Carbohydrate Polymers 135 (2016) 18–26

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Preparation of nanocellulose from micro-crystalline cellulose: The

effect on the performance and properties of agar-based composite
films
Shiv Shankar, Jong-Whan Rhim ∗
Department of Food Engineering and Bionanocomposite Research Institute, Mokpo National University, 61 Dorimri, Chungkyemyon, Muangun,
534729 Jeonnam, Republic of Korea

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

Article history: A facile approach has been performed to prepare nanocellulose (NC) from micro-crystalline cellulose
Received 23 July 2015 (MCC) and test their effect on the performance properties of agar-based composite films. The NC was
Received in revised form 15 August 2015 characterized by STEM, XRD, FTIR, and TGA. The NC was well dispersed in distilled water after sonication
Accepted 25 August 2015
and their size was in the range of 100–500 nm. The XRD results revealed the crystallinity of NC. The
Available online 29 August 2015
crystallinity index of NC (0.71) was decreased compared to the MCC (0.81). The effect of NC or MCC
content (1, 3, 5 and 10 wt% based on agar) on the mechanical, water vapor permeability (WVP), and
Keywords:
thermal properties of the composites were studied. The NC obtained from MCC can be used as a reinforcing
Nanocellulose
Micro-crystalline agent for the preparation of biodegradable composites films for their potential use in the development
Cellulose of biodegradable food packaging materials.
Agar © 2015 Elsevier Ltd. All rights reserved.
Composite films

1. Introduction (Paula et al., 2015), chitosan (Li, Zhou, & Zhang, 2009), and ther-
moplastic starch (Savadekar, Karande, Vigneshwaran, Kadam, &
Recently, the demands of products made from renewable and Mhaske, 2015) have been studied to make biodegradable pack-
sustainable resources that are biodegradable, carbon neutral, non- aging films. However, owing to the poor mechanical and barrier
petroleum based, and have low environmental, animal/human properties of biopolymer-based packaging films, the industrial uti-
health and safety risks are increasing by the consumers, indus- lization has not been materialized (Rhim et al., 2013). With the
try, and the government (Duncan, 2011; Moon, Martini, Nairn, advent of nanotechnology, various types of nano-sized filler mate-
Simonsen, & Youngblood, 2011; Rhim, Park, & Ha, 2013). Thus the rials such as nanoclays, metallic nanoparticles, and nanocelluloses
suitability of biomaterials, especially biopolymers, for film pro- have been utilized to improve the physical, mechanical, and gas bar-
duction have been intensively studied. Various biopolymers such rier properties of biopolymer films (Rhim et al., 2013). Among the
as polysaccharides, proteins, and lipids have been considered as nanofillers, nanocellulose isolated from various natural resources,
an alternative biodegradable packaging materials because of their such as jute, flax, rice straw, paper mulberry pulp, and by-products
abundant availability, renewable, inexpensive, environmentally of agricultural crops have been gaining considerable attention due
friendly, biodegradable, and biocompatible nature (Duncan, 2011; to their unique and attractive features such as biodegradability,
Rhim et al., 2013; Tang, Kumar, Alavi, & Sandeep, 2012). Among biocompatibility, renewability, abundance, high aspect ratio, and
these materials, polysaccharides are most widely utilized and stud- light weight (Kalia et al., 2011). However, acid hydrolysis tech-
ied for food packaging applications due to their good film forming niques has been utilized to prepare nanocellulose from cellulose
property and good mechanical and gas barrier properties compared microfiber (Reddy & Rhim, 2014). In the present study, nanocellu-
with other biopolymeric materials (Rhim et al., 2013). Various lose (NC) was prepared from micro-crystalline cellulose using facile
polysaccharides such as agar (Shankar, Teng, & Rhim, 2014), car- approach without using any harmful chemicals like sulfuric acid
boxymethyl cellulose (Ghanbarzadeh & Almasi, 2011), carrageenan and hydrochloric acid.
Among the biopolymers, agar has a potential for making
biodegradable films for packaging applications (Kanmani & Rhim,
∗ Corresponding author. 2014; Shankar et al., 2014). Agar is a polysaccharide derived
E-mail address: [email protected] (J.-W. Rhim). from red algae contains two components such as agarose and

http://dx.doi.org/10.1016/j.carbpol.2015.08.082
0144-8617/© 2015 Elsevier Ltd. All rights reserved.
 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26 19

agaropectin. Agar is soluble in hot water and it forms biodegradable (D) of the MCC and NC was calculated in nm using the Scherrer’s
film when added with proper plasticizer. Agar has been also suc- equation (Reddy & Rhim, 2014):
cessfully used to make nanocomposites with clay, silver, and copper
K
nanoparticles and reported that the mechanical and water vapor D=
permeability have been improved with the addition of nanofillers ˇ1/2 cos 
(Rhim et al., 2013; Shankar et al., 2014; Shankar & Rhim, 2015).
where K is the Scherrer constant (0.94),  is the X-ray wavelength
However, there are no reports on reinforcement of nanocellulose
( = 0.154056 nm), ˇ1/2 is the full width at the half maximum of the
from regenerated MCC into agar based composites available in the
deflection peak (FWHM), and  is Bragg’s angle.
literature. Therefore, in the present study, nanocellulose prepared
from MCC was reinforced into agar polymer film and the films were
2.4. Preparation of agar/cellulose nanocomposite film
characterized using various techniques.

Agar, agar/MCC, and agar/NC composite films were prepared

2. Materials and methods using a solution casting method following the method of Shankar
et al. (2014). For the preparation of agar film, 3 g of agar powder was
2.1. Materials dissolved into 150 mL of distilled water including 0.9 g of glycerol (a
plasticizer) while mixing vigorously for 30 min at 95 ◦ C using a mag-
All chemicals were procured from Sigma–Aldrich (St. Louis, MO, netic stirrer. The solubilized film forming solution was cast evenly
USA) unless otherwise stated. Food grade agar was procured from onto a leveled Teflon film (Cole-Parmer Instrument Co., Chicago, IL,
Gel Tec Co., Ltd. (Seoul, Korea). USA) coated glass plate (24 cm × 30 cm), then dried for about 48 h at
room temperature. In addition, agar/MCC and agar/NC composite
films were prepared with different concentration of MCC and NC
2.2. Preparation of cellulose nanoparticles
(1, 3, 5, and 10 wt% based on agar weight). For this, predetermined
amount of MCC or NC was dispersed in 150 mL of distilled water
Nanocellulose (NC) was prepared from microcrystalline cellu-
and stirred vigorously for 1 h using a magnetic stirrer. The fully dis-
lose (MCC) by the method of Adsul, Soni, Bhargava, and Bansal
persed suspensions were homogenized using a high shear mixer
(2012). For this, 5 g of microcrystalline cellulose (MCC) was added
(T25 basic, Ika Labotechnik, Janke & Kunkel GmbH & Co., KG Staufen,
into 100 mL solution of 7% (w/w) NaOH and 12% (w/w) urea and the
Germany) at 12,000 rpm for 10 min. Then, 0.9 g of glycerol were
solution was stirred for 30 min at room temperature, then kept in a
dissolved into the suspension while mixing vigorously for about
freezer at −20 ◦ C for 16 h. The content was stirred using a magnetic
20 min. Finally, 3 g of agar was added slowly into the solution and
stirrer at 1000 rpm for 10 min until the MCC dissolved completely.
mixed vigorously at 95 ◦ C using a hot plate stirrer and followed the
The dissolved MCC was regenerated by adding an excess amount
same procedure for the preparation of the neat agar film. The dried
(∼1000 mL) of double distilled water (DDW). The regenerated cel-
films were peeled off from the casting plate and conditioned in a
lulose was separated by centrifugation at 3000 rpm for 10 min. The
constant temperature–humidity chamber set at 25 ◦ C and 50% rela-
regenerated cellulose was washed with double distilled water for
tive humidity (RH) for at least 48 h before further characterization.
6–7 times to remove the remnant of urea and NaOH and bring down
the pH around 7–9. The regenerated MCC was suspended in DDW
and ultrasonicated for 5 min using a 2 mm probe at 75% amplitude 2.5. Characterization of agar/AgNPs nanocomposite films
(SONICS Vibra cell; model VC 130, Sonic & Materials, Inc., Newtown,
CT, USA). The dispersed cellulose, nanocellulose (NC), was kept at 2.5.1. Morphological observation, FTIR, and XRD
4 ◦ C until further used. Scanning electron microscopy (SEM) analysis was used to
observe surface microstructure of agar and agar-based composite
films. Film sample was cut into small pieces and directly mounted
2.3. Characterization of cellulose nanoparticles on a specimen holder and analyzed using a field emission scanning
electron microscopy (FE-SEM, S-4800, Hitachi Co., Ltd., Matsuda,
The morphology and size of MCC were determined by FE-SEM Japan) with an accelerating voltage of 5.0 kV.
(S-4800, Hitachi Co., Ltd., Matsuda, Japan). The size and shape of Fourier transform infrared (FTIR) spectra of the films were
NC was determined by scanning transmission electron microscopy obtained using an attenuated total reflectance-Fourier transform
(STEM, FE-SEM, S-4800, Hitachi Co., Ltd., Matsuda, Japan). The aque- infrared (ATR-FTIR) spectrophotometer (TENSOR 37 Spectropho-
ous suspension of NC was dropped on a carbon coated copper grid, tometer with OPUS 6.0 software, Billerica, MA, USA) operated at a
air dried and image was analyzed using FE-SEM in the transmission resolution of 4 cm−1 . Film samples were cut into rectangular shapes
mode. (5 cm × 5 cm) and directly placed on the ray exposing stage. The
Fourier transform infrared (FTIR) spectra of MCC and NC were spectrum was recorded at wave number of 500–4000 cm−1 .
collected using a FTIR spectroscopy (TENSOR 37 Spectrophotome- X-ray diffraction (XRD) pattern of the films was analyzed
ter with OPUS 6.0 software, Billerica, MA, USA) at a resolution by X-ray diffractometer (PANalytical X’pert pro MRD diffrac-
of 4 cm−1 in the range of 500–4000 cm−1 . The crystalline nature tometer, Amsterdam, Netherlands). Rectangular film samples
of MCC and NC was determined by XRD (PANalytical Xpert pro (2.5 cm × 2.5 cm) were placed on a glass slide and the spectra
MRD diffractometer, Amsterdam, Netherlands) by placing MCC and were recorded using Cu K␣ radiation (wavelength of 0.1541 nm)
NC powders on a glass slide and spectra were recorded using Cu and a nickel monochromator filtering wave at 40 kV and 30 mA.
K␣ radiation at 40 kV and 30 mA with scanning at 2 = 5–50◦ . The The diffraction pattern was obtained at diffraction angles between
degree of crystallinity (DC) of MCC and NC was calculated using 2 = 30 and 80◦ with scanning speed of 0.4◦ /min at room tempera-
following equation (Shankar & Rhim, 2015): ture.
I2 0 0 − Iam
DC = 2.5.2. Surface color and optical properties
I2 0 0
The surface color of the films was measured using a Chroma
where I2 0 0 is the intensity of the peak at (2 0 0) lattice and Iam is meter (Konica Minolta, CR-400, Tokyo, Japan). A white standard
the intensity of the peak at amorphous region. The crystallite size color plate (L = 97.75, a =−0.49, and b = 1.96) was used as a
20 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26

background for color measurements. Hunter color (L, a, and b) val- Optics Co., Ltd., Kunpo, Korea). Each film sample was cut into rect-
ues were averaged from five readings from each sample. The total angular pieces (3 cm × 10 cm) and directly placed on the horizontal
color difference (E) was calculated as follows: movable stage (Black Teflon coated steel) fixed with the contact
angle analyzer. 10 ␮L of water was dropped on surface of the film
2 2 2 0.5
E = [(L) + (a) + (b) ] using a micro syringe and measured the contact angle on both
sides of the water drop to assume symmetry and horizontal level
where L, a, and b are the difference between each color value (Shankar & Rhim, 2015).
of standard color plate and film samples, respectively.
Optical properties of the films were determined by measuring
2.5.5. Thermal stability
the light absorption spectra and the transmittance of the films.
Thermogravimetric analysis (TGA) was performed to analyze
Each film was cut into a square piece (5 cm × 5 cm) and directly
the thermal properties of agar and agar-based composite films
mounted between magnetic holders of the spectrophotometer, and
using a TGA analyzer (TGA; Hi-Res TGA 2950 Thermo Gravimet-
the absorbance and percent transmittance were measured using
ric Analyzer, TA Instrument, New Castle, DE, USA). The samples
a UV-vis spectrophotometer (Mecasys Optizen POP Series UV/Vis,
(∼5 mg) were taken in standard aluminum cups and empty cup was
Seoul, Korea). Transparency of the films was tested by measuring
used as a reference. The TGA thermogram was obtained by heating
percent transmittance at 280 nm (T280 ) and 660 nm (T660 ) (Shankar,
the samples from room temperature to 600 ◦ C with the heating rate
Reddy, Rhim, & Kim 2015).
of 10 ◦ C/min under a nitrogen flow of 50 cm3 /min. Derivative forms
of TGA (DTG) was obtained by differentials of TGA using the central
2.5.3. Thickness and tensile properties of the films finite difference method as follows (Shankar & Rhim, 2015):
The film thickness was measured with a hand-held microme-
ter (Dial Thickness Gauge 7301, Mitutoyo Corporation, Kanagawa, Wt+t − Wt−t
DTG =
Japan) with a 0.01-mm accuracy. Thickness was measured at five 2t
random points of film samples and the average value was used. where Wt+t − Wt−t are the residual weight of the sample at time
For the tensile test, film sample was cut into rectangular strips t + t and t − t, respectively, and t is the time interval for reading
(2.54 cm × 15 cm) using a precision double blade cutter (model residual sample weight. The char content at 600 ◦ C was determined
LB.02/A, Metrotec, S.A., San Sebastian, Spain), and tensile proper- from the TGA curve (Shankar & Rhim, 2015).
ties such as tensile strength (TS), elongation at break (E), and elastic
modulus (EM) of the film was measured according to the standard 2.6. Statistical analysis
test method of ASTM D 882-88 using an Instron Universal Testing
Machine (Model 5565, Instron Engineering Corporation, Canton, Film properties were measured with individually prepared films
MA, USA). The instrument was operated in tensile mode with an ini- in triplicate as the replicated experimental units and the results
tial grip separation and crosshead speed of 50 mm and 50 mm/min, were provided with means (standard deviation) values. One way
respectively. The TS (MPa) was determined by dividing the maxi- analysis of variance (ANOVA) was performed, and the significance
mum load (N) by the initial cross-sectional area (m2 ) of the films. of each mean value was determined (p < 0.05) with the Duncan’s
The E (%) was determined by dividing the extension at the rupture multiple range test of the statistical analysis system using the SPSS
of the film by the initial length of the film (50 mm) multiplied by computer program (SPSS Inc., Chicago, IL, USA).
100. The EM (MPa) was determined from the slope of the linear
portion of the stress–strain curve, which corresponds to the stress 3. Results and discussion
divided by the strain of the film sample. Ten measurements were
carried out for each film, and the average values were presented. 3.1. Characterization of cellulose nanoparticles

2.5.4. Water vapor permeability (WVP) and water contact angle The MCC was dissolved completely in the mixed solution of
(WCA) NaOH and urea to become a transparent solution. After regener-
The water vapor permeability of the film samples was deter- ation and ultrasonication for 5 min, the MCC was readily dispersed
mined gravimetrically according to the method of ASTM E96-95 into distil water to make homogeneous dispersion without for-
with slight modification (Gennadios, Weller, & Goodings, 1994). mation of any precipitation. The color of the dispersed cellulose
Rectangular film samples (7.5 cm × 7.5 cm) were mounted hori- solution was milky white. To observe the morphological structure
zontally on the WVP measuring cups which were made of flaxy of the regenerated cellulose, the dispersed cellulose was analyzed
glass, poly(methylmethacrylate), with depth and diameter of 2.5 by the STEM. The SEM images of MCC and STEM micrograph of
and 6.8 cm, respectively, filled with distilled water up to 1 cm dispersed regenerated nanocellulose (NC) are shown in Fig. 1. The
underneath the film. The cups were placed in the humidity cham- MCC was irregular in shape with the size in the range of 20–200 ␮m.
ber controlled at 25 ◦ C and 50% RH with air current circulation of However, the size decreased to the range of 100–500 nm after for-
198 m/min. The cup was weighed every hour for a period of 8 h. mation of NC. STEM micrograph showed that the NC was well
Water vapor transmission rate (WVTR) was determined from the separated from each other and dispersed in the aqueous solution.
slopes of the steady state portion of weight loss of the cup versus The size of the NC was further analyzed using a dynamic light scat-
time curve. Then the WVP of the films in g/m2 Pa s was calculated tering (DLS) method and found that the average particle size was
as follows: 646.9 ± 94.9 nm with the polydispersity index of 0.2522. The differ-
WVTR × L ence between particle sizes of the NC determined by STEM and DLS
WVP = methods was attributed to that the STEM determined the physical
p
size of particles in the dried form, whereas the DLS determined the
where WVTR is the measured water vapor transmission rate hydrodynamic radius of particles dispersed in a solution. Since cel-
(g/m2 s) through a film, L was the mean thickness of the film (m), lulose molecules are highly hydrophilic and swells easily in water,
and p was partial water vapor pressure difference (Pa) across the the particle size determined by the DLS method was higher com-
film. pared to that determined by the STEM method.
The contact angle of water (WCA) on the film surface was mea- The MCC and NC were further characterized by XRD to check
sured using the WCA analyzer (model Phoenix 150, Surface Electro the crystallinity and the results were shown in Fig. 2. The MCC
 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26 21

Fig. 1. (A) SEM micrograph of MCC and (B) STEM image of NC.

exhibited its characteristic crystalline peaks around 2 = 22.6, 14.8, FTIR spectroscopy was performed to understand the change in
16.2, and 34.5 corresponding to (2 0 0), (−1 1 0), (1 1 0), and (4 0 0) the chemical structure between MCC and NC (Fig. 3). In general,
lattice planes, respectively, which is a characteristic of cellu- the change in the crystal structure of cellulose leads to disappear-
lose I structure (Yu, Yan, Lei, Qin, & Yao, 2014). However, NC ance or reduction in intensity of certain FTIR peaks of the crystalline
showed the characteristic crystalline peaks at 2 = 12.03, 20.19, and domains of cellulose (Adsul et al., 2012). The FTIR absorption peak
22.5 corresponding to (1 1 0), (−1 1 0), and (2 0 0) lattice planes, at 1430 cm−1 corresponds to CH2 bending vibration that attributed
respectively, which is a characteristic of cellulose II structure (Yu to the “crystallinity band” in the cellulose. FTIR spectrum of NC
et al., 2014). shows a significant reduction in the intensity of this crystallinity
The peak at 22.6 is broader in the case of NC compared with the band compared to that of MCC, which indicates that the level of
MCC, which indicates that the crystallite size of NC was smaller than crystallinity of cellulose crystals decreased during the process of
that of the MCC. The crystallite size of the NC and MCC calculated cellulose regeneration and dispersion. On the other hand, the inten-
using the Scherrer equation was 4.27 and 5.73 nm, respectively. sity of vibrational peak at 893 cm−1 increased in NC compared with
The crystallite size of MCC decreased slightly after treatment with MCC. The band at 893 cm−1 was attributed to the C O C stretching
NaOH and urea. However, the crystallite size of cellulose nanocrys- vibration of ␤-(1 → 4)-glycosidic linkages of cellulose, which was
tals prepared by acid hydrolysis of cellulose was found to be larger considered as an “amorphous band”. The result on the reduction in
than that of untreated cellulose fiber (Maiti et al., 2013). The crys- the crystallinity of NC obtained by the FTIR test agreed well with
tallinity index (CI) of MCC and NC was 0.81 and 0.70, respectively. that obtained by the XRD test.
The decrease in the CI of the NC was probably due to the treatment The thermal stability of MCC and NC was tested using TGA and
of MCC with NaOH and urea. It has been reported that the urea had the results are shown in Fig. 4. MCC showed only one step of thermal
no direct interaction with cellulose in the dissolution process, but it degradation, however, NC showed two steps of thermal degrada-
promoted the decrease in the crystallinity (Isobe et al., 2013). Sim- tion. The weight loss usually observed by the moisture evaporation
ilar results of reduced crystallinity and crystallite size have been in the range of 80–100 ◦ C (Reddy & Rhim, 2014; Shankar et al.,
observed when cellulose was treated with high concentration (7%, 2014) was not observed in the present study since the samples were
w/v) of NaOH (Mittal, Katahira, Himmel, & Johnson, 2011). dried previously in an oven at 70 ◦ C. The main thermal degradation

Fig. 2. XRD of MCC and NC. Fig. 3. FTIR spectra of NC and MCC.
22 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26

Fig. 4. (a) TGA and (b) DTG curves of MCC and NC.

was occurred in the range of 300–370 ◦ C, which was mainly due to 3.2. Characterization of agar composite films
the thermal degradation of cellulosic materials. In the case of NC,
additonal thermal degration step was observed at lower temper- 3.2.1. Apparent color and transparency
ature (220–240 ◦ C), which was presumably due to the amorphous Apparently, all the films including agar, agar/MCC, and agar/NC
materials remained the NC. The TGA results exhibited that the ther- composite films prepared using the solvent casting method were
mal stability of the NC was significantly increased compared to flexible, free-standing, and homogeneous. The color of the film
the MCC. The addition of urea increased the dissolution of cel- is of importance because it might influence the product demand
lulose in the solvent and improved the thermal stability of the and consumer preference. Table 1 shows the surface color prop-
cellulose. Urea might help alkali hydrate to penetrate into the crys- erties and the total color difference (E) of the composite films.
talline region of cellulose by stabilizing the alkali-swollen cellulose The total color difference, which takes into account of three
molecules, led to an increase in dissolved fraction of cellulose. color parameters, such as lightness (L), greenness–redness (a), and
The increased thermal stability may be due to the local accumu- blueness–yellowness (b) of surface color, is used as an index for
lation of urea on the hydrophobic surface, which prevents the the color difference between the sample and the standard color.
hydrophobic association of dissolved cellulose molecules (Isobe As shown in Table 1, The L-values did not change significantly
et al., 2013). after incorporation of both crystalline cellulose, however, a-values

Fig. 5. SEM micrograph of agar/cellulose composite films.

 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26 23

Table 1
Color and percent transmittance of agar and agar/cellulose composite films.

Film L a b E T280 (%) T660 (%)

Agar 92.70 ± 0.21bc −0.58 ± 0.02e 4.90 ± 0.09a 2.74 ± 0.19ab 55.5 ± 0.8g 89.0 ± 0.5g
NC1% 92.93 ± 0.16c −0.62 ± 0.01d 4.93 ± 0.10a 2.64 ± 0.20a 54.3 ± 0.5g 86.1 ± 0.5f
NC3% 92.71 ± 0.10bc −0.67 ± 0.01c 5.18 ± 0.11ab 2.95 ± 0.10bc 49.0 ± 1.4e 82.2 ± 0.5e
NC5% 92.71 ± 0.10bc −0.70 ± 0.01c 5.31 ± 0.06bc 3.05 ± 0.08bc 45.0 ± 1.2d 77.6 ± 1.7d
NC10% 92.63 ± 0.10bc −0.74 ± 0.05b 5.53 ± 0.29c 3.28 ± 0.24cd 40.5 ± 2.1c 71.5 ± 2.1c
MCC1% 92.57 ± 0.13b −0.50 ± 0.01f 5.01 ± 0.07ab 2.91 ± 0.12bc 51.5 ± 1.4f 83.1 ± 2.1e
MCC3% 92.60 ± 0.20bc −0.58 ± 0.02e 5.25 ± 0.12bc 3.07 ± 0.19bc 47.8 ± 1.8e 77.0 ± 1.9d
MCC5% 92.46 ± 0.47ab −0.70 ± 0.07c 5.82 ± 0.55d 3.61 ± 0.71d 37.3 ± 7.1b 63.8 ± 5.6b
MCC10% 92.21 ± 0.48a −0.84 ± 0.03a 6.25 ± 0.31e 4.13 ± 0.46e 31.2 ± 3.2a 51.5 ± 5.0a

Values with the same superscript letter in the same column indicate that they are not statistically different (p < 0.05).

decreased and b-values increased with the incorporation of crys- incident and transmitted light, a: extinction coefficient, l: path
talline celluloses. Decrease in a-values and increase in b-values of length of light), it is reasonable to conclude the transmittance of
both agar/NC and agar/MCC composite films was dependent on the the composite films decreased logarithmically with increase in the
content of crystalline cellulose. Consequently, the total color differ- content of NC or MCC.
ence (E) of agar/crystalline cellulose composite films increased
by incorporation of either NC or MCC. The increase in E value 3.2.2. Surface morphology of nanocomposite films
was higher in the case of MCC compared to NC incorporation. Simi- Surface morphology of agar, agar/NC, and agar/MCC compos-
lar results were obtained when they incorporated paper-mulberry ite films were observed by FE-SEM and the results are shown in
pulp nanocellulose into agar matrix (Reddy & Rhim, 2014). Fig. 5. As shown in the SEM images, the pristine agar film exhibited
Transparency of the film samples was determined by measuring smooth and homogeneous surface and the NC particles were found
transmittance at 660 nm, using UV–vis spectroscopy analysis and to be well dispersed in the agar matrix to form smooth surfaced
the results are also shown in Table 1. The neat agar film was trans- homogeneous composite films, however, the surface became less
parent with transmittance of 89.0 ± 0.5%. However, it decreased homogeneous when the concentration of NC increased to 10 wt%.
slightly after the formation of composite with NC and MCC. The This might be caused by the agglomeration of NC in the agar matrix
decrease in the transmittance at 660 nm (T660 ), which indicates at high concentration (Agustin et al., 2013). On the contrary, incor-
transparency of the film, was lower in case of NC compared to MCC. poration of MCC into agar matrix showed very rough surface with
In addition, the transparency of the composite films decreased sig- aggregated form. MCC particles could be seen embedded in the agar
nificantly with increase in the both crystalline celluloses. Similar matrix.
trend of decrease in transmittance at 280 nm (T280 ), which indicates
transmission or blocking UV-light, was observed with both agar/NC 3.2.3. Thickness and mechanical properties of the composite films
and agar/MCC composite films. The decrease in the transparency of Table 2 shows thickness and tensile properties of the agar,
the composite film is mainly due to the blocking of light passage agar/NC, and agar/MCC composite films. The thickness of neat agar
through the film by the crystalline cellulose. Previously, Reddy and film was 42.3 ± 2.0 ␮m, which was increased with the increas-
Rhim (2014) reported that the transmittance (T660 ) of agar films ing concentration of NC or MCC. The thickness was increased up
was affected by the addition of paper-mulberry pulp nanocellulose to 50.7 ± 2.0 and 88.7 ± 4.2 ␮m when 10 wt% of NC and MCC was
in the manner of linear relationship between the transmittance and incorporated, respectively. The increase in the film thickness is
the nanocellulose content. The present test results also showed that mainly due to the content of solid. Reddy and Rhim (2014) also
both transmittance (T660 and T280 ) values decreased with increase reported that the thickness of agar/cellulose nanocrystal composite
in both crystalline cellulose. However, they decreased logarithmi- films increased linearly with increase in their content of cellu-
cally rather than linearly with increase in the content of NC or lose nanocrystals. It is interesting to note that the dependence of
MCC. A linear relationship was observed between ln T vs. C with film thickness on the concentration of crystalline cellulose is quite
the coefficient of determination (R2 ) of 0.95–0.98 when a logarith- different between NC and MCC. The increase in thickness of the
mic model (ln T = aC + b, where T: transmittance, C: concentration of composite film was more pronounced with MCC compared with NC,
NC or MCC, a and b: constants) has been fitted to the transmittance which means that the NC forms a more compact structure with agar
data. On the contrary, when a linear model (T = aC + b) has been polymer matrix than MCC. This is mainly due to the reduced size of
fitted to the transmittance data, they also showed a linear rela- NC (100–500 nm) compared to MCC (20–200 ␮m) as observed with
tionship between T vs. C, however, the R2 values (0.94–0.97) were SEM and STEM images. This result indicates that the thickness of the
lower than those obtained from the logarithmic model. Consider- agar films was strongly dependent not only on the concentration
ing the Lambert–Beer law (ln Io /I = aCl, where Io and I: intensity of of crystalline cellulose but also the type of cellulose nanocrystals.

Table 2
Thickness and tensile properties of agar and agar/cellulose composite films.

Film Thickness (␮m) TS (MPa) E (%) EM (GPa)

Agar 42.3 ± 2.0a 46.7 ± 3.7e 15.7 ± 2.2bcd 1.34 ± 0.09f

NC1% 44.3 ± 1.8ab 48.7 ± 2.1e 15.4 ± 2.3bcd 1.37 ± 0.07f
NC3% 44.4 ± 1.6b 52.8 ± 3.3f 15.8 ± 2.5bcd 1.39 ± 0.10f
NC5% 46.3 ± 1.8b 45.3 ± 3.3e 17.6 ± 2.2e 1.23 ± 0.10e
NC10% 50.7 ± 2.0c 38.4 ± 2.6d 17.8 ± 2.5e 1.15 ± 0.15d
MCC1% 59.9 ± 4.2d 30.1 ± 3.2c 14.9 ± 2.5bc 0.96 ± 0.09c
MCC3% 70.9 ± 3.2e 24.9 ± 1.8b 16.4 ± 2.1cde 0.77 ± 0.08b
MCC5% 74.6 ± 5.0f 24.3 ± 1.9b 14.6 ± 2.7b 0.81 ± 0.07b
MCC10% 88.7 ± 4.2g 19.5 ± 2.4a 12.5 ± 2.7a 0.66 ± 0.07a

Values with the same superscript letter in the same column indicate that they are not statistically different (p < 0.05).
24 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26

Table 3
Moisture content, water vapor permeability, and water contact angle of agar and
agar/cellulose composite films.

Film MC (%) WVP (×10−9 g/m2 Pa s) WCA (◦ )

Agar 14.3 ± 0.5c

1.32 ± 0.06b
33.5 ± 0.8a
NC1% 14.3 ± 0.3c 1.18 ± 0.16b 34.9 ± 0.9b
NC3% 14.3 ± 1.2c 0.97 ± 0.03a 35.4 ± 1.3b
NC5% 13.4 ± 0.8bc 1.22 ± 0.08b 37.9 ± 2.9c
NC10% 12.9 ± 0.5b 1.32 ± 0.07b 37.0 ± 2.6bc
MCC1% 14.3 ± 0.7c 1.89 ± 0.25c 36.0 ± 0.9c
MCC3% 14.3 ± 1.1c 2.24 ± 0.24cd 36.2 ± 0.6c
MCC5% 9.4 ± 0.4a 2.48 ± 0.23d 37.7 ± 0.5c
MCC10% 9.1 ± 0.1a 3.07 ± 0.42e 37.2 ± 1.8c

Values with the same superscript letter in the same column indicate that they are
not statistically different (p < 0.05).

The mechanical properties of agar film were also greatly influ-

enced by the blending with the NC and MCC and they were also
influenced with the amount of the MCC or NC addition. The TS of
agar film increased from 46.7 MPa of the pristine agar film up to
52.8 MPa with 3 wt% addition of the NC, then, it decreased with Fig. 6. XRD of agar/NC and agar/MCC composite films (a) agar, (b) agar/NC3, and (c)
higher concentration of the NC. However, the strength of the com- agar/MCC3.
posite film with the highest amount of the NC (10 wt%) was still
comparable to that of the pristine agar film. In case of agar/MCC with a less compact interaction between MCC and agar polymer
composite films, the TS decreased with increasing concentration matrix to make a loosely formed composite film.
of MCC. This behavior may be attributed to the nano size of NC Results on the water contact angle (WCA) of the agar, agar/NC,
with high surface area, which may induce a strong interfacial inter- and agar/MCC composite films are also shown in Table 3. The WCA
action through hydrogen or ionic bonds between the NC and the is used as an indicator for the degree of surface hydrophobicity or
polymer matrix (Siqueira, Bras, & Dufresene, 2010). However, such hydrophilicity and the wettability of the film samples (Su et al.,
reinforcing effect of the NC was reduced at high level of NC addition 2010). Generally, a film with higher than 65◦ of WCA is considered
(10 wt%), probably due to the formation of aggregation of the NC as as hydrophobic (Vogler, 1998). The WCA of the control agar film
observed in the SEM analysis. High amount of NC may reduce the was 33.5 ± 0.8◦ , which indicates the agar film is very hydrophilic.
interaction between the NC and polymer matrix due to the non- Though the WCA increased significantly (p < 0.05) when both NC
uniform stress distribution in the film, and consequently reduce and MCC have been incorporated, they are still considered as
the strength of the composite film (Reddy & Rhim, 2014). The ten- hydrophilic. It is mainly due to the hydrophilic nature of both agar
sile test result suggests that there is an optimum concentration and celluloses.
of the filler to induce the maximum increment of the strength of Since 3 wt% of NC showed the best WVP and mechanical proper-
the composite films. Similar mechanical performance of cellulose ties, for XRD, FTIR and thermogravimetric analysis, films reinforced
nanofibers have been reported with the other biopolymers such as with 3 wt% were chosen for analysis.
starch and agar (Agustin et al., 2013; Reddy & Rhim, 2014). The elas-
tic modulus (EM) of the composite film, which measures degree of
3.2.5. XRD and FTIR of agar/cellulose composite films
stiffness of a film, showed similar trend as the TS. The elongation
X-ray diffraction (XRD) patterns of the agar, agar/NC, and
at break (E), which measures flexibility or ductility, of the film was
agar/MCC composite films are shown in Fig. 6. The control film
not changed significantly by the addition of NC and MCC.
showed a peak at 19.0◦ and slight shoulder at 13.9◦ , which

3.2.4. Moisture content (MC), water vapor permeability (WVP),

and water contact angle (WCA)
Table 3 shows the MC, WVP, and WCA of agar, agar/NC, and
agar/MCC composite films. The MC of the neat agar film was
14.3 ± 0.5%, and it decreased significantly (p < 0.05) when more
than 5 wt% of NC or MCC has been incorporated. This is probably
due to that the NC and MCC with less hydrophilic in nature
caused holding of less moisture in the film. The WVP of agar
film was (1.32 ± 0.06) × 10−9 g/m2 Pa s, and it decreased signifi-
cantly (p < 0.05) when NC was incorporated, while it increased
significantly when MCC was incorporated. The WVP decreased
linearly down to (0.97 ± 0.03) × 10−9 g/m2 Pa s when the NC was
added up to 3 wt%, then it increased slightly with increase in the
concentration of NC, however, it still less than or comparable
to the WVP value of neat agar films. The NC in the composite
film functions as an impermeable barrier against water vapor
permeation which increased the tortuous path for water vapor to
diffuse through the film and resulted in the decreased WVP of the
agar/NC composite films (Su, Huang, Yuan, Wang, & Li, 2010). On
the contrary, the incorporation of MCC resulted in the increase of
the WVP value, which might be due to the bulky structure of MCC Fig. 7. FTIR spectra of agar-based composite films.
 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26 25

Fig. 8. (a) TGA and (b) DTG of agar/cellulose composite films.

indicated high degree of crystallinity of agar (Freile-Pelegrín et al., films, the 4th stage of thermal degradation was observed around
2007; Wu, Geng, Chang, Yu, & Ma, 2009). There were no change in 330–360 ◦ C, which might be due to the thermal degradation of NC
the peak position of agar and even extra peaks of cellulose did not and MCC. The agar/crystalline cellulose composite films exhibited
appear might be because of less concentration of cellulose in the slightly higher thermal stability than the pristine agar film, which
agar matrix. was attributed to the presence of cellulose in the crystalline form
Fourier transform infrared spectroscopy (FTIR) spectra of the of NC and MCC. After the final thermal decomposition, there were
agar, agar/NC, and agar/MCC composite films are shown in Fig. 7. no significant difference in the residuals left at 600 ◦ C between the
The spectrum of agar exhibited various characteristic peaks in neat agar and agar/crystalline cellulose composite films. However,
the range from 3500 cm−1 to 700 cm−1 . The characteristic broad Reddy and Rhim (2014) reported that there was lower residual char
absorption band at about 3350 cm−1 indicates the stretching of left at 600 ◦ C inn the TGA test with agar films reinforced with cel-
hydroxyl (O H) groups (Wu et al., 2009). The peak at 2930 cm−1 lulose nanocrystals obtained by acid hydrolysis of paper mulberry
is attributed to the stretching of C H group of methane (Shankar pulp.
et al., 2014). The peak appeared at 1640 cm−1 is due to the
stretching vibration of the conjugated peptide bond formation by 4. Conclusions
amine (NH) and acetone groups. The vibrational peak at 1373 cm−1
is assigned to ester sulfate. The peaks at 1040 and 930 cm−1 Nanocellulose (NC) was prepared from micro-crystalline cel-
indicated C O stretching group of 3,6-anhydrogalactose. The peak lulose (MCC) using NaOH/urea dissolution method and followed
at 886 cm−1 is due to the C H stretching of ␤-galactose (Kanmani by regeneration, neutralization and ultrasonication. SEM and XRD
& Rhim, 2014). In case of agar/cellulose composite films, additional analysis on both NC and MCC revealed that the size of micro-
peaks were observed in addition to the peaks of agar polymer. crystalline cellulose decreased to nano level with decrease in the
These additional peaks are attributed to the NC and MCC in the crystallinity. NC and MCC was reinforced into the agar and film
composite films. The vibrational peak at 1430 cm−1 corresponds properties were evaluated. The film properties such as mechan-
to CH2 bending vibration that attributed to the crystalline peak of ical and water vapor barrier properties have been significantly
cellulose, however, the vibrational peak at 893 cm−1 is attributed improved with lower amount of the NC reinforcement (up to 3 wt%).
to the C O C stretching vibration of amorphous band of cellulose. This was due to the reinforcement of nanosize NC with high surface
The crystalline peak was more intense in agar/MCC, whereas, area caused the strong interaction between the homogeneously
amorphous peak was more intense in agar/NC composite films. dispersed NC and the polymer matrix. The agar/NC bionanocom-
posite film, which is completely biodegradable and biocompatible,
3.2.6. Thermal analysis of agar/cellulose composite films has a high potential to be used for biodegradable food packaging or
Thermal stability of neat agar, agar/NC, and agar/MCC compos- biomedical applications.
ite films was tested using TGA analysis and the results are shown
in Fig. 8. TGA curves show the weight loss pattern on thermal Acknowledgement
decomposition of the films (Fig. 8a) and DTG curves clearly show
the maximum decomposition temperature at each step of ther- This research was supported by the Agriculture Research Cen-
mal decomposition (Fig. 8b). All the films exhibited multiple steps ter (ARC 710003) program of the Ministry of Agriculture, Food and
of thermal degradation. The initial weight loss for all films was Rural Affairs, Korea.
observed around 90 ◦ C, which was due to the removal of mois-
ture from the films, which accounted for about 5–10% of weight References
loss (Shankar et al., 2014). The subsequent steps of degradation
were varied depending on the types of film. The second decom- Adsul, M., Soni, S. K., Bhargava, S. K., & Bansal, V. (2012). Facile approach for the
position begins at around 200 ◦ C and reached a maximum around dispersion of regenerated cellulose in aqueous system in the form of
nanoparticles. Biomacromolecules, 13, 2890–2895.
250 ◦ C, which was ascribed to the volatilization of glycerol added Agustin, M. B., Ahmmad, B., Richel, E. P. D. L., Jerico, L. B., Joel, R. S., & Fumihiko, H.
as a plasticizer (Martucci & Ruseckaite, 2010). Then the major (2013). Starch-based biocomposite films reinforced with cellulose
thermal decomposition were observed around 300 ◦ C, which was nanocrystals from garlic stalks. Polymer Composites, 34, 1325–1332.
Duncan, T. V. (2011). Application of nanotechnology in food packaging and food
attributed to the thermal decomposition of agar polymer (Shankar safety: Barrier materials, antimicrobials and sensors. Journal of Colloid and
et al., 2014). In the case of agar/crystalline cellulose composite Interface Science, 363, 1–24.
26 S. Shankar, J.-W. Rhim / Carbohydrate Polymers 135 (2016) 18–26

Freile-Pelegrín, Y., Madera-Santana, T., Robledo, D., Veleva, L., Quintana, P., & from mixtures of ␬-carrageenan, ␫-carrageenan, and alginate. Food
Azamar, J. A. (2007). Degradation of agar films in a humid tropical climate: Hydrocolloids, 47, 140–145.
Thermal, mechanical, morphological and structural changes. Polymer Reddy, J. P., & Rhim, J. W. (2014). Characterization of bionanocomposite films
Degradation and Stability, 92, 244–252. prepared with agar and paper-mulberry pulp nanocellulose. Carbohydrate
Gennadios, A., Weller, C. L., & Goodings, C. H. (1994). Measurement errors in water Polymers, 110, 480–488.
vapor permeability of high permeable hydrophilic edible films. Journal of Food Rhim, J. W., Park, H. M., & Ha, C. S. (2013). Bio-nanocomposites for food packaging
Engineering, 21, 395–409. applications. Progress in Polymer Science, 38, 1629–1652.
Ghanbarzadeh, B., & Almasi, H. (2011). Physical properties of edible emulsified Savadekar, N. R., Karande, V. S., Vigneshwaran, N., Kadam, P. G., & Mhaske, S. T.
films based on carboxymethyl cellulose and oleic acid. International Journal of (2015). Preparation of cotton linter nanowhiskers by high-pressure
Biological Macromolecules, 48, 44–49. homogenization process and its application in thermoplastic starch. Applied
Isobe, N., Noguchi, K., Nishiyama, Y., Kimura, S., Wada, M., & Kuga, S. (2013). Role of Nanoscience, 5, 281–290.
urea in alkaline dissolution of cellulose. Cellulose, 20, 97–103. Shankar, S., Reddy, J. P., Rhim, J. W., & Kim, H. Y. (2015). Preparation,
Kalia, S., Dufresne, A., Cherian, B. M., Kaith, B. S., Avérous, L., Njuguna, J., et al. characterization, and antimicrobial activity of chitin nanofibrils reinforced
(2011). Cellulose-based bio- and nanocomposites: A review. International carrageenan nanocomposite films. Carbohydrate Polymers, 117, 468–475.
Journal of Polymer Science,. Article ID: 837875. Shankar, S., & Rhim, J. W. (2015). Amino acid mediated synthesis of silver
Kanmani, P., & Rhim, J. W. (2014). Antimicrobial and physical–mechanical nanoparticles and preparation of antimicrobial agar/silver nanoparticles
properties of agar-based films incorporated with grapefruit seed extract. composite films. Carbohydrate Polymers, 130, 353–363.
Carbohydrate Polymers, 102, 708–716. Shankar, S., Teng, X., & Rhim, J. W. (2014). Properties and characterization of
Li, Q., Zhou, J., & Zhang, L. (2009). Structure and properties of the nanocomposite agar/CuNPs bionanocomposite films prepared with different copper salts and
films of chitosan reinforced with cellulose whiskers. Journal of the Polymer reducing agents. Carbohydrate Polymers, 114, 484–492.
Science B: Polymer Physics, 47, 1069–1077. Siqueira, G., Bras, J., & Dufresene, A. (2010). Cellulosic bionanocomposites: A
Maiti, S., Jayaramudu, J., Das, K., Reddy, S. M., Sadiku, R., Ray, S. S., et al. (2013). review of preparation properties and applications. Polymers, 2, 728–765.
Preparation and characterization of nano-cellulose with new shape from Su, J. F., Huang, Z., Yuan, X. Y., Wang, X. Y., & Li, M. (2010). Structure and properties
different precursor. Carbohydrate Polymers, 98, 562–567. of carboxymethyl cellulose/soy protein isolate blend edible films crosslinked
Martucci, J. F., & Ruseckaite, R. A. (2010). Three-layer sheets based on gelatin and by maillard reactions. Carbohydrate Polymers, 79, 145–153.
poly(lactic acid), part 1: Preparation and properties. Journal of Applied Polymer Tang, X. G., Kumar, P., Alavi, S., & Sandeep, K. P. (2012). Recent advances in
Science, 118, 3102–3110. biopolymers and biopolymer-based nanocomposites for food packaging
Mittal, A., Katahira, R., Himmel, M. E., & Johnson, D. K. (2011). Effects of alkaline or materials. Critical Reviews in Food Science and Nutrition, 52, 426–442.
liquid-ammonia treatment on crystalline cellulose: Changes in crystalline Vogler, E. A. (1998). Structure and reactivity of water at biomaterial surfaces.
structure and effects on enzymatic digestibility. Biotechnology for Biofuels, 4, Advances in Colloid and Interface Science, 74, 69–117.
41–56. Wu, Y., Geng, F., Chang, P. R., Yu, J., & Ma, X. (2009). Effect of agar on the
Moon, R. J., Martini, A., Nairn, J., Simonsen, J., & Youngblood, J. (2011). Cellulose microstructure and performance of potato starch film. Carbohydrate Polymers,
nanomaterials review: Structure, properties and nanocomposites. Chemical 76, 299–304.
Society Review, 40, 3941–3994. Yu, H., Yan, C., Lei, X., Qin, Z., & Yao, J. (2014). Novel approach to extract thermally
Paula, G. A., Benevides, N. M. B., Cunha, A. P., de Oliveira, A. V., Pinto, A. M. B., stable cellulose nanospheres with high yield. Materials Letters, 131,
Morais, J. P. S., et al. (2015). Development and characterization of edible films 12–15.