Nanocellulose-tannin films: From trees to sustainable active packaging

Accepted Manuscript Nanocellulose-tannin films: From trees to sustainable active packaging André L. Missio, Bruno D. Mattos, Daniele de F. Ferreira, Washington L.E. Magalhães, Daniel A. Bertuol, Darci A. Gatto, Alexander Petutschnigg, Gianluca Tondi PII:

S0959-6526(18)30527-4

DOI:

10.1016/j.jclepro.2018.02.205

Reference:

JCLP 12152

To appear in:

Journal of Cleaner Production

Received Date: 5 September 2017 Revised Date:

18 January 2018

Accepted Date: 20 February 2018

Please cite this article as: Missio AndréL, Mattos BD, Ferreira DdF, Magalhães WLE, Bertuol DA, Gatto DA, Petutschnigg A, Tondi G, Nanocellulose-tannin films: From trees to sustainable active packaging, Journal of Cleaner Production (2018), doi: 10.1016/j.jclepro.2018.02.205. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Nanocellulose + Tannin gel

Tannin + Nanocellulose

Connection between tannin and nanocellulose

EP

TE D

M AN U

SC

RI PT

Cellulose

AC C

Tannin

Active Packaging

- Antioxidant capacity - Barrier properties - Chemical resistance (min)

ACCEPTED MANUSCRIPT

Nanocellulose-tannin films: from trees to sustainable active packaging

André L. Missioa; Bruno D. Mattosb; Daniele de F. Ferreirac; Washington L. E. Magalhãesb,d;

a

RI PT

Daniel A. Bertuole; Darci A. Gattoa; Alexander Petutschniggf; Gianluca Tondif*

Laboratório de Produtos Florestais (PPGEF), Centro de Ciências Rurais, Universidade

Federal de Santa Maria, P.O. Box 221, ZIP code 97105-900. Santa Maria, Brazil.

Programa de Pós-Graduação em Engenharia e Ciência dos Materiais (PIPE), Universidade

SC

b

Federal do Paraná, Centro Politécnico, P.O Box 19011, ZIP code 81531-990, Curitiba, Brazil. Departamento de Tecnologia e Ciência dos Alimentos (PPGCTA), Universidade Federal de

M AN U

c

Santa Maria, P.O. Box 221, ZIP code 97105-900. Santa Maria, Brazil. d

Centro Nacional de Pesquisas Florestais - Embrapa Florestas, Estrada da Ribeira, Km 111 -

Bairro Guaraituba, P.O. Box 319 – ZIP code 83411-000 - Colombo, Brazil. de Processos Ambientais (LAPAM), Departamento de Engenharia Química,

TE D

e Laboratório

Universidade Federal de Santa Maria, ZIP code 97105-900. Santa Maria, Brazil. f

Salzburg University of Applied Sciences, Forest Products Technology and Timber

EP

Construction Department, Marktstraße 136a, 5431 Kuchl, Austria. *Corresponding author.

AC C

[email protected]

1

ACCEPTED MANUSCRIPT ABSTRACT Cellulose nanofibrils and condensed tannins were chosen to prepare a strong, sustainable packaging material. Cellulose nanofibrils provided the physical and mechanical requirements, while the tannin was incorporated due to its antioxidant properties. Herein, an easy, one-step

RI PT

method was designed to prepare a film containing 190 mg/g of active molecules. The incorporation of tannin into the cellulose matrix was carried out through the mechanical

fibrillation of cellulose pulp and tannin mixtures. The tannin-incorporated cellulose films

SC

presented high density and improved surface hydrophobicity, which resulted in a 6-fold

enhancement in their air-barrier properties. A slow release of antioxidant components was

M AN U

verified upon soaking, as well as good resistance to several solvents. The mechanical features of the cellulose matrix were not significantly affected by the incorporation of such phenolic molecules. These properties are key factors to put forward the utilization of tannin-added films as a valid green, nontoxic packaging material for food and pharmaceutical products.

1. Introduction

TE D

KEYWORDS: polyphenols; NFC; Acacia mearnsii; wettability, shelf-life, preservative.

EP

Almost all the items in a super market are packaged. Packaging plays a fundamental role in the protection of goods from physical damage, external contamination and

AC C

deterioration (Wang and Wang, 2017). Synthetic plastics such as low-density polyethylene (LDPE) and linear low-density polyethylene (LLDPE) are the favored materials for packaging as they are light, inert and easily available. As such, the production of plastics has increased exponentially over the last 65 years (Bernstad Saraiva et al., 2016). In Europe, for instance, the production of oil-derived plastics reached 49 million of T/y, almost 40% of which is used for packaging purposes (Plastic, 2017). Traditional packaging materials are becoming outdated, especially in the food sector, where there is an increased interest in the interaction between the packaging material and the product. Nowadays, besides the traditional physic2

ACCEPTED MANUSCRIPT mechanical resistance and barrier features, new functionalities are required to extend the shelf-life or the expiration date of the products. Safer packaged food can be addressed if these new materials, called “active packaging”, were used to absorb unwanted substances such as heavy-metals or exhausted oils, and to protect against oxidation, UV and moisture. In special

preservatives or antioxidants on-demand (EU, 2009).

RI PT

cases, active packaging material can be designed to deliver protective chemicals like

In the last few years, research in the field of active/sustainable packaging has

SC

increased significantly (Beitzen-Heineke et al., 2017). Functional materials prepared using polycarbonate and polyethylene added of phenolic substances (Krepker et al., 2017) or

M AN U

inorganic compounds of copper (Al-Naamani et al., 2016) and zinc (De Vietro et al., 2017) have been successfully presented as antimicrobial packages. Very recently, bio-based films based on carboxymethylcellulose - chitosan - oleic acid added of zinc oxide (ZnO) were considered suitable for prolonging the shelf-life of sliced wheat bread (Noshirvani et al.,

TE D

2017). Future trends indicate the frontier of research in this field moving towards the preparation of active packaging that uses green and sustainable resources. Herein, forestbased structures and molecules were explored in order to prepare green, renewable active

EP

packaging.

Cellulose nano- and microfibrils obtained from wood are strong, lightweight,

AC C

chemically versatile, non-toxic and biodegradable (Dufresne, 2013). Its production is estimated at 1011 T/y, making it the most abundantly available bio-macromolecule on Earth (Douglass et al., 2016). Cellulose’s peerless properties are fundamental for a broad range of applications that go from green composite films (Duan et al., 2016) to optical devices (Cai et al., 2016). Nanometric scaled cellulose fibrils, namely nanofibrillated cellulose (CNF or NFC), are produced through simple, feasible mechanical grinding procedures. CNF chains present diameter varying from 5 to 60 nm with length of a few micrometers, making it a flexible and strong matrix (Hubbe et al., 2017). CNF can be assembled in films with excellent 3

ACCEPTED MANUSCRIPT modulus of elasticity (Lee et al., 2012) and good barrier properties (Hubbe et al., 2017), and therefore, they have been successfully proposed as bio-based packaging material (Lavoine et al., 2016). In parallel, tannin provides an excellent opportunity to improve the activeness of cellulose films. Condensed tannins are the most abundantly extracted natural substances on

RI PT

Earth, with ca. 200 thousand tons extracted each year (Pizzi, 2008). The bark of Acacia mearnsii, commonly known as black wattle or mimosa, is the most important source of

commercially available condensed tannins (Duval and Avérous, 2016). Robinetinidol is the

SC

basic monomeric unit constituent of this extract (Fig. 1) that polymerizes in a branched

network to form an oligomer of prorobinetinidin (Pizzi, 1994). These polyhydroxyphenols

M AN U

have a strong antioxidant (Missio et al., 2017), which makes them an excellent candidate to incorporate antioxidant properties into cellulose films.

Therefore, the goal of this research is to prepare tannin-incorporated CNF films and then characterize their chemical, mechanical and physical properties. Two methodologies

TE D

were investigated to incorporate tannin into the cellulose matrix: (i) addition of tannin in the nanocellulose gel to further filter it and (ii) co-grinding of the components followed by filtering. The features of the obtained mixtures were investigated by transmission electron

EP

microscopy, while the derived films were thoroughly investigated using Fourier transform infrared spectroscopy, scanning electron microscopy, apparent contact angle, air-permeability,

AC C

thermogravimetric analysis, antioxidant activity, tensile strength and chemical resistance.

2. Material and methods 2.1.Raw materials

The active films were prepared using nanofibrillated cellulose as the structural matrix material and wood tannin as the source of bioactive molecules. The nanofibrillated cellulose was prepared according to previous work (Kumode et al., 2017). Briefly: bleached Kraft pulp (from Suzano Celulose e Papel®) was turned into fluffy-like cellulose by using a high-shear 4

ACCEPTED MANUSCRIPT mixer MH-100. Then, a 2 wt.% cellulose aqueous suspension was passed through an ultrafine friction grinder (Masuko Sangyo Super Masscollider) at 1,500 rpm until a homogeneous cellulose hydrogel was obtained. The tannin was obtained from Acacia mearnsii bark (from SETA® industrial process), and is composed by 75% of the condensable tannins and 25% of

RI PT

hydrocolloid gums, sugars and other small molecules such as the flavan-3-ol (Arbenz and Averous, 2015).

6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), 2,2’-azobis(2-

SC

amidinopropane) dihydrochloride (AAPH) and sodium fluorescein were purchased from

Sigma-Aldrich. Monobasic and dibasic potassium phosphate were purchased from Vetec.

M AN U

Ultrapure MiliQ water (18.2 MΩ·cm, Millipore Corp., Bedford, MA) was used in the antioxidant study.

2.2. Incorporation of tannin into the nanocellulose matrix

TE D

The production process for the nanocellulose films with and without tannin is summarized in Fig. 1a. Two strategies to incorporate tannin into the nanocellulosic matrix were tested. In the first (i), tannin powder was manually mixed with the nanocellulose

EP

dispersion at a tannin-to-cellulose ratio equal to 5, and then the mixture was stirred for 5 min. In the second (ii), tannin was added in the initial dispersion of fluffy cellulose at the same

AC C

ratio in (i), followed by mechanical fibrillation of the mixture. In this process (ii) the formation of cellulose nanofibrils occurred in the presence of tannin. A filtration system (Whatman®) with a nylon membrane (diameter = 47 mm; porosity

= 0.22 µm) connected with a vacuum-pump was used to produce the films. Neat nanocellulose (CNF) and nanocellulose-tannin (CNF-T) obtained by process (ii) were used for the preparation of the films. 1.4 g of the dispersions were mixed with 50 mL of distilled water and deposited on the nylon membrane in order to obtain films with a targeted grammage of 30 g.m-2 and diameter of 35 mm. The films were air-dried for 4 hours, removed 5

ACCEPTED MANUSCRIPT from the membrane and finally stored in climate chamber at 20°C and 65% of relative humidity.

2.3.Morphology studies

RI PT

The morphologic features of the tannin-added nanocellulose dispersions were investigated using a transmission electron microscope (TEM) JEOL, model JEM-1200 EXII. For this, the dispersions were diluted to 0.1% w/v, dispersed in an ultrasonic bath for 30 min,

SC

and then casted on a copper TEM sample grid. In addition, the films’ surface was observed by scanning electron microscopy (SEM) using an accelerating voltage of 5 kV (Tescan, VEGA-

ensure the conductivity of the sample.

2.4.Density measurement

M AN U

3SBU, Czech Republic). For the SEM images, all samples were coat-sputtered with Au to

TE D

The density of the films was calculated by taking the weight and volume of 10 films per formulation. Weight was obtained in an analytic balance. The diameter of the films was obtained using a digital caliper, and the thickness was measured using the procedures

EP

described in T 411 om-97 (TAPPI, 1997).

AC C

2.5.FT-IR analysis

The films were scanned with a Perkin-Elmer Frontier infrared spectrometer equipped

with an attenuated total reflectance (ATR) Miracle diamond crystal. Each sample was scanned 3 times, registering the spectra in the spectral range between 4000 and 600 cm-1 applying 16 scans with a resolution of 4 cm-1. The spectra were averaged and evaluated after baseline correction and area normalization.

2.6.Thermogravimetric analysis (TGA) 6

ACCEPTED MANUSCRIPT TG curves were obtained in a Setaram equipment - SetSys Evolution model. The samples (ca. 8 mg) were heated from 30 to 600°C at a constant heating rate of 15°C.min-1 under argon flow rate of 20 ml.min-1. The mass loss (TG) and the first derivative (DrTG)

RI PT

curves were analyzed.

2.7.Wettability measurements

The wettability of the films was investigated in a Drop Shape Analyzer – DSA25

SC

equipment, using the sessile drop method. The water contact angle was calculated from 5 to 60 seconds after the casting of a 20 µL droplet of deionized water on the surface of the

M AN U

samples. The measurements were replicated 10 times and the results were reported as averages with their relative standard deviation.

2.8.Air permeability barrier

TE D

Air permeability of the cellulose and tannin-incorporated films was measured using a Lorentzen & Wettre SE 114- Bendtsen instrument, according to ISO 5636-3 normative. The measurement area was 5 cm2. The presented results are the averages of triplicated tests. White

AC C

materials.

EP

paper (70 g.cm-1) and common polypropylene-based packages were also tested as comparison

2.9.Antioxidant capacity The determination of the antioxidant activity of the films was carried out according to

the method described by Kaya et al. (2017). 6-hydroxy-2,5,7,8-tetramethylchroman-2carboxylic acid (Trolox), 2,2’-azobis(2-amidinopropane) dihydrochloride (AAPH) and sodium fluorescein were used. Monobasic and dibasic potassium phosphate were used to prepare the buffer (pH 7.4), as well as working solutions of AAPH, Trolox and sodium fluorescein. Ultrapure water was obtained using a MilliQ system. The fluorescence 7

ACCEPTED MANUSCRIPT monitoring for Oxygen Radical Absorbance Capacity (ORAC) assay was carried out using a microplate reader (Sense, HIDEX, Turku, Finland). The antioxidant capacity was evaluated in triplicate by the ORAC method (Ou et al., 2001). For this analysis, 100 mg of the film were submerged in 10 ml of MilliQ water and the

RI PT

antioxidant capacity was measured after 1, 4, 6, 8, 12, 24, 36 and 48 hours. The samples and Trolox were diluted in a potassium phosphate buffer (75 mmol.L-1) at concentration of 50 mg.L-1 of sample and the Trolox at different concentrations (4 to 98 µM). 25 µL of each

SC

sample and Trolox solutions were added to a black microplate (96 wells), as well as 150 µL of fluorescein (81 nmol.L-1), followed by incubation at 37°C for 10min. Right after, 25 µL of

M AN U

AAPH solution (152 mmol.L-1) were added as a peroxyl radical generator. The fluorescence was monitored (excitation and emission wavelengths were 485 ± 10 nm and 535 ± 20 nm) in a microplate reader with 120 measuring cycles during 120 min at 37°C. The ORAC values expressed as equivalent to µmol of Trolox per g of sample were based on the Area Under the

TE D

Curve (AUC) of fluorescence decay over time, calculated as described in Equation 1: AUC=1+ f1ൗf0 + f2ൗf0 + f3ൗf0 +… fnൗf0

Equation 1.

EP

in which ݂0 is initial fluorescence (t = 0) and ݂݊ is the fluorescence obtained in a read cycle.

2.10. Tensile strength, leaching tests and chemical resistance

AC C

The tensile strength of films (sizing 30x10 mm2) was measured using a Zwick/Roell Z

250 universal testing machine equipped with paper tensile test apparatus at the rate of 2 mm.min-1 according to the DIN EN31 (DIN, 1993). Five measurements per formulation were tested and reported as average with their relative standard deviation. The chemical resistance of the films was investigated by dipping the films in different chemicals for 7 days according to ASTM D543-14 (ASTM, 2014). Distilled water, ethanol, ethyl acetate, hexane, vegetable corn oil, sodium hydroxide (10%) and sulfuric acid (30%)

8

ACCEPTED MANUSCRIPT were used to test chemical resistance. Sample weight before and after the soaking time was taken in order to quantify the amount of tannin (or cellulose) leached or decomposed during the experiment. Tensile strength measurements were used to quantify the chemical resistance.

RI PT

All tests were performed in triplicates.

3. Results and Discussion

3.1.Evaluation of the nanocellulose-tannin interactions in the film formation process

SC

and resulting material.

The incorporation of tannin into the nanocellulose network was performed using two

M AN U

different procedures as represented in Fig. 1a. The addition of tannin in the cellulose dispersion through simple stirring resulted in a not-homogeneous mixture. It was possible to observe several spherical tannin agglomerates, completely different from the cellulose nanofibrils (Fig. 1b). On the other hand, when the cellulose pulp was ultra-fine ground in the

TE D

presence of tannin (Fig. 1a, process ii) an intimate interface interaction between the tannin and the cellulose nanofibrils can be observed (Fig. 1c). These two procedures presented dramatically different behavior upon filtration. The mixtures prepared through single stirring

EP

resulted in a complete loss of all added tannin, resulting in a white film (negligible tannin content). Conversely, the ultra-fine ground mixtures resulted in a brown film with a high mass

AC C

content of tannin. Hence, the joint ultra-fine grinding process of tannin and cellulose played a fundamental role for a successful intimate interconnection between the two forest-based resources. Upon mechanical processing, the freshly exposed nanofibril surfaces immediately form H-bonds with water (Douglass et. al., 2016). Hypothetically, if tannin is added after that, the layer of water onto cellulose surface hinders possible H-bonding between cellulose and tannin. However, when the new nanofibril surfaces are exposed in the presence of tannin, Hbonds at the tannin-cellulose interface can also occur. This phenomenon could explain why the mechanical process of tannin-cellulose dispersion effectively incorporated tannin in the 9

ACCEPTED MANUSCRIPT cellulose matrix, while single stirring did not. The resulting brown film, called nanocellulosetannin film (CNF-T), was the subject of this work and was compared with a pure

EP

TE D

M AN U

SC

RI PT

nanocellulose (CNF) film.

Fig. 1. Scheme of the systematic preparation of the tannin-incorporated nanocellulose films,

AC C

with detail of the basic unit of prorobinetinidin chemical structure, the most abundant repeating unit in the Mimosa tannin extract (a). Transmission electron microscopy images show that the lack of interaction between tannin and cellulose nanofibrils resulted from a simple stirring procedure (b), and the improved interaction between tannin and cellulose nanofibrils after mechanical processing (c).

The amount of tannin incorporated in the nanocellulose matrix, following process (ii), was estimated by a gravimetric approach using 10 films to obtain a valuable average. The 10

ACCEPTED MANUSCRIPT pure nanocellulose films weighed 24.4 ± 1.2 mg, while the tannin-incorporated films weighed 30.3 ± 1.0 mg. The difference in mass was attributed to the amount of tannin embedded in the film, which was estimated to be ca. 196 mg tannin by g of film. Multiple interactions are expected to happen between the cellulose surface and the

RI PT

phenolic-based molecules in the tannin. These interactions can be different at the molecular or macromolecular levels. The basic tannin moieties, the phenolics, present significant H-

bonding capacity, enabling it to interact strongly with practically any hydrophilic substrate

SC

(Ejima et al. 2013). For instance, using gallic acid as a model molecule of the tannin basic moieties, Guo et al. (2016) stated that multiple H-bonding and quadrupolar interactions can

M AN U

take place between tannin and hydrophilic surfaces (Fig 2a). In addition, the electrostatic potential mapping of the macromolecular state of tannin at equilibrium conformation (represented here as a typical linear polymerization of prorobinetinidin, Fig. 2b) shows a decisively positive structure, with some isolated negative areas. Thus, besides H-bonding and

AC C

EP

cellulose surfaces.

TE D

quadrupolar, electrostatic interactions are favored between tannin and negatively charged

Fig. 2. Expected interactions that can take place between the basic tannin moieties (a) or the tannin macromolecules (b) and the hydrophilic, negatively charged cellulose surface.

11

ACCEPTED MANUSCRIPT Although several physical interactions can take place, covalent bonds between both unmodified tannin and cellulose are not expected. FT-IR analysis was applied to investigate the chemical structures of both components and mixtures (Fig 3.) The spectrum of the

RI PT

nanocellulose film (CNF) presents typical absorptions in the regions between 1,440-1,400 and 1,390-1,290 cm-1 which are attributed to C-H bending and wagging (Chung et al., 2004). A broad complex band at 1,200-900 cm-1 corresponds to the C-O and C-O-C stretching of the

SC

carbohydrates (Schwanninger et al., 2004). The typical, multiple hydroxyl bonds are

observed at 3,300 cm-1. The spectrum of the tannin-containing film (CNF-T) shows a very

M AN U

similar trend, but there was an increase in the peaks at 1,605, 1,510 and 1,450 cm-1, which are assigned to aromatic/phenolic vibrations of the mimosa tannin (Tondi and Petutschnigg, 2015). As expected, tannin is principally embedded in the cellulose net but only few or no covalent bonds are observed between the tannin and the nanocellulose. The only part of the

TE D

spectra which does not exactly follow the expected trend is the C-O-C region between 1,400 and 1,200 cm-1. Here, the CNF-T presents slightly weaker/shifted signals that could indicate the presence of few ether bonds. However, several secondary interactions could have

EP

happened between tannin and cellulose. Van der Waals, short-range dipole-dipole (through quadrupole) or even multiples H-bonding are possible interactions (Fig. 2) at the tannin-

AC C

cellulose interface that would not have been tracked in the FT-IR analysis.

12

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 3. FT-IR spectra of the films of nanocellulose with (orange line) and without tannin (grey line). A control spectrum of the tannin powder (black line) was taken for comparison.

surface level.

TE D

3.2. Evaluation of the barrier properties of the films and water interactions at the

All films presented similar morphologies at both low and high magnifications (Fig. 4).

EP

The SEM images taken from the films’ surfaces showed a very compact, nonporous material at a micrometric scale. The compactness of the films resulted in high densities for both films.

AC C

The values were slightly higher for the tannin-incorporated film (1.50 ± 0.15 g.cm-³) when compared with that of the pure nanocellulose films (1.44 ± 0.21 g.cm-3). The densities of the films are comparable with the density of the cellulose nanofibrils, which is ca. 1.5-1.6 g.cm-3 (Dufresne, 2013), meaning that the designed films have low porosity, high compactness and, as consequence, high resistance against air permeation is expected (Kisonen et al., 2015). Nanofibrillated cellulose films are materials with very high barrier against air (Lavoine et al., 2012), as the nanofeatures of the fibrils increase the surface contact between single fibrils optimizing their interlocking upon filtration. CNF and CNF-T films underwent 13

ACCEPTED MANUSCRIPT air-barrier tests and the results are reported in Table 1. The permeability of the CNF and CNF-T films is 14-fold lower than that registered for regular A4 paper (70 g.m-2). The addition of tannin abated the permeability up to 3.1 mL min-1, more than 6 times less than the pure CNF film. This value is close to that of the polypropylene plastic package for chocolate

RI PT

that perfectly seals the product so that no air is permeated. Regarding barrier properties, it is expected that the tannin incorporation into a cellulose matrix puts forward the utilization of sustainable packaging material for near-future commercial package as substitutes of synthetic,

AC C

EP

TE D

M AN U

SC

pollutant materials.

Fig. 4. SEM observation of the nanocellulose films’ surface without (a, c) and with the incorporation of tannin (b, d).

14

ACCEPTED MANUSCRIPT Table 1. Air barrier properties of CNF and CNF-T in comparison with commercial packaging materials. Air permeability (mL.min-1) Packaging materials

19.27 ± 1.5

CNF-T

3.1 ± 1.1

Regular paper

277.2 ± 9.7

PP-based package

0.0

SC

CNF

RI PT

Average ± Standard deviation

M AN U

Good barrier properties are characterized by high densities, but for nanocellulose films, a limitation is represented by its high affinity with water (Hubbe et al., 2017). The CNF-T films have shown decreased surface wettability, which was accessed through apparent contact angle measurements (Fig. 5). The incorporation of tannin rendered the cellulosic film

TE D

higher hydrophobicity, since the apparent contact angle was higher than 90° during the first minute of measurement. A graphical representation (Fig. 5b) shows that the CNF-T films do not promote significant adsorption of the water droplet, which allowed the deposition of

EP

several droplets at the same time. The hydrophobization of the surface can be explained by the interactions taking place at the tannin-cellulose interface (Fig. 2). Tannin is macro

AC C

molecularly structured with both hydrophilic and hydrophobic domains. The adsorption of tannin on the cellulose surface is driven by their hydrophilic domains via H-bonding. In that case, the hydrophobic domains of the tannin macromolecules are the ones exposed at the very top molecular surface. As the wettability is a surface phenomenon, even if the tannin is leachable upon water soaking, the water contact angle can be high. The high density of the film combined with the enhanced hydrophobicity, placed the tannin-cellulose films as a very promising candidate for both water and air protection (Lavoine et al., 2012).

15

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 5. Apparent contact angle of water on the nanocellulose films containing or not tannin (a). Graphical representation of several droplets on the CNF-T films, showing that no water

TE D

adsorption (or negligible) happens on these films for 1 minute (b).

3.3. Thermogravimetric analysis

EP

Regardless of the water desorption, the thermal degradation of the nanocellulose film

AC C

occurred in a single thermal event (Fig. 6a). The mass loss related to the structural decomposition of the cellulose film took place at 340 °C, with onset temperature at 250 °C. From this temperature onward, dehydration of the cellulose happens, saturated carbonaceous structures are formed, and a small fraction of charcoal is produced (Lengowski et al., 2016). The thermal degradation pathway observed for the pure tannin was completely different from the cellulose. The mass loss started at an earlier temperature (150 °C) and kept decreasing with a low mass loss rate until 200 °C, where another thermal degradation event occurred. It was also observed that at higher temperatures (>350°C) the tannin imparts a certain

16

ACCEPTED MANUSCRIPT degradation resistance due to the aromatic rearrangement (Tondi et al., 2008). This thermal degradation pattern highlights the more complex chemical structure of the tannin when compared with cellulose. The thermal behavior of the CNF-T films is dominated by the thermal degradation patterns of cellulose, as the content of tannin is much smaller. However,

RI PT

the resultant CNF-T thermogram is not an overlapped result of the single components (Fig. 6b), stating that multiple interactions could take place between tannin and nanocellulose, and/or there was modification of the tannin after the mechanical processing. Also, the DrTG

SC

curve of the CNF-T clearly confirms that the tannin-cellulose film is a “new” material, and not a simple mixture of the two starting components where no interactions took place.

M AN U

From the thermogravimetric results it is possible to infer a molecular adsorption of tannin on the cellulose surface, which occurred due the water solubility of the initial tannin and its ability to make H-bonds with cellulose. The high residual mass of tannin powder after 600ºC is a result of the elevated recalcitrant character of polyphenolic branched

TE D

macromolecules. However, if tannin is adsorbed at the molecular level on a substrate, namely cellulose, it loses recalcitrance, as the degree of polymerization and the molecular weight tend to be lower than in the macromolecular state. In other words, tannin loses the ability to form

EP

char after its adsorption. At the macromolecular level, thermal energy is first demanded to break the cohesive intermolecular forces and then to separate the tannin macromolecules into

AC C

small units for further decomposition. In the molecular state, thermal energy is demanded only to decompose the tannin molecules as they are already separately adsorbed on the cellulose fibrils. The molecular adsorption of tannin justifies the lower thermal decomposition temperature observed for CNF-T film when compared with CNF films. It happened because the lower thermally-stable tannin molecules induced the early degradation of the nanocellulose, which is the major component of the film. This suggests an intimate contact between tannin and nanocellulose so that the tannin only slightly anticipates the degradation

17

ACCEPTED MANUSCRIPT of the nanocellulose fibrils. This analysis confirms that the tannin molecules are deeply

AC C

EP

TE D

M AN U

SC

RI PT

embedded in the cellulose net.

Fig. 6. Thermogravimetric curves (a) and their first derivative (b) of CNF, CNF-T films and tannin.

3.4.Antioxidant activity The antioxidant capacity of the CNF and CNF-T films were scavenged as a function of time, from 1 to 48 hours. The Trolox equivalence was calculated under water exposure (Fig. 7b) after measurement of the concentration decay as registered by the ORAC method (Fig.

18

ACCEPTED MANUSCRIPT 7a). The water in which the CNF films were soaked did not present any antioxidant capacity over time. On the other hand, after soaking, the CNF-T films presented significant antioxidant activity after 1 h, peaking at 8 h. A very high activity was still measured even after 48 h (Fig. 7b). These results suggest that the CNF-T films possess an on-demand release characteristic,

RI PT

where antioxidant molecules are released in a controllable manner upon water soaking. The antioxidant activity of the released tannins slightly decreases after 12h; however, even after 48 h it is still 17.7% higher than the antioxidant activity of pure Trolox (Fig. 6a). This

SC

property demonstrates the CNF-T is an excellent alternative active packaging (López-deDicastillo et al., 2012), especially for on-demand antioxidant action, which is usually needed

EP

TE D

M AN U

in elevated humidity or under moisture.

AC C

Fig. 7. Concentration decay of the antioxidant capacity over the time of tannin films obtained by ORAC assay (a). Antioxidant capacity measured as a function of µmol Trolox equivalence by g of film (b).

It is interesting to observe how the antioxidant activity decreases with a relatively contained slope. Understanding the rates in which tannin is released out from the nanocellulose matrix, and how it consequently promotes antioxidant activity, it is possible to further design CNF films containing different amounts of tannin so that tailored antioxidant 19

ACCEPTED MANUSCRIPT activity can be achieved. It is also possible to start pointing these films toward several technical applications that demand different antioxidant values at different times, from food (Zhou et al., 2016) to the pharmaceutical industry (Klemm et al., 2011).

RI PT

3.5.Tensile strength, leaching and chemical resistance The CNF and CNF-T films were exposed for seven days to different organic solvents, and to strong alkali and acid in order to evaluate their mechanical integrity. The average

SC

tensile strength of the tested films is presented together with their relative mass loss (Fig. 8). Before any chemical resistance test, the CNF-T resulted in ca. 20% weaker than the tannin-

M AN U

free CNF film. It is a result of the presence of tannin in-between cellulose nanofibrils, which hinders optimal conditions for intermolecular hydrogen bonds between the nanofibrils. However, the mechanical performance of the CNF-T film is comparable to that registered in others nanostructured cellulosic-based films (Olejar et al., 2014), and higher than plastics

AC C

EP

TE D

(Magalhães et al., 2013) and glass fiber-reinforced polyethylene (Bajpai, 2017).

Fig. 8. Tensile strength of films before and after chemical resistance tests. Same letters above the bars, separately for CNF and CNF-T films, indicate that the results are not statistically 20

ACCEPTED MANUSCRIPT different according to LSD Fisher test at 5% significance. The values inside the bars indicate the mass loss (in %) of the film after the chemical resistance test.

The mass loss observed for the CNF-T films soaked in water was ca. 11%, while

RI PT

negligible mass loss was observed for the CNF films. The mass loss in the CNF-T films occurs due the leaching of the embedded tannin, which, as discussed above, is the factor that drives the antioxidant capacity of these films. After seven days of drastic water exposure, it is

SC

possible to affirm that 45% of the incorporated tannin remained in the cellulose matrix. Hypothetically, these molecules are the first layer of the tannin-cellulose interface,

M AN U

representing the strongest possible interactions in this system. They are expected to be leached only after soaking for several days.

The CNF and CNF-T films showed good resistance and negligible mass loss for all the organic solvents. In some cases, for instance with corn oil, there was mass uptake that shows

TE D

how these films can be used to absorb exuded, undesirable oily substances. The tensile strength of the CNF-T was either slightly enhanced or negligibly altered, which means that the tannin does not contribute to the mechanical properties of the film and, hence, the active

EP

packaging material does not lose its mechanical performance after exposure to solvents. CNFT films showed improved performance after dipping in ethanol and ethyl acetate, even if no

AC C

significant mass changes were registered. This phenomenon can be explained by the increased mobility of the flavonoids exposed to polar solvents. The tannin molecules became more mobile (but not enough to become soluble) and they rearranged in a manner that allows the optimization of the secondary forces (between cellulose and tannin fractions) upon drying. The good chemical resistance observed against solvents was not repeated for alkali and acids (Fig. 8). The exposure to alkali reduced the tensile strength by 68 and 84% for CNF and CNFT. The acid exposure, followed by drying led to the degradation of the films. In sum, the films were highly resistant against organic solvents, but they were dramatically damaged by strong 21

ACCEPTED MANUSCRIPT alkali and acid conditions. Water is the only solvent that significantly interacts with the tannin-containing nanocellulose film, confirming the release of tannin also during weekly exposures.

RI PT

4. Conclusions Nanocellulose films with ca. 196 mg of intimately incorporated tannin by gram of film were designed and produced. The CNF-T films are 100% natural and biodegradable in their

SC

composition. They have a dense and highly hydrophobic surface, which resulted in

extraordinary air barrier properties (3.1 mL.min-1). These films are thermally stable until ca.

M AN U

230°C and chemically resistant against organic solvents. When in contact with water, the films released around 55% of the original tannin in one week and the tannin released offers a good antioxidant activity for at least 48 hours. From the results obtained here, it is possible to conclude that the CNF-T films have a wide range of potential applications in packaging

TE D

technology. For instance, it would be possible to increase the shelf-life of dry foods like rice or pasta, as well as preserved fruits, vegetables and meat, but also of external packages of oxidation-sensitive pharmaceuticals or cosmetics.

EP

Systematic studies on the addition of different tannin loads could be helpful to broaden the utilization of tannin-nanocellulose based materials. In addition, a deeper analytical

AC C

understanding about the cellulose-tannin interactions, as well as solvent-free routes to improve their interactions can be considered the next steps toward cleaner packaging technologies for safer food and materials.

Acknowledgements The authors wish to thank Ms. Nayara Lunkes for the graphic works performed, SETA® industry for the donation of the tannic extract used in the research and we also thank Miss Bia Carneiro for the English revision of the manuscript. 22

ACCEPTED MANUSCRIPT Funding This work was supported by CAPES (Coordination for the Improvement of Higher Education Personnel - Brazil), under the Science without Borders Program – CsF, process number: 88881.068144/2014-01 and CNPq (National Counsel of Technological and Scientific

RI PT

Development).

5. References

SC

Al-Naamani, L., Dobretsov, S., Dutta, J., 2016. Chitosan-zinc oxide nanoparticle composite coating for active food packaging applications. Innovative Food Sci. Emerging Technol. 38,

M AN U

Part A, 231-237. DOI: 10.1016/j.ifset.2016.10.010

Arbenz, A., Averous, L., 2015. Chemical modification of tannins to elaborate aromatic biobased macromolecular architectures. Green Chem. 17, 2626-2646. DOI: 10.1039/C5GC00282F

TE D

ASTM, 2014. Standard Practices for Evaluating the Resistance of Plastics to Chemical Reagents - ASTM D543-14. ASTM International, West Conshohocken.

EP

Bajpai, P., 2017. Chapter 3 - Structure and Properties of Cellulose and Nanocellulose, Pulp

AC C

and Paper Industry. Elsevier, pp. 27-40. Beitzen-Heineke, E.F., Balta-Ozkan, N., Reefke, H., 2017. The prospects of zero-packaging grocery stores to improve the social and environmental impacts of the food supply chain. J. Cleaner Prod. 140, Part 3, 1528-1541. DOI: 10.1016/j.jclepro.2016.09.227 Bernstad Saraiva, A., Pacheco, E.B.A.V., Gomes, G.M., Visconte, L.L.Y., Bernardo, C.A., Simões, C.L., Soares, A.G., 2016. Comparative lifecycle assessment of mango packaging made from a polyethylene/natural fiber-composite and from cardboard material. J. Cleaner Prod. 139, 1168-1180. DOI: 10.1016/j.jclepro.2016.08.135 23

ACCEPTED MANUSCRIPT Cai, J., Chen, J., Zhang, Q., Lei, M., He, J., Xiao, A., Ma, C., Li, S., Xiong, H., 2016. Wellaligned cellulose nanofiber-reinforced polyvinyl alcohol composite film: Mechanical and optical properties. Carbohydr. Polym. 140, 238-245. DOI: 10.1016/j.carbpol.2015.12.039 Chung, C., Lee, M., Choe, E.K., 2004. Characterization of cotton fabric scouring by FT-IR

RI PT

ATR spectroscopy. Carbohydr. Polym. 58, 417-420. DOI: 10.1016/j.carbpol.2004.08.005 De Vietro, N., Conte, A., Incoronato, A.L., Del Nobile, M.A., Fracassi, F., 2017. Aerosolassisted low pressure plasma deposition of antimicrobial hybrid organic-inorganic Cu-

M AN U

Technol. 41, 130-134. DOI: 10.1016/j.ifset.2017.02.010

SC

composite thin films for food packaging applications. Innovative Food Sci. Emerging

DIN, 1993. Spanplatten und Faserplatten; Bestimmung der Zugfestigkeit senkrecht zur Plattenebene; Deutsche Fassung EN 319:1993, Deutsches Institut für Normung, Berlin. Douglass, E.F., Avci, H., Boy, R., Rojas, O.J., Kotek, R., 2016. A Review of Cellulose and

TE D

Cellulose Blends for Preparation of Bio-derived and Conventional Membranes, Nanostructured Thin Films, and Composites. Polym. Rev. 1-62. DOI:

EP

10.1080/15583724.2016.1269124

Duan, J., Reddy, K.O., Ashok, B., Cai, J., Zhang, L., Rajulu, A.V., 2016. Effects of spent tea

AC C

leaf powder on the properties and functions of cellulose green composite films. J. Environ. Chem. Eng. 4, 440-448. DOI: 10.1016/j.jece.2015.11.029 Dufresne, A., 2013. Nanocellulose: a new ageless bionanomaterial. Mater. Today 16, 220227. DOI: 10.1016/j.mattod.2013.06.004 Duval, A., Avérous, L., 2016. Oxyalkylation of Condensed Tannin with Propylene Carbonate as an Alternative to Propylene Oxide. ACS Sustainable Chem. Eng. 4, 3103-3112. DOI: 10.1021/acssuschemeng.6b00081 24

ACCEPTED MANUSCRIPT EU, 2009. EU Guidance to the Commission Regulation (EC) No 450/2009 of 29 May 2009 on active and intelligent materials and articles intended to come into contact with food. , in: Sustainability, European Commission - Health and Consumers Directorate-General Directorate-General Directorate E-Safety of the Food chain. E6- Innovation and

RI PT

Sustainability, Ed. 2009. Ejima, H., Richardson, J. J., Liang, K., Best, J. P., Koeverden, M. P. van, Such, G. K., Cui, J., Caruso, F., 2013. One-Step Assembly of Coordination Complexes for Versatile Film and

SC

Particle Engineering. Science 341, 154-157. DOI: 10.1126/science.1237265

M AN U

Guo, J., Tardy, B. L., Christofferson, A. J., Dai, Y., Richardson, J. J., Zhu, W., Hu, M., Ju, Y., Cui, J., Dagastine, R. R., Yarovsky, I., Caruso, F., 2016. Modular assembly of superstructures from polyphenol-functionalized building blocks. Nature Nanotechnol. 11, 1105-1112. DOI: 10.1038/NNANO.2016.172

TE D

Hemingway, R.W., Karchesy, J.J., Branham, S.J., 1989. Chemistry and Significance of Condensed Tannins. Springer.

Hubbe, M.A., Ferrer, A., Tyagi, P., Yin, Y., Salas, C., Pal, L., Rojas, O.J., 2017.

EP

Nanocellulose in Thin Films, Coatings, and Plies for Packaging Applications: A Review.

AC C

BioResources 12, 2143-2233. DOI: 10.1002/adem.200400097 Kaya, M., Akyuz, L., Sargin, I., Mujtaba, M., Salaberria, A.M., Labidi, J., Cakmak, Y.S., Koc, B., Baran, T., Ceter, T., 2017. Incorporation of sporopollenin enhances acid–base durability, hydrophobicity, and mechanical, antifungal and antioxidant properties of chitosan films. J. Ind. Eng. Chem. 47, 236-245. DOI: 10.1016/j.jiec.2016.11.038 Kisonen, V., Prakobna, K., Xu, C., Salminen, A., Mikkonen, K.S., Valtakari, D., Eklund, P., Seppälä, J., Tenkanen, M., Willför, S., 2015. Composite films of nanofibrillated cellulose and O-acetyl galactoglucomannan (GGM) coated with succinic esters of GGM showing potential 25

ACCEPTED MANUSCRIPT as barrier material in food packaging. J. Mater. Sci. 50, 3189-3199. DOI: 10.1007/s10853015-8882-7 Klemm, D., Kramer, F., Moritz, S., Lindström, T., Ankerfors, M., Gray, D., Dorris, A., 2011. Nanocelluloses: A New Family of Nature-Based Materials. Angew. Chem., Int. Ed. 50, 5438-

RI PT

5466. DOI: 10.1002/anie.201001273

Krepker, M., Shemesh, R., Danin Poleg, Y., Kashi, Y., Vaxman, A., Segal, E., 2017. Active food packaging films with synergistic antimicrobial activity. Food Control 76, 117-126. DOI:

SC

10.1016/j.foodcont.2017.01.014

M AN U

Kumode, M.M.N., Bolzon, G.I.M., Magalhães, W.L.E., Kestur, S.G., 2017. Microfibrillated nanocellulose from balsa tree as potential reinforcement in the preparation of ‘green’ composites with castor seed cake. J. Cleaner Prod. 149, 1157-1163. DOI: 10.1016/j.jclepro.2017.02.083

TE D

Lavoine, N., Desloges, I., Dufresne, A., Bras, J., 2012. Microfibrillated cellulose – Its barrier properties and applications in cellulosic materials: A review. Carbohydr. Polym. 90, 735-764.

EP

DOI: 10.1016/j.carbpol.2012.05.026

Lavoine, N., Guillard, V., Desloges, I., Gontard, N., Bras, J., 2016. Active bio-based food-

AC C

packaging: Diffusion and release of active substances through and from cellulose nanofiber coating toward food-packaging design. Carbohydr. Polym. 149, 40-50. DOI: 10.1016/j.carbpol.2016.04.048 Lee, K.-Y., Tammelin, T., Schulfter, K., Kiiskinen, H., Samela, J., Bismarck, A., 2012. High Performance Cellulose Nanocomposites: Comparing the Reinforcing Ability of Bacterial Cellulose and Nanofibrillated Cellulose. ACS Appl. Mater. Interfaces 4, 4078-4086. DOI:10.1021/am300852a

26

ACCEPTED MANUSCRIPT Lengowski, E.C., Magalhães, W.L.E., Nisgoski, S., Muniz, G.I.B.d., Satyanarayana, K.G., Lazzarotto, M., 2016. New and improved method of investigation using thermal tools for characterization of cellulose from eucalypts pulp. Thermochim. Acta 638, 44-51. DOI: 10.1016/j.tca.2016.06.010

RI PT

López-de-Dicastillo, C., Gómez-Estaca, J., Catalá, R., Gavara, R., Hernández-Muñoz, P., 2012. Active antioxidant packaging films: Development and effect on lipid stability of brined sardines. Food Chem. 131, 1376-1384. DOI: 10.1016/j.foodchem.2011.10.002

SC

Magalhães, W.L.E., Pianaro, S.A., Granado, C.J.F., Satyanarayana, K.G., 2013. Preparation

M AN U

and characterization of polypropylene/heart-of-peach palm sheath composite. J. Appl. Polym. Sci. 127, 1285-1294. DOI: 10.1002/app.37633

Missio, A.L., Tischer, B., Santos, P.S.B., Codevilla, C., Menezes, C.R., Barin, J.S., Haselein, C.R., Labidi, J., Gatto, D.A., Petutschnigg, A., Tondi, G., 2017b. Analytical characterization

TE D

of purified Mimosa (Acacia mearnsii) industrial tannin extract: Single and sequential fractionation. Sep. Purif. Technol. 186, 218–225. DOI: 10.1016/j.seppur.2017.06.010 Noshirvani, N., Ghanbarzadeh, B., Rezaei Mokarram, R., Hashemi, M., 2017. Novel active

EP

packaging based on carboxymethyl cellulose-chitosan-ZnO NPs nanocomposite for increasing

AC C

the shelf life of bread. Food Packag. Shelf Life 11, 106-114. DOI: 10.1016/j.fpsl.2017.01.010 Olejar, K.J., Ray, S., Ricci, A., Kilmartin, P.A., 2014. Superior antioxidant polymer films created through the incorporation of grape tannins in ethyl cellulose. Cellulose 21, 4545-4556. DOI: 10.1007/s10570-014-0447-4 Ou, B., Hampsch-Woodill, M., Prior, R.L., 2001. Development and Validation of an Improved Oxygen Radical Absorbance Capacity Assay Using Fluorescein as the Fluorescent Probe. J. Agric. Food Chem. 49, 4619-4626. DOI: 10.1021/jf010586o

27

ACCEPTED MANUSCRIPT Pizzi, A., 2008. Tannins: Major Sources, Properties and Applications, in: Belgacem, M.N., Gandini, A. (Eds.), Monomers, Polymers and Composites from Renewable Resources. Elsevier, Amsterdam, p. 553.

CRC Press.

RI PT

Pizzi, A., 1994. Advanced wood adhesives technology. Ch.5 Tannin-based wood adhesives.

Plastic, E., 2017. Plastics - the Facts 2016, An analysis of European plastics production,

SC

demand and waste data. Association of Plastic Manufacture.

Schwanninger, M., Rodrigues, J.C., Pereira, H., Hinterstoisser, B., 2004. Effects of short-time

M AN U

vibratory ball milling on the shape of FT-IR spectra of wood and cellulose. Vib. Spectrosc. 36, 23-40. DOI: 10.1016/j.vibspec.2004.02.003

TAPPI, 1997. Thickness (caliper) of paper, paperboard, and combined board. T 411 om-97,

TE D

Technical Association of the Pulp and Paper Industry Atlanta, USA. Tondi, G., Petutschnigg, A., 2015. Middle infrared (ATR FT-MIR) characterization of industrial tannin extracts. Ind. Crops Prod. 65, 422-428. DOI: 10.1016/j.indcrop.2014.11.005

EP

Tondi, G., Pizzi, A., Pasch, H., Celzard, A., 2008. Structure degradation, conservation and rearrangement in the carbonisation of polyflavonoid tannin/furanic rigid foams – A MALDI-

AC C

TOF investigation. Polym. Degrad. Stab. 93, 968-975. DOI: 10.1016/j.polymdegradstab.2008.01.024 Wang, H., Wang, L., 2017. Developing a bio-based packaging film from soya by-products incorporated with valonea tannin. J. Cleaner Prod. 143, 624-633. DOI: 10.1016/j.jclepro.2016.12.064

28

ACCEPTED MANUSCRIPT Zhou, B., Hu, X., Zhu, J., Wang, Z., Wang, X., Wang, M., 2016. Release properties of tannic acid from hydrogen bond driven antioxidative cellulose nanofibrous films. Int. J. Biol.

AC C

EP

TE D

M AN U

SC

RI PT

Macromol. 91, 68-74. DOI: 10.1016/j.ijbiomac.2016.05.084

29

ACCEPTED MANUSCRIPT

Tannin is embedded in nanocellulose matrix through mechanical process; Barrier properties of the cellulose film are further enhanced by tannin addition;

RI PT

Tannin-embedded films are thermal resistant up to 230oC;

AC C

EP

TE D

M AN U

SC

Tannin-nanocellulose films displayed antioxidant activity upon water soaking.