wt.% neat
MF (c), PLA (d),
in crystallinityPLA
neat PLA (d), films
of PLA reinforced
PHAfilms with
filmsreinforced 2 wt.%
with the additionCNC
with 2 wt.% of(e), reinforced
bothCNC types
(e), of
with 20
reinforcement wt.%
reinforced with MF (f).
is also
20 wt.%reflected
MF (f).in the increase in the contact angle. It might be explained by the
crystalline nature of the reinforcements, which leads to an increase in the hydrophobicity of the
Contact
The
matrix, thusangle
increase inmeasurement
apparent
increasing porewas
the contact size
angleconducted
for PHA
[27]. isto evaluate
accompanied
Therefore, the water thebywettability
an increase
barrier ofinthe
properties pristine
roughness,
of this type PHAas is
of
and
shown PLA
films by may film
AFM and the nanocomposite
in Figure 8. The
be functionalized films after
differenttreatment
by thermal their
mean roughness combination
accordingvalues with different
for unreinforced
to the desired concentrations
properties.PHA, Thus,1% whenCNC of
CNC
andMF and
is added
20% MF. Contact angle
to both types
MF reinforced measurements
of biopolymers,
samples are 117.7 nm, and
PLA surface
andnm
189.6 PHA, roughness
anda318.7
morenm, may be
hydrophobic related
respectively. to
performance each other
may be
Surface roughness [27].
Contact
is achieved angle
an essential by results
film (Table for
quenching.
parameter 3) reveal that thecomposites
multi-layer hydrophobicity, sinceand it therefore
may affect thethe
water barrier properties
interfacial bonding
between the polymer and reinforcement layers. Surface roughness promotes an enhancement ofthan
of the nanocomposite
In the case of film,
samples are somewhat
reinforced with related
MF, a to
roughthe crystallinity
surface with a degree
dense of the
network films,
of rather
microfibers the
thesurrounded
concentration
interfacial by a of
adhesion fillers.
thinandcoatingWhen
allows the Xc isstress
of polymer
greater iscompared
observed
transfer with
(Figure the contact
6c,f).
between The angle
matrixfor
the average every
length
and ofsample
the fibers
reinforcements, after
theafter
firstcasting
reducing heating is stage,
332.2 μm
the capacity aofhigher
with
fiberan crystallinity
average width
debonding. degreevalue
Therefore, leads
oftheto increase
11.3 higher
μm. The hydrophobicity
length
in of the
surface for both
fibers
roughness typesthe
decreases
with of
during
polymer. the
The high-shear
increase in mixing process
crystallinity of compared
PHA films to the
with initial
the morphological
addition
addition of MF may benefit the application of this type of material in multi-layer composites. of both characterization.
types of reinforcement The is
increase
also reflected in apparent porosity
in the increase in is
thealso confirmed
contact angle.by It SEM
might analysis. Figureby
be explained 7 shows the surface
the crystalline of PHA
nature of the
films with no reinforcement
reinforcements, which leads to(a), anwith 1% CNC
increase in the(b) and with 20% of
hydrophobicity of the
MF.matrix,
In spitethusof reducing
increasing thethe
number of apparent pores, it is clear that the addition of CNC increases
contact angle [27]. Therefore, the water barrier properties of this type of films may be functionalized the pore size compared to
byunreinforced
thermal treatment PHA, from an average
according to thepore size of
desired 2.2 μm upThus,
properties. to a 7.1 μm. MF
when A similar
is addedtrend to isboth
observed
types of
in PLA samples, where apparent pore size increases from 2.7
biopolymers, PLA and PHA, a more hydrophobic performance may be achieved by film quenching. μm up to 8.1 μm. With the addition of
In the case of samples reinforced with MF, a rough surface with a dense network of microfibers
surrounded by a thin coating of polymer is observed (Figure 6c,f). The average length of the fibers after
 Figure
Molecules 2020,6.25,
Surface
4653 optical micrographs of neat PHA films (a), PHA films reinforced with 2 wt.% CNC
12 of 17
(b), reinforced with 20 wt.% MF (c), neat PLA (d), PLA films reinforced with 2 wt.% CNC (e),
reinforced with 20 wt.% MF (f).
casting is 332.2 µm with an average width value of 11.3 µm. The length of the fibers decreases during
The increase
the high-shear in apparent
mixing process pore size for
compared to PHA is accompanied
the initial morphologicalby an increase in roughness,
characterization. as is
The increase
shown by AFM in Figure 8. The different mean roughness values for unreinforced
in apparent porosity is also confirmed by SEM analysis. Figure 7 shows the surface of PHA films PHA, 1% CNC
and 20%
with MF reinforced(a),
no reinforcement samples are CNC
with 1% 117.7 (b)
nm,and
189.6
withnm20%
and of
318.7
MF.nm, respectively.
In spite Surface
of reducing roughness
the number of
is an essential
apparent pores, parameter forthe
it is clear that multi-layer
addition of composites
CNC increasessincethe
it pore
may size
affect the interfacial
compared bonding
to unreinforced
between
PHA, theanpolymer
from averageand porereinforcement
size of 2.2 µmlayers.
up to aSurface
7.1 µm.roughness promotes
A similar trend an enhancement
is observed of the
in PLA samples,
interfacial
where adhesion
apparent andincreases
pore size allows greater
from 2.7stress
µm uptransfer between
to 8.1 µm. theaddition
With the matrix ofand
MF,reinforcements,
the number of
reducing
pores the capacity
remains at the same of level
fiber as
debonding. Therefore,
for unreinforced the increase
samples, while theinaverage
surfacepore
roughness with the
size increases up
addition
to 3.8 µm.of MF may benefit the application of this type of material in multi-layer composites.

Figure Scanningelectron
7. Scanning
Figure 7. electron microscope
microscope of films
of films surface
surface of neat
of neat PHAPHA (a), PHA
(a), PHA + 2%(b),
+ 2% CNC CNCPHA(b),
+
PHA + 20% MF (c) and fracture surface of neat PLA (d), PLA + 2% CNC (e) and PLA
20% MF (c) and fracture surface of neat PLA (d), PLA + 2% CNC (e) and PLA + 20% MF (f). + 20% MF (f).

The increase in apparent pore size for PHA is accompanied by an increase in roughness, as is
AFM also confirms the modification of the surface roughness with the addition of fibers for the
shown by AFM in Figure 8. The different mean roughness values for unreinforced PHA, 1% CNC and
PLA composite samples (Figure 8). As noted for PHA, in PLA films the roughness is also increased,
20% MF reinforced samples are 117.7 nm, 189.6 nm and 318.7 nm, respectively. Surface roughness is
from 74.4 nm (plain PLA) up to 88.7 and 174.8 nm for 1% CNC and 20% MF samples, respectively.
an essential parameter for multi-layer composites since it may affect the interfacial bonding between
the polymer and reinforcement layers. Surface roughness promotes an enhancement of the interfacial
adhesion and allows greater stress transfer between the matrix and reinforcements, reducing the
capacity of fiber debonding. Therefore, the increase in surface roughness with the addition of MF may
benefit the application of this type of material in multi-layer composites.
AFM also confirms the modification of the surface roughness with the addition of fibers for the
PLA composite samples (Figure 8). As noted for PHA, in PLA films the roughness is also increased,
from 74.4 nm (plain PLA) up to 88.7 and 174.8 nm for 1% CNC and 20% MF samples, respectively.
Once again, the gain in roughness is more notorious for samples reinforced with MF, which can be
translated into a better interlaminar bonding in multi-layer composite applications. In the case of PLA
samples, SEM (Figure 7) reveals that fractured surfaces, even after their necking, are rougher with
the addition of different reinforcements. In Figure 7d, a neat PLA sample of around 2 µm thickness
(cross-section area is around 100 times smaller than the average produced films with 200 µm thickness)
presents a smooth surface, while Figure 6e,f clearly shows rougher surfaces.

3.6. Ultraviolet-Visible Spectroscopy

In the case of UV protective films (UV-PF), the light-resistant capacity of the UV-PF is the main
indicator of packaging materials for UV susceptible products. In this study, fiber-reinforced thin
transparent films produced by solvent casting were investigated by light transmittance analysis
(Figure 9). PLA films showed no UV light transmission in the lower range of UV-C (100–230 nm)
Molecules 2020, 25, 4653 13 of 17

and started to show transmission (up to 40%) in the higher range of UV-C (230–280 nm) (Figure 9a).
From this point, the transmittance of PLA increases to ~90% for UV-B (280–315 nm) and 315–400 nm
(UV-A), which remained constant in the wavelength range of 350–800 nm. The addition of CNC to PLA
films has an insignificant effect on the PLA film transmittance in this region. For wavenumbers below
350 nm, the addition of CNC reduces the transmittance by around 5% for the higher concentration
contents (2 and 3 wt.%). In contrast, all PLA-MF composites showed decreased transmittance across all
Molecules 2020, 25, x FOR PEER REVIEW 13 of 17
wavelengths. This variation in transmittance is related to the presence of UV-absorbing chromophores
in the reinforcing particles of MF. The lignin content of the MF may explain this effect, as observed in
Once again, the gain in roughness is more notorious for samples reinforced with MF, which can be
the thermogravimetric analysis [35]. Lignin is well known to be UV absorbent because of the presence
translated into a better interlaminar bonding in multi-layer composite applications. In the case of
ofPLA
UV-absorbing
samples, SEM chromophores such asthat
(Figure 7) reveals phenolic and ketone
fractured groups
surfaces, even [38–40]. Therefore,
after their necking, MF-reinforced
are rougher
PLA
withfilms were ableoftodifferent
the addition absorb UV, and the higher
reinforcements. In the contents
Figure 7d, aofneat
MF PLA
in thesample
film (10,
of 20 and 302 wt.%),
around μm
the higher the UV absorption of the composite, displaying transmittance values
thickness (cross-section area is around 100 times smaller than the average produced films withof 79.1, 70.3200
and
40.3%, respectively.
μm thickness) presents a smooth surface, while Figure 6e,f clearly shows rougher surfaces.

Molecules 2020, 25, x FOR PEER REVIEW 14 of 17

1.1, 33.1 and 244%, respectively. This UV blocking effect is more evident for shorter wavelengths.
These results
Figure
Figure indicate
8.8. Atomic
Atomic that
Force both PHA
Microscopy andmap
height
height PLAdiagram
map composites
diagram and reinforced
andmean
mean with
roughness
roughness MF of
values
valuespresent
neat
of better
PHA
neat PHA UV-
(a),(a),
protecting
PHA++ 2%
PHA properties compared
CNC (b), PHA ++20%
2% CNC to
20%MF samples
MF(c), with
(c),neat
neatPLA CNC.
PLA(d), PLA+ +
(d),PLA 2%2%CNC
CNC(e)(e)
and PLA
and PLA + 20%
+ 20% MFMF(f). (f).

3.6. Ultraviolet-Visible Spectroscopy

In the case of UV protective films (UV-PF), the light-resistant capacity of the UV-PF is the main
indicator of packaging materials for UV susceptible products. In this study, fiber-reinforced thin
transparent films produced by solvent casting were investigated by light transmittance analysis
(Figure 9). PLA films showed no UV light transmission in the lower range of UV-C (100–230 nm) and
started to show transmission (up to 40%) in the higher range of UV-C (230–280 nm) (Figure 9a). From
this point, the transmittance of PLA increases to ~90% for UV-B (280–315 nm) and 315–400 nm (UV-
A), which remained constant in the wavelength range of 350–800 nm. The addition of CNC to PLA
films has an insignificant effect on the PLA film transmittance in this region. For wavenumbers below
350 nm, the addition of CNC reduces the transmittance by around 5% for the higher concentration
contents (2 and 3 wt.%). In contrast, all PLA-MF composites showed decreased transmittance across
all wavelengths. This variation in transmittance is related to the presence of UV-absorbing
chromophores in the reinforcing particles of MF. The lignin content of the MF may explain this effect,
as observed in the thermogravimetric analysis [35]. Lignin is well known to be UV absorbent because
of the Figure 9. UV-Vis
presence spectrograms of
of UV-absorbing the PHA (a) and
chromophores PLAas
such (b)phenolic
films reinforced with CNC
and ketone and MF.
groups [38–40].
Figure 9. UV-Vis spectrograms of the PHA (a) and PLA (b) films reinforced with
Therefore, MF-reinforced PLA films were able to absorb UV, and the higher the contents of MF in the CNC and MF.
Regarding neat PHA samples presented in Figure 9b, the transmittance values are lower compared
film (10, 20 and 30 wt.%), the higher the UV absorption of the composite, displaying transmittance
to 4.
PLAConclusions
films, with values of around 70% at 600 nm. In contrast to PLA films, the addition of high
values of 79.1, 70.3 and 40.3%, respectively.
The addition
Regarding neat of
PHAMF samples
cellulose-based
presentedfibers is presented
in Figure 9b, theastransmittance
an alternative to reinforcing
values are lower
biopolymers films, offering some advantages compared to the addition of CNC.
compared to PLA films, with values of around 70% at 600 nm. In contrast to PLA films, the addition Despite the higher
ofspecific surface area ofofCNC
high concentrations CNCand, therefore,
(2 and 3 wt.%)theinadvantages
PHA specimens associated
leadswith
to ait,15%
it also reduces
decrease in the
the
possibility ofthrough
transmittance introducing higher
the entire concentrations
spectrum. The major of this type of
difference filler inPHA
between composite
and PLA applications,
films is the
especially in commodity
improvement in the UV ones.
light Moreover, the high
barrier ability production
of the PHA films. cost This
of CNC UValso limits the
blocking application
effect is more
Molecules 2020, 25, 4653 14 of 17

concentrations of CNC (2 and 3 wt.%) in PHA specimens leads to a 15% decrease in the transmittance
through the entire spectrum. The major difference between PHA and PLA films is the improvement in
the UV light barrier ability of the PHA films. This UV blocking effect is more pronounced when MF is
added, in comparison to samples reinforced with CNC. At 600 nm, when MF is added in different
concentrations (10, 20 and 30 wt.%) in PHA, the transmittance is reduced by 1.1, 33.1 and 244%,
respectively. This UV blocking effect is more evident for shorter wavelengths. These results indicate
that both PHA and PLA composites reinforced with MF present better UV-protecting properties
compared to samples with CNC.

4. Conclusions
The addition of MF cellulose-based fibers is presented as an alternative to reinforcing biopolymers
films, offering some advantages compared to the addition of CNC. Despite the higher specific
surface area of CNC and, therefore, the advantages associated with it, it also reduces the possibility
of introducing higher concentrations of this type of filler in composite applications, especially in
commodity ones. Moreover, the high production cost of CNC also limits the application of this type of
filler for high added value purposes, whereas MF from kraft pulp may help to reduce not only the cost
of the filler but the whole composite with the addition of up to 20 wt.% of reinforcement.
Compared to CNC, MF have a better reinforcing effect on PHA samples, increasing both tensile
strength and Young’s modulus. Combined with this improvement of the mechanical performance of
PHA films, the addition of MF causes a notorious increase in the roughness of the films, compared to
both neat PLA and PHA, and composites reinforced with CNC. This added roughness may have
interesting applications in multi-layer composites where improved interlaminar bonding is required.
Moreover, the surface affinity of the films to water may be controlled by the addition of MF and
specific cooling conditions since the crystallinity degree of the film is related to the contact angle with
water. In addition, the use of MF as reinforcement in biopolymer films may be applied in UV-blocking
elements since MF clearly diminishes UV-VIS transmittance in the whole wavelength range.
After this overall characterization of biopolymer films reinforced with MF, a more in-depth
characterization should be addressed in order to better understand the performance of this type
of reinforcement for specific applications. Considering the benefits of both types of reinforcements
explored here, CNC and MF, a possible hierarchical reinforcement can be conducted to understand their
synergic effect on the properties of this type of composite. Moreover, possible treatments on cellulose
fibers would introduce additional functionalities and improved properties to bio-based composites.
Along with the improved mechanical properties of the biopolymers evaluated in this work with
the addition of MF, the use of MF brings an important economic advantage. This reduction in cost
is related to the price of this type of filler. In this work, the authors used commercial kraft paper to
avoid any heterogeneity in the quality of the microfibers and, therefore, to obtain consistent films.
Even using this high-quality raw material, the price of the filler is much cheaper compared with PHA.
Then, this filler, when dosed at 30 wt.%, clearly reduces the cost of the polymer composite. However,
the great advantage of using MF is the possibility of reusing recycled kraft paper once is discarded
from its initial purpose. Recycled kraft paper from a selective waste collection after its processing
could be up to 10 times cheaper than technical grade kraft paper.
It is remarkable that the defibrillation process applied in this study only involves immersion in
water, low-energy mechanical stirring and drying. In the end, in industrial processes, up to 90% of
the energy demand of this process is due to water evaporation. Using efficient drying procedures
such as Through Air Drying (TAD), water removal involves 4800 kJ/kg of water evaporated [41].
This requires around 695 kWh to obtain a ton of dried microfibers. In addition, the water required in
this process could be used in a loop cycle after decontamination (impurities, oil and pigments) through
straightforward flocculation/filtration techniques.

Author Contributions: Conceptualization, G.M. and R.F.; methodology, G.M.; validation, G.M., C.G. and R.F.;
formal analysis, G.M.; investigation, G.M.; resources, R.F.; data curation, C.G.; writing—original draft preparation,
Molecules 2020, 25, 4653 15 of 17

G.M.; writing—review and editing, C.G.; visualization, G.M.; supervision, R.F.; project administration, R.F.;
funding acquisition, R.F. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by TSSiPRO; NORTE-01-0145-FEDER-000015-project, Technologies for
Sustainable and Smart Innovative Products, which involves this research work and financed by the European
Regional Development Fund (ERDF) through the Support System for Scientific and Technological Research
(Structured R & D & I Projects) of the Regional Operational Program for Northern Portugal 2020.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Shekar, H.S.; Ramachandra, M. Green composites: A review. Mater. Today Proc. 2018, 5, 2518–2526. [CrossRef]
2. Ramesh, M.; Palanikumar, K.; Reddy, K.H. Plant fibre based bio-composites: Sustainable and renewable
green materials. Renew. Sustain. Energy Rev. 2017, 79, 558–584. [CrossRef]
3. Christian, S.; Billington, S.L. Mechanical response of PHB and cellulose acetate natural fiber-reinforced
composites for construction applications. Compos. Part B Eng. 2011, 42, 1920–1928. [CrossRef]
4. Zhang, Q.; Shi, L.; Nie, J.; Wang, H.; Yang, D. Study on poly (lactic acid)/natural fibers composites. J. Appl.
Polym. Sci. 2012, 125, E526–E533. [CrossRef]
5. Saba, N.; Jawaid, M.; Al-Othman, O. An overview on polylactic acid, its cellulosic composites and applications.
Curr. Org. Synth. 2017, 14, 156–170. [CrossRef]
6. Mathuriya, A.S.; Yakhmi, J.V.; Martínez, L.M.T.; Kharissova, O.V.; Kharisov, B.I. Polyhydroxyalkanoates:
Biodegradable plastics and their applications. In Handbook of Ecomaterials; Springer: Cham, Switzerland,
2017; pp. 1–29.
7. Bugnicourt, E.; Cinelli, P.; Lazzeri, A.; Alvarez, V. Polyhydroxyalkanoate (PHA): Review of synthesis,
characteristics, processing and potential applications in packaging. Express Polym. Lett. 2014, 8, 791–808.
[CrossRef]
8. Chen, G.-Q.; Chen, X.-Y.; Wu, F.-Q.; Chen, J.-C. Polyhydroxyalkanoates (PHA) toward cost competitiveness
and functionality. Adv. Ind. Eng. Polym. Res. 2020, 3, 1–7. [CrossRef]
9. Koay, S.C.; Yeng, C.M.; May, C.P.; Yeow, T.K.; Kiat, O.T.; How, C.K. Effect of coupling agent content on
properties of composites made from polylactic acid and chrysanthemum waste. J. Vinyl Addit. Technol. 2019,
26, 10–16. [CrossRef]
10. Seggiani, M.; Cinelli, P.; Verstichel, S.; Puccini, M.; Vitolo, S.; Anguillesi, I.; Lazzeri, A. Development of
fibres-reinforced biodegradable composites. Chem. Eng. Trans. 2015, 1813–1818. [CrossRef]
11. Nery, T.B.R.; Dos Santos, Z.I.G.; José, N.M. Desenvolvimento e caracterização de biocompósitos de
polihidroxibutirato e fibra de bananeira. Matéria Rio Jan. 2018, 23. [CrossRef]
12. Machado, M.L.C.; Pereira, N.C.; De Miranda, L.F.; Terence, M.C.; Pradella, J.G.C. Estudo das propriedades
mecânicas e térmicas do polímero Poli-3-hidroxibutirato (PHB) e de compósitos PHB/pó de madeira. Polímeros
2010, 20, 65–71. [CrossRef]
13. Shojaeiarani, J.; Bajwa, D. Functionalized Cellulose Nanocrystals for Improving the Mechanical Properties of Poly
(Lactic Acid); American Society of Mechanical Engineers: Pittsburgh, PA, USA, 2018; Volume 12.
14. Mokhena, T.C.; Sefadi, J.S.; Sadiku, E.; John, M.J.; Mochane, M.J.; Mtibe, A. Thermoplastic processing of
PLA/cellulose nanomaterials composites. Polymers 2018, 10, 1363. [CrossRef] [PubMed]
15. Zheng, T.; Zhang, Z.; Shukla, S.; Agnihotri, S.; Clemons, C.M.; Pilla, S. PHBV-graft-GMA via reactive extrusion
and its use in PHBV/nanocellulose crystal composites. Carbohydr. Polym. 2019, 205, 27–34. [CrossRef]
[PubMed]
16. Patrício, P.S.D.O.; Pereira, F.V.; Dos Santos, M.C.; De Souza, P.P.; Roa, J.P.B.; Orefice, R.L. Increasing the
elongation at break of polyhydroxybutyrate biopolymer: Effect of cellulose nanowhiskers on mechanical
and thermal properties. J. Appl. Polym. Sci. 2012, 127, 3613–3621. [CrossRef]
17. Trache, D.; Hussin, M.H.; Haafiz, M.K.M.; Mishra, Y.K. Recent progress in cellulose nanocrystals: Sources and
production. Nanoscale 2017, 9, 1763–1786. [CrossRef] [PubMed]
18. Seoane, I.T.; Manfredi, L.B.; Cyras, V.P.; Torre, L.; Fortunati, E.; Puglia, D. Effect of cellulose nanocrystals and
bacterial cellulose on disintegrability in composting conditions of plasticized PHB nanocomposites. Polymers
2017, 9, 561. [CrossRef]
Molecules 2020, 25, 4653 16 of 17

19. Rigotti, D.; Checchetto, R.; Tarter, S.; Caretti, D.; Rizzuto, M.; Fambri, L.; Pegoretti, A. Polylactic acid-lauryl
functionalized nanocellulose nanocomposites: Microstructural, thermo-mechanical and gas transport
properties. Express Polym. Lett. 2019, 13, 858–876. [CrossRef]
20. Deetuam, C.; Samthong, C.; Choksriwichit, S.; Somwangthanaroj, A. Isothermal cold crystallization kinetics
and properties of thermoformed poly (lactic acid) composites: Effects of talc, calcium carbonate, cassava starch
and silane coupling agents. Iran. Polym. J. 2020, 29, 103–116. [CrossRef]
21. Singh, M.; Kumar, P.; Ray, S.; Kalia, V.C. Challenges and opportunities for customizing polyhydroxyalkanoates.
Indian J. Microbiol. 2015, 55, 235–249. [CrossRef]
22. Oksman, K.; Skrifvars, M.; Selin, J.-F. Natural fibres as reinforcement in polylactic acid (PLA) composites.
Compos. Sci. Technol. 2003, 63, 1317–1324. [CrossRef]
23. Torres-Tello, E.V.; Robledo-Ortíz, J.R.; González-García, Y.; Pérez-Fonseca, A.A.; Jasso-Gastinel, C.F.;
Mendizábal, E. Effect of agave fiber content in the thermal and mechanical properties of green composites
based on polyhydroxybutyrate or poly (hydroxybutyrate-co-hydroxyvalerate). Ind. Crops Prod. 2017,
99, 117–125. [CrossRef]
24. Graupner, N.; Herrmann, A.S.; Müssig, J. Natural and man-made cellulose fibre-reinforced poly (lactic acid)
(PLA) composites: An overview about mechanical characteristics and application areas. Compos. Part A Appl.
Sci. Manuf. 2009, 40, 810–821. [CrossRef]
25. Barkoula, N.; Garkhail, S.; Peijs, T. Biodegradable composites based on flax/polyhydroxybutyrate and its
copolymer with hydroxyvalerate. Ind. Crops Prod. 2010, 31, 34–42. [CrossRef]
26. Di Vona, M.L. Annealing of polymer membranes. In Encyclopedia of Membranes; Drioli, E., Giorno, L., Eds.;
Springer: Berlin/Heidelberg, Germany, 2014; pp. 1–2.
27. Kumar, R.; Kumari, S.; Rai, B.; Das, R.; Kumar, G. Effect of nano-cellulosic fiber on mechanical and barrier
properties of polylactic acid (PLA) green nanocomposite film. Mater. Res. Express 2019, 6, 125108. [CrossRef]
28. Gupta, A.; Simmons, W.; Schueneman, G.T.; Hylton, D.; Mintz, E.A. Rheological and thermo-mechanical
properties of poly (lactic acid)/lignin-coated cellulose nanocrystal composites. ACS Sustain. Chem. Eng. 2017,
5, 1711–1720. [CrossRef]
29. Sato, H.; Murakami, R.; Padermshoke, A.; Hirose, F.; Senda, K.; Noda, I.; Ozaki, Y. Infrared spectroscopy
studies of CH···O hydrogen bondings and thermal behavior of biodegradable poly (hydroxyalkanoate).
Macromolecules 2004, 37, 7203–7213. [CrossRef]
30. Socrates, G. Infrared and Raman Characteristic Group Frequencies: Tables and Charts, 3rd ed.; Wiley: Chichester,
UK, 2010.
31. Zhang, J.; Sato, H.; Noda, I.; Ozaki, Y. Conformation rearrangement and molecular dynamics of poly
(3-hydroxybutyrate) during the melt-crystallization process investigated by infrared and two-dimensional
infrared correlation spectroscopy. Macromolecules 2005, 38, 4274–4281. [CrossRef]
32. Padermshoke, A.; Katsumoto, Y.; Sato, H.; Ekgasit, S.; Noda, I.; Ozaki, Y. Melting behavior of poly
(3-hydroxybutyrate) investigated by two-dimensional infrared correlation spectroscopy. Spectrochim. Acta
Part A Mol. Biomol. Spectrosc. 2005, 61, 541–550. [CrossRef]
33. Brizuela, A.B.; Bichara, L.C.; Romano, E.; Yurquina, A.; Locatelli, S.; Brandán, S.A. A complete characterization
of the vibrational spectra of sucrose. Carbohydr. Res. 2012, 361, 212–218. [CrossRef]
34. Borkotoky, S.S.; Dhar, P.; Katiyar, V. Biodegradable poly (lactic acid)/Cellulose nanocrystals (CNCs) composite
microcellular foam: Effect of nanofillers on foam cellular morphology, thermal and wettability behavior.
Int. J. Biol. Macromol. 2018, 106, 433–446. [CrossRef]
35. Bai, J.; Chen, X.; Shao, J.; Jia, C.; Wang, Q. Study of breakage of main covalent bonds during co-pyrolysis of
oil shale and alkaline lignin by TG-FTIR integrated analysis. J. Energy Inst. 2019, 92, 512–522. [CrossRef]
36. Ojha, N.; Das, N.; Ojhaa, N. A Statistical approach to optimize the production of Polyhydroxyalkanoates
from Wickerhamomyces anomalus VIT-NN01 using response surface methodology. Int. J. Biol. Macromol.
2018, 107, 2157–2170. [CrossRef] [PubMed]
37. Zou, H.; Yi, C.; Wang, L.; Liu, H.; Xu, W. Thermal degradation of poly (lactic acid) measured by
thermogravimetry coupled to Fourier transform infrared spectroscopy. J. Therm. Anal. Calorim. 2009,
97, 929–935. [CrossRef]
38. Lanzalunga, O.; Bietti, M. Photo and radiation chemical induced degradation of lignin model compounds.
J. Photochem. Photobiol. B Biol. 2000, 56, 85–108. [CrossRef]
Molecules 2020, 25, 4653 17 of 17

39. Park, S.Y.; Kim, J.-Y.; Youn, H.J.; Choi, J.W. Utilization of lignin fractions in UV resistant lignin-PLA
biocomposites via lignin-lactide grafting. Int. J. Biol. Macromol. 2019, 138, 1029–1034. [CrossRef]
40. Qian, Y.; Qiu, X.; Zhu, S. Lignin: A nature-inspired sun blocker for broad-spectrum sunscreens. Green Chem.
2015, 17, 320–324. [CrossRef]
41. Stenström, S. Drying of paper: A review 2000–2018. Dry. Technol. 2019, 38, 825–845. [CrossRef]

Sample Availability: Samples of the compounds made out of PHA and PLA are available from the authors.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).