Journal Information
Download PDF
More article options
Original Article
DOI: 10.1016/j.jmrt.2019.02.005
Open Access
Available online 9 April 2019
Sisal cellulose and magnetite nanoparticles: formation and properties of magnetic hybrid films
Daiana M. Furlana, Daniella Lury Morgadoa, Adilson J.A. de Oliveirab, Ângelo D. Facetob, Daniel A. de Moraesc, Laudemir C. Varandac, Elisabete Frollinia,
Corresponding author

Corresponding author.
a Macromolecular Materials and Lignocellulosic Fibers Group, Center for Research on Science and Technology of BioResources, Institute of Chemistry of São Carlos, University of São Paulo, São Carlos, São Paulo, Brazil
b Physics Department, Federal University of São Carlos, São Carlos, São Paulo, Brazil
c Colloidal Materials Group, Center for Research on Science and Technology of BioResources, Institute of Chemistry of São Carlos, University of São Paulo, São Carlos, São Paulo, Brazil
This item has received

Under a Creative Commons license
Received 07 September 2018, Accepted 28 February 2019
Article information
Full Text
Download PDF
Figures (8)
Show moreShow less
Tables (2)
Table 1. Nitrogen and lithium contents, determined by elemental analysis and atomic absorption, respectively, for the starting cellulose and films (FC, FCFe0.5, FCFe1.4, and FCFe3.0).
Table 2. Temperatures of the thermal decomposition of FC, FCFe0.5, FCFe1.4, and FCFe3.0.
Show moreShow less

In this study, sisal cellulose/magnetite-nanoparticle (Fe3O4 NPs; 0.5, 1.4, and 3.0gL−1) hybrid films (denoted as FCFe0.5, FCFe1.4, and FCFe3.0, respectively) were prepared by casting, using the solvent system LiCl/DMAc. Sisal was chosen as a cellulose source because it is a fast-growing plant, in contrast to the long cycle of woody trees, and Brazil accounts for most of the sisal produced in the world. Fe3O4 NPs were chosen owing to their excellent properties (superparamagnetic behavior at room temperature, high chemical stability, and low toxicity). The synthesized magnetite NPs (coated with oleic acid and oleylamine to prevent agglomeration during synthesis) were spherical with an average diameter of 5.1±0.5nm (transmission electron microscopy analysis; TEM). X-ray diffraction analysis showed that the NPs were satisfactorily incorporated into the cellulose films (as confirmed by TEM) and that their presence favored the formation of cellulose crystalline domains. FCFe1.4 and FCFe3.0 exhibited higher tensile strengths (14.3MPa and 12.1MPa, respectively) than the neat cellulose film (9.9MPa). The moduli of elasticity of FCFe0.5, FCFe1.4, and FCFe3.0 were 1650, 1500MPa, and 780MPa, respectively, lower than that of the cellulose film (1860MPa), indicating that the incorporation of NPs in the cellulosic matrix decreased the films’ stiffness. Hybrid films exhibited high magnetizations at 300K, i.e., 23.0emug−1 (FCFe0.5), 31.0emug−1 (FCFe1.4), and 37.0emug−1 (FCFe3.0), as well as no magnetic hysteresis and remanent magnetization (Mr) null, namely, a superparamagnetic behavior at room temperature. The results obtained suggest several applications of hybrid films based on cellulose and magnetite, such as biomedical applications, miniaturized electronic devices, and advanced catalysis.

Sisal cellulose
Magnetite nanoparticles
Magnetic hybrid films
Full Text

The sustainable use of biomass and its recovery to convert it into profitable products and materials require effort, mainly with regard to its exploration in applications different from those already considered. In this scenario, cellulosic compounds emerge as some of the most important raw materials to consider.

Cellulose is the most abundant natural polymer obtained from renewable sources, including jute, ramie, sisal, sugarcane, and cotton. Sisal, as the other plants mentioned, is a source with fast growth and regeneration cycle. Also, the world's largest producer of sisal is Brazil, and these aspects led to the choice of this fiber as a source of cellulose in the present study [1–5]. The chemical structure of cellulose differs from those of many synthetic polymers owing to its hydrophilic character, stereoregularity, and polyfunctionality [6]. These characteristics, combined with its low cost, make cellulosic pulp very attractive for use as starting material in the preparation of new materials and composites [7], cellulose nanofibril hydrogel [8], microcapsules containing cellulose nanofibrils [9], and organic–inorganic hybrids, such as those considered in this study.

Magnetic nanoparticles (MNPs) can be incorporated into different materials, allowing the creation of end-products with diverse properties, besides magnetic properties [10,11]. Composites consisting of MNPs and cellulose have attracted great interest owing to the synergy between the properties derived from the individual components. This type of hybrid material may exhibit different structural and functional properties, as it also represents a new concept material with advantages such as biocompatibility and biodegradability [12–16]. Moreover, the dispersion of magnetic nanoparticles (MNPs) in an organic polymer to form magnetic composites may improve its optical, mechanical, and other properties [17]. Several applications can be envisaged for cellulose–MNP materials, such as conductive paper [18], nanocatalysts for removal of textile dyes [19], catalytic and antibacterial activities [20], and transparent films for magneto-optical applications [21].

Iron oxides, in particular magnetite (Fe3O4), represent the magnetic particles that are most commonly associated with a polymer matrix in a nanometric scale owing to their excellent magnetic properties, such as superparamagnetic behavior at room temperature, high chemical stability, exceptional biocompatibility of their surfaces [22], low toxicity compared to both metals or other metal-oxide NPs, and the facility and low cost of the procedures available for their preparation [23].

Cellulosic materials have been used for the preparation of materials containing an organic–inorganic interface, in the form of fibers, films, and microspheres [12–14]. Hybrid films obtained by mixing the magnetic component with a cellulose derivative, namely, cellulose acetate, have been reported by Shim et al. [24]. However, the number of studies related to obtaining magnetic composite films from underivatized cellulose is still small [25], mainly because cellulose dissolves in a limited number of solvents. In this context, herein we prepared sisal cellulose/Fe3O4-nanoparticle (FCFe) hybrid films by casting, using LiCl/DMAc as the solvent system. The influence of different amounts of MNPs on the properties of cellulose films (FCs) was evaluated, with the aim of preparing cellulosic films with enhanced mechanical and magnetic properties.

2Materials and methods2.1Materials

Sisal pulp was provided by the Lwarcel Company (Lençois Paulista, São Paulo, Brazil), where it was obtained by the Kraft pulping process, using an elemental chlorine-free bleaching sequence [25]. Lithium chloride (LiCl, Synth), N,N-dimethylacetamide (DMAc, Vetec), sodium hydroxide, ethanol, methanol, and hexane (Synth), cupriethylenediamine (Qeel), 1,2-hexadecanodiol, benzyl ether, oleic acid, oleylamine, and iron (III) acetylacetonate (Sigma-Aldrich) were used without further purification—except for DMAc, which was purified by distillation with CaH2, followed by storage over 4Å molecular sieves, and for LiCl, which was dried under vacuum overnight at 60°C and stored in a desiccator before use.

The properties of the sisal pulp (viscometric degree of polymerization=723 (DPv), crystallinity=70%, and α-cellulose=88.7%) were determined as described elsewhere [26]. Prior to its use, the pulp was treated with NaOH 20% (wt%) solution (mercerization treatment), as previously described [26,27]. The mercerized sisal cellulose was dried at room temperature for 24h and subsequently in an air-circulating oven at 105°C until it had a constant weight. The mercerized sisal cellulose exhibited DPv=670, crystallinity=58.0%, and α-cellulose content=90.5%, which upgraded its quality as a dissolving pulp [1], mainly owing to the decrease in crystallinity.

2.2Methods2.2.1Synthesis of Fe3O4 NPs

The superparamagnetic Fe3O4 NPs were prepared with the thermal decomposition method in the presence of polyols, following an adapted experimental procedure described previously [28]. In a three-necked round-bottom flask, iron (III) acetylacetonate (1mmol), 1,2-hexadecanodiol (5mmol), oleic acid (3mmol), oleylamine (3mmol), and benzyl ether (20mL) were mixed and heated to 100°C under magnetic stirring for 15min and a nitrogen flow. The NPs were coated with oleic acid and oleylamine to prevent agglomeration during synthesis. Afterwards, under a blanket of nitrogen, the mixture was heated to 200°C for 2h and then subjected to reflux (298°C) for 1h. The black-colored mixture was cooled to room temperature by removing the heat source. Under ambient conditions, ethanol and hexane were added to the mixture, and a black material was isolated by centrifuging (9000rpm, 10min). This procedure was applied until the complete removal of the undispersed residue, and the resulting material was then redispersed into hexane and stored at room temperature.

2.2.2Elaboration of films

Cellulose film (FC). The FC was prepared in a LiCl/DMAc solvent system following a previously described experimental procedure [4,26]. Cellulose (1.5g) and LiCl (1.5g) were mixed in a glass reactor equipped with a mechanical stirrer and a vacuum pump. The system was heated to 110°C and maintained at that temperature for 30min. Next, 45mL of DMAc was added, and the vacuum pump was turned off and replaced by a reflux condenser under a flow of nitrogen. The mixture was heated to 160°C and stirred for 1.5h, cooled down to room temperature, and allowed to stir overnight to obtain a homogeneous solution. After this period, the solution was immediately cast on glass Petri dishes (diameter 9.6cm) and left to stand at room temperature until a gel-like substance with the minimal consistency for washing was formed. The material was then exhaustively washed with distilled water until the solvent system was entirely removed. To confirm the complete elimination of LiCl in this step, the conductivity of the washing residue was measured repeatedly until it equaled that of the starting water. The residual water was eliminated by drying the formed film at room temperature for one day, then sandwiching it between two Teflon® plates, and finally drying it at 60°C under reduced pressure until it had a constant weight.

FCFe hybrid films. The hybrid films were prepared from LiCl/DMAc solutions, following the same procedure as that mentioned for the FC. Fe3O4 NPs (dried for one day at room temperature to eliminate hexane) were dispersed in DMAc using an ultrasound bath just before the use of the solvent. A set of the films of cellulose/Fe3O4 NPs were prepared by changing the concentration of Fe3O4 NPs dispersed in DMAc to 0.5, 1.4, and 3.0gL−1, and the hybrid films were denoted as FCFe0.5, FCFe1.4, and FCFe3.0, respectively.

2.3Characterization of starting materials, FC and FCFe films

Elemental analysis and atomic absorption. The possible presence of residual solvent system (LiCl/DMAc) in the films was evaluated by elemental analysis (nitrogen content; CE Instruments CHNS-O model EA-1110) and atomic absorption (lithium content; Varian atomic absorption spectrometer model AA240FS). Prior to elemental analysis, the films were sprayed using liquid nitrogen. The atomic absorption analysis was performed via wet dissolution, digesting 4mg of the dried samples (pure FC or hybrid films) in 1mL of concentrated H2SO4 plus heating, followed by the addition of hydrogen peroxide dropwise until a clear solution was observed. The solution was cooled to room temperature and diluted to a 50-mL volume using a KCl 2ppm solution.

Crystallinity index (Ic). X-ray diffraction (XRD) analysis was used to evaluate the Ic of cellulose, Fe3O4 NP phase formation, and its incorporation into the hybrid films. Measurements were performed using a Rigaku-Denki® RINT2000 diffractometer with a horizontal goniometer of high angle operating at 50kV/100mA and a Bruker D8 Advanced diffractometer, both using Cu Kα radiation (λ=1.542Å). The diffraction patterns were recorded in the range 2θ=10–80° and at a scanning speed of 2°min−1 and 0.02°/0.2s, respectively. The Ic of the materials was evaluated as described elsewhere [1,3], using Eq. (1):

I1 is the minimum intensity of the noncrystalline part of cellulose (18° for cellulose I and 16° for cellulose II),

I2 is the maximum intensity of the crystalline part of cellulose (22°2θ23° for cellulose I and 18°2θ22° for cellulose II).

Thermogravimetric analysis (TGA). TGA was conducted using a Shimadzu TGA-50 instrument for samples of cellulose, Fe3O4 NPs, and the hybrid films. Approximately 5.0mg of the sample was weighed and heated from room temperature to 800°C (20°Cmin−1) under N2 flow (50mLmin−1).

The mass percentage of Fe3O4 incorporated into the films was determined via TGA under air atmosphere (figures not shown) under the same conditions described above, from the difference of the final masses of FCFe and FC.

Morphological analysis. The morphology of the films was examined by scanning electron microscopy (SEM) using a LEO 440 ZEISS/LEICA SEM system at an accelerating voltage of 20kV, equipped with an energy-dispersive X ray spectrometer (EDX; EDX LINK ANALYTICAL, Isis System Series 200). Samples were metalized with gold before examination. SEM images of the surface and of the cross section of the films were assessed. The presence of iron-oxide NPs in the hybrid films was investigated by EDX. The size, shape and particle-size distribution of the Fe3O4 NPs were analyzed by transmission electron microscopy (TEM) using a Philips CM120 microscope operating at 120kV and a JEOL JEM2100 microscope operating at 200kV. A sample of as-synthesized NPs dispersed in hexane was diluted, sonicated, and then dropped on a Formvar®- and carbon-coated copper grid, and the hybrid films were fragmented, suspended in isopropanol, and dropped onto the sample holder. In both cases, the solvent was slowly evaporated at room temperature and the copper grids remained under vacuum for 12h. A statistical analysis (average diameter and standard deviation) was performed by counting around 200 NPs using the free software ImageJ version 1.42q.

Tensile strength. The tensile tests were performed using a TA Instruments model DMA Q800 operating in tension-film mode at room temperature, according to the equipment manual instructions for this type of analysis. The film samples were strained at a constant rate of 1Nmin−1 until 18N or failure, with a preload of 0.001N. Samples 5.7mm in width and 10mm in gauge length were used for the tests. At least three samples from each film were tested, and the errors concerning the average results reported (section 3.3.5) ranged from ±3 to ±7%.

Magnetic measurements. The hysteresis loops of the Fe3O4 NPs and the hybrid films were measured by a vibrating sample magnetometer (VSM; EG&G Princeton Applied Research) at300K, with a static magnetic field applied parallel to the film. The applied field was systematically varied from −10.0 to 10.0kOe. Magnetic susceptibility measurements were performed using an MPMS3 from Quantum Design Squid magnetometer. The magnetic susceptibility data of the Fe3O4 particles were obtained from field-cooling/zero-field cooling (ZFC) curves by heating the sample from 5 to 300K with an applied field of 50Oe. First, the sample was cooled down to 5K without any applied field to obtain ZFC measurements; then, the data were obtained after cooling down from 300K to 5K with an applied field of 50Oe, i.e., field-cooling protocol.

3Results and discussion3.1Nitrogen and lithium analyses

In order to confirm the absence of DMAc, the solvent used in the preparation of the films, the content of nitrogen was evaluated by elemental analysis. Another source of nitrogen in the hybrid films was oleylamine (C18H37N), which was added during the synthesis, along with oleic acid, to avoid aggregation of the NPs. The starting sisal cellulose and its film, as well as the hybrid films, were also analyzed by atomic absorption to verify the presence of lithium from residual LiCl, which was used together with DMAc as the solvent system in the preparation of the films (Table 1).

Table 1.

Nitrogen and lithium contents, determined by elemental analysis and atomic absorption, respectively, for the starting cellulose and films (FC, FCFe0.5, FCFe1.4, and FCFe3.0).

Samples  Nitrogen content (%)  Lithium content (mgL−1
Starting cellulose  0.005  a 
FC  0.2  0.001 
FCFe0.5  0.4  0.001 
FCFe1.4  0.2  0.003 
FCFe3.0  0.4  0.005 

LoD, limit of detection; LoD=0.001mgLiL−1.

The results showed that the successive washes with distilled water were efficient in promoting the elimination of DMAc, because very small amounts of nitrogen (equal to or less than 0.4%) were found in the films after the washing and drying process. The amounts of nitrogen in the hybrid films may also be assigned to the presence of oleylamine on the surfaces of magnetite NPs, as mentioned. The atomic absorption results showed that practically no lithium residues were found in all samples.


The average size, shape, and size distribution of the as-synthesized magnetite NPs was investigated by TEM (Fig. 1).

Fig. 1.

(a) TEM images of the as-synthesized magnetite NPs and (b) Gaussian fit of the particle-size distribution of the magnetite.


The TEM image (Fig. 1a) indicates that the synthesis of magnetite using the modified polyol method gave rise to spherical NPs with an average diameter of 5.1±0.5nm. Varanda et al. [29] obtained Fe3O4 NPs with 5.0±0.4nm diameter by employing the same protocol. Owing to the low degree of polydispersity (σ<10%, Fig. 1b) obtained for this synthesis, the NP system can be considered as monodisperse [30]. MNPs behave as a magnetic monodomain below a certain critical diameter, usually around 10–20nm, and exhibit a superparamagnetic behavior. Thus, individual NPs have a large constant magnetic moment and behave like a giant paramagnetic atom with a quick response to applied magnetic fields. These characteristics make the superparamagnetic NPs very attractive for a wide range of applications, such as in the biomedical field, owing to the size of the NPs being compatible with that of biological agents, such as cells [31].

3.3Film characterization3.3.1SEM–EDX

SEM images of the surfaces of the FC and hybrid films (FCFe0.5, FCFe1.4, and FCFe3.0) are shown in Fig. 2, together with their respective EDX spectra.

Fig. 2.

SEM micrographs of the surfaces of the films and the respective EDX spectrum, acquired from SEM images of (a) FC, (b) FCFe0.5, (c) FCFe1.4, and (d) FCFe3.0.


The SEM images of the FC (control sample, Fig. 2a) and hybrid films (Fig. 2b–d) showed a fibrous structure on their surfaces. These structures were probably formed from the cellulose chain aggregates, which further self-organized during the removal of the solvent from the films, forming fibrous structures. The surface morphology of the hybrid films (FCFe0.5, FCFe1.4, and FCFe3.0) were similar to that of the FC, suggesting that the introduction of Fe3O4 NPs in the cellulosic matrix at concentrations of 0.5, 1.4, and 3.0gL−1 did not affect the surface morphology of the cellulose films.

The EDX spectra of the hybrid films (Fig. 2b–d) indicated the presence of C, O (from cellulose chains), and Fe in these films, which shows that Fe3O4 NPs at different concentrations were efficiently incorporated into the cellulosic matrix during its dissolution in DMAc/LiCl. The NPs remained impregnated in the cellulosic matrix after the formation of the films, and were not removed by washing with water, which indicates strong cellulose–magnetite interactions. In addition to the C, O, and Fe, Al and Si were detected in the hybrid film FCFe1.4 (Fig. 2c), which are also present in the control film (FC). Silica (SiO2) represents the major inorganic portion of the ash in higher plants. Behera et al. [32] reported an ash content of 1.27% for sisal, indicating a very low content of Si in this fiber. The presence of aluminum may be due to residues of the processing.

Fig. 3 displays the cross-section SEM images of the films. The cross section of the FC (Fig. 3a) was more homogeneous than those of the hybrid films (Fig. 3b–d). When Fe3O4 NPs were present in the cellulosic matrix, the cross section of the films showed that there was a greater distance from the cellulose microfibers (shown by arrows) and, consequently, an increase in the thickness of the films (Fig. 3b–d) compared to the FC (Fig. 3a).

Fig. 3.

SEM micrographs of the cross section of the films: (a) FC, (b) FCFe0.5, (c) FCFe1.4, and (d) FCFe3.0.


In the XRD image of the FC (Fig. 4a) the typical diffraction pattern of the polymorphic form attributed to cellulose II (mercerized pulp) [1–5], with reflections at 2θ values of 22°, 20°, and 13° corresponding to the (002), (101), and (101¯) planes, respectively, can be observed [33]. Besides the diffraction peaks of crystal cellulose II, hybrid films display distinct peaks that are characteristic of the crystallographic structure of the inverse spinel of magnetite. As expected, the relative intensity of the magnetite diffraction increased with NP concentration in the hybrid film, as indicated in the amplified range region in Fig. 4b. As shown in Fig. 4b, in the film containing the highest concentration of magnetite (FCFe3.0), we observed reflections of magnetite at 2θ values of 30.5°, 35.8°, 43.2°, 57.6°, and 63.0° (Joint Committee on Powder Diffraction Standards, JCPDS, card no. 19-629), assigned to the (220), (311), (400), (511), and (440) planes, respectively. In all hybrid film samples, the XRD reflections from magnetite were broadened, in agreement with the nanometer size shown in the TEM analysis. Therefore, magnetic Fe3O4 NPs were satisfactorily incorporated into the cellulose in the preparation of the films and were not removed after thorough washing with water, which indicates a strong cellulose–magnetite interaction.

Fig. 4.

XRD patterns of the films (a) obtained with a scanning angle from 5° to 80° in 2θ; (b) amplification of the region corresponding to the diffraction of magnetite (Fe3O4).


The cellulosic matrix of the FCFe0.50, FCFe1.4, and FCFe3.0 hybrid films exhibited a larger proportion of crystalline regions (Fig. 4), of 70.0%, 64.0%, and 72.0%, respectively, than that of the neat film (FC, 60.0%), indicating that the presence of magnetite NPs favored the formation of cellulose-chain crystalline domains.


The FCFe0.5, FCFe1.4, and FCFe3.0 hybrid films analyzed by TEM were prepared directly on the copper sample holder, which eliminated the need to use an ultramicrotome to cut the films. To confirm the presence of NPs, the TEM images were obtained from the hybrid films and the distribution of NPs was also evaluated qualitatively in the cellulosic matrix. Fig. 5 shows the TEM images for the FCFe0.5 and FCFe1.4 hybrid films, and the TEM images for the FCFe3.0 (not shown) was similar to that of FCFe1.4. The observed dark regions (Fig. 5c) may be a consequence of NP agglomeration, which may have been facilitated by the formation of the films analyzed via TEM directly on the reduced size surface of the sample support. Nonetheless, the NPs are well dispersed in most of the image shown (Fig. 5c), as they were coated with oleic acid and oleylamine to avoid agglomeration.

Fig. 5.

TEM images of the hybrid films: (a) FCFe0.5 and (b) FCFe1.4.


Fig. 6 shows the thermogravimetric (TG) and differential TG (DTG) curves of the control film (FC) and of the hybrid films.

Fig. 6.

(a) TG (insert: magnification from 25°C to 120°C) and (b) DTG curves of the control film (FC) and its hybrid films (FCFe0.5, FCFe1.4, and FCFe3.0) (N2 atmosphere, 50mL min−1).


Table 2 summarizes the TGA data derived from the TG and DTG curves (Fig. 6). It also shows the magnetite content, in weight percentage with respect to cellulose, which was determined using TGA in air atmosphere (Section 2.3, figures not shown), because in this condition the percentage of the residue is indicative of the amount of magnetite incorporated into the film. It should be noted that the values shown in Table 2 correspond to the weight percentage of magnetite NPs related to the total weight of the film, whereas 0.5, 1.4, and 3.0 correspond to the concentration (gL−1) of NPs dispersed in LiCl/DMAc during the preparation of the film.

Table 2.

Temperatures of the thermal decomposition of FC, FCFe0.5, FCFe1.4, and FCFe3.0.

Sample  Second stepaFe3O4 weight (wt%)b 
  Weight loss (Ti to Tf) (%)  Ti (°C)  Tf (°C)  Td (°C)   
FC  66  294  392  355  – 
FCFe0.5  63  272  363  343  1.3 
FCFe1.4  61  276  362  343  4.1 
FCFe3.0  56  280  365  343  6.8 

N2 atmosphere: Ti, initial temperature; Tf, final temperature; Td, temperature of maximum rate of weight loss.


From TGA performed in air atmosphere (figures not shown).

The TG curves (Fig. 6) of the raw material, FC, and the three hybrid films (FCFe0.5, FCFe1.4, and FCFe3.0) showed three stages of weight loss. The low amount of DMAc in the films, associated to their boiling point (about 160°C), indicates that the first step of weight loss observed in the films’ TGA should be attributed mainly to humidity, and not to DMAc, according to the results presented of the elemental analysis (Table 1). The second step corresponds to the main decomposition of cellulose, leading to levoglucosan, which is produced by the scission of the 1,4-glucosidic linkage in cellulose, followed by the intramolecular rearrangement of the monomer units [34]. This was observed for cellulose and FC from 280 to 400°C.

It can be seen from Fig. 6a that in the TG curves obtained for the hybrid films the second step of weight loss started at a lower temperature compared to FC (Table 2). FC exhibited a Td (the temperature of maximum rate of weight loss) of approximately 355°C, and the presence of magnetite slightly shifted Td to a lower temperature (Fig. 6b).

3.3.5Tensile properties

The tensile strength and elongation at break (%) of the control FC and its hybrid films with different concentrations of magnetite were evaluated using experimental stress–strain curves (not shown). The FC exhibited tensile strength and percentage elongation at break values of 9.9±0.1MPa and 0.66±0.07%, respectively. The hybrid film with the lower concentration of magnetite (FCFe0.5) showed values close to those of FC (9.6±0.5MPa and 0.68±0.01%). The films prepared with higher concentrations of magnetite, FCFe1.4 and FCFe3.0, showed higher tensile strength and elongation than FCFe0.5 and the FC. The tensile strength and elongation of FCFe1.4 were 14.3±1.2MPa and 1.1±0.2%, and for FCFe3.0 were 12.1±0.6MPa and 1.55±0.05%, respectively. These results indicate that the increase of NP concentration in the cellulosic matrix favored sliding between the cellulose chains, thereby facilitating the stretching of the films during stress.

The low percentage of elongation shown by the films characterizes them as rigid, which is a consequence of the low flexibility of the cellulose chains, of the strong intermolecular hydrogen bonds between the hydroxyl groups, and of their high Ic (between 60.0% and 72.0%, as mentioned). The moduli of elasticity are shown in Fig. 7.

Fig. 7.

Modulus of elasticity for the FC and its hybrid films, FCFe0.5, FCFe1.4, and FCFe3.0.


The moduli of elasticity of the hybrid films FCFe0.5, FCFe1.4, and FCFe3.0 were 1650±160MPa, 1500±190MPa, and 780±100MPa, respectively, and that of the FC was 1860±280MPa. These results indicate a tendency of the modulus to decrease from FC to FCFe1.4 and a significant decrease in this property for FCFe3.0. The incorporation of magnetite NPs in the cellulosic matrix provided lower stiffness to the hybrid films compared to the FC (Fig. 7). Although the presence of NPs has favored the formation of crystalline domains, as mentioned, the results of elongation and modulus indicated that the presence of NPs among the cellulose chains had gradually decreased the resistance to deformation of the films.

3.3.6Magnetic properties

Fig. 8a shows the magnetization curve of the magnetite as a function of the applied magnetic field, obtained by the SQUID technique. The analyses were performed at two different temperatures, 5K and 300K, in order to evaluate the magnetic behavior of magnetite NPs as a function of temperature.

Fig. 8.

(a) Magnetization versus applied magnetic field of magnetite at different temperatures (5K and 300K). Inset: amplification of the magnetization curve in the region of low external field (−400 to 400Oe); (b) zero-field cooling (ZFC) and field-cooling (FC) curves for magnetite NPs, obtained by applying an external magnetic field of 50Oe; (c) magnetization curves of the FC, FCFe0.5, FCFe1.4, and FCFe3.0 films at 300K as a function of applied external field parallel to the film surfaces (g: mass of Fe3O4 in the films, determined by TGA in air).


The hysteresis loop of the magnetite NPs as a function of the applied magnetic field in a range from −70kOe to 70kOe showed that the saturation magnetization (Ms; where all the magnetic moments of the materials remain aligned in the applied magnetic field direction) values at 5K and 300K (room temperature) are 68.0 and 58.0emug−1, respectively (Fig. 8a). These values are lower than that of the saturation magnetization of bulk magnetite and maghemite (dimensions greater than nanometer scale), 92.0emug−1 and 82.0emug−1, respectively, as reported in the literature [35]. This is due to the fact that the contribution to the magnetization from the outermost layers of the material is small owing to the disorganization of the magnetic moments of the surface atoms (surface anisotropy), which promotes the formation of a magnetically inactive layer on the surface of the particles (dead surface layer). A similar behavior has been already reported in the literature [36–38].

The magnification of the magnetization curves around the central region, shown in the inset of Fig. 8a, showed that at 300K, the magnetite presented a remnant magnetization (Mr; residual magnetization of the material when the external magnetic field is removed) equal to 0.6emug−1 and the coercivity (Hc; field intensity necessary to override the remanent magnetization) reaches 26Oe. This field is associated to the remanence of the superconductor magnet used in the MPMS3 SQUID magnetometer.

The very low coercivity and almost zero remanent magnetization values indicate that the magnetite synthesized in this study has a superparamagnetic behavior at room temperature, with absence of magnetic hysteresis. At 5K, the magnetite showed a blocked superparamagnetic behavior, with a slight opening of the hysteresis loop, showing Hc and Mr values of around 50.0Oe and 4.0emug−1, respectively (Fig. 8a). These results were confirmed by ZFC and field-cooling curves (Fig. 8b), which were obtained to characterize the superparamagnetic behavior of the NPs, as well as to determine their blocking temperature, TB (the temperature from which the thermal fluctuations of the magnetization are destabilized or unblocked).

In order to obtain the ZFC curve, the sample's temperature is first lowered slowly without any external magnetic field, until a low temperature below the blocking temperature (for superparamagnetic materials). Then, a small magnetic field (usually of the order of tens of Oersteds) is applied to the sample, and the magnetization of the system is monitored as a function of the increase in the temperature, keeping the field fixed throughout the measurement. Unlike the ZFC curve, the field-cooling curve is obtained by measuring the magnetization while the sample is cooled below its critical temperature in the presence of an applied external magnetic field [31,39].

The ZFC/field-cooling curves between 5 and 300K showed that from 40K, the two curves coincide and separate below the quoted temperature. Thus, 40K was found to be the blocking temperature for the magnetite NPs synthesized in this study (Fig. 8b). It can be observed that the maximum ZFC curve coincides with the TB value, which is normally observed for the ZFC/field-cooling curves for superparamagnetic NPs [40].

Above the blocking temperature, in this case TB=40K, the NPs have enough energy to overlap the anisotropic energy barrier, i.e., the required energy to revert the magnetization from one stable state to another and become superparamagnetic and unblocked (their magnetic moment varies with the time of the magnetic measurement). This behavior is in accordance with the magnetization curves as a function of the applied magnetic field (Fig. 8a), where Hc and Mr at 300K have almost zero values because 300K is above the TB for magnetite (40K), where the NPs have superparamagnetic behavior. However, at 5K, a slight opening of the hysteresis loop of the NPs and a superparamagnetic unblocked behavior are observed because the temperature of the analysis (5K) is lower than TB (40K). Superparamagnetic NPs are very attractive for a wide range of biomedical applications, mainly because aggregation of the particles is avoided when the external magnetic field is removed owing to the magnetization of superparamagnetic NPs disappearing because the magnetic moment remains randomly oriented like in a paramagnetic material, favoring the stability of the NPs [40].

The magnetization curves as a function of applied external field parallel to the film surfaces of the FC and of the hybrid films are shown in Fig. 8c.

The control film (FC), which did not contain a magnetic component in its composition, behaved as a diamagnetic material, as expected, and did not contribute to the magnetization of the sample in the presence of the applied external field (Fig. 8c). Because the magnetization is expressed in electromagnetic units per units of mass of the analyzed material, the magnetization curves of the hybrid films were obtained considering the magnetite mass contained in the portion of the film analyzed (determined by TGA in air) to be 1.3%, 4.1%, and 6.8% in the FCFe0.5, FCFe1.4, and FCFe3.0 films, respectively (Table 2). The total mass of the film evaluated was not considered in this work because the cellulose is diamagnetic, as mentioned. Thus, if the total mass was considered in the magnetization calculations, the cellulose present would contribute to the mass increase, but would not contribute to the magnetization, and the saturation magnetization (Ms) values for the films would be underestimated.

It was observed that the (Ms) of the hybrid films increased with the concentration of magnetite NPs dispersed in the films, as expected. High magnetizations for the hybrid films were obtained at room temperature, corresponding to 23.0emug−1 for FCFe0.5, 31.0emug−1 for FCFe1.4, and 37.0emug−1 for FCFe3.0 (Fig. 8c).

The Ms values of the hybrid films were lower than that of the neat NPs (58.0emug−1, at 300K; Fig. 8a); i.e., the magnetic response of the magnetite NPs in the films to the application of an external field was lower than that of the neat NPs, which may be attributed to the fact that the magnetic NPs in the films are surrounded by the diamagnetic cellulose matrix. Hybrid films showed no magnetic hysteresis at 300K and zero remnant magnetization (Mr) and, therefore, behaved as superparamagnetic materials at room temperature.


Sisal cellulose/magnetite-nanoparticle hybrid films containing three different nanoparticle concentrations were successfully prepared by a casting method. The possibility of using underivatized cellulose represents an advantage because it eliminates the cost involved in the derivatization reaction.

The hydrophobic magnetic nanoparticles with an average diameter of 5nm were synthesized by the modified polyol process. After purification methodology, the nanoparticles remain coated with a mixture of oleic acid and oleylamine onto the surface that favors their dispersion in the film medium preparation avoiding significant nanoparticles aggregation.

The presence of the magnetic nanoparticles into the hybrid films influenced their mechanical properties. FCFe1.4 and FCFe3.0 exhibited higher tensile strengths (14.3MPa and 12.1MPa, respectively) than the neat cellulose film (9.9MPa). The moduli of elasticity of FCFe0.5, FCFe1.4, and FCFe3.0 were 1650, 1500MPa, and 780MPa, respectively, lower than that of the cellulose film (1860MPa), indicating that the incorporation of nanoparticles in the cellulosic matrix decreased the films’ stiffness.

Hybrid films exhibited high magnetizations at 300K with values of 23.0emug−1 (FCFe0.5), 31.0emug−1 (FCFe1.4), and 37.0emug−1 (FCFe3.0), as well as the superparamagnetic behavior at room temperature. The results showed that hybrid films with useful properties, including magnetization, can be obtained from cellulose and magnetite.

The properties of the bio-based nanocomposites prepared in the present study point to a wide range of possible applications, such as in sensors, micromechanical devices, advanced catalysis, and biomedical applications, among others. In addition, the methodology can also be easily adjusted to promote the incorporation of different hydrophobic nanoparticles in the cellulose matrix, conferring new properties such as optical and electronic ones.

Conflicts of interest

There are no conflicts to declare.


The authors would like to thank CAPES (Coordination for the Improvement of Higher Level or Education Personnel) for providing a fellowship for D.M.F., as well as CNPq (National Council of Scientific Research) for a research productivity fellowship to E.F. and for financial support (Process 426847/2016-4), as well as FAPESP (State of São Paulo Research Foundation, Brazil) for financial support (Process 2012/00116-6 and 2013/01284-2, 2009/54082-2, 2013/07296-2).

L.A. Ramos, J.M. Assaf, O.A. El Seoud, E. Frollini.
Influence of the supramolecular structure and physicochemical properties of cellulose on its dissolution in a lithium chloride/N,N-dimethylacetamide solvent system.
Biomacromolecules, 6 (2005), pp. 2638-2647
J.J. Kaschuk, E. Frollini.
Effects of average molar weight, crystallinity, and hemicelluloses content on the enzymatic hydrolysis of sisal pulp, filter paper, and microcrystalline cellulose.
Ind Crops Prod, 115 (2018), pp. 280-289
B.V.M. Rodrigues, E. Heikkilä, E. Frollini, P. Fardim.
Multi-technique surface characterization of bio-based films from sisal cellulose and its esters: a FE-SEM, μ-XPS and ToF-SIMS approach.
Cellulose, 21 (2014), pp. 1289-1303
E.V.R. Almeida, D.L. Morgado, L.A. Ramos, E. Frollini.
Sisal cellulose and its acetates: generation of films and reinforcement in a one-pot process.
Cellulose, 20 (2013), pp. 453-465
D.L. Morgado, V.C.A. Martins, A.M.D.G. Plepis, E. Frollini.
Aggregation of chains of cellulose acetates in LiCl/DMAc: evaluation via viscometry.
Polím Ciênc Tecnol, 21 (2011), pp. 143-145
D. Klemm, B. Heublein, H.-P. Fink, A. Bohn.
Cellulose: fascinating biopolymer and sustainable raw material.
Angew Chem Int Ed English, 44 (2005), pp. 3358-3393
N. Phinichka, S. Kaenthing.
Regenerated cellulose from high alpha cellulose pulp of steam-exploded sugarcane bagasse.
J Mater Res Technol, 7 (2018), pp. 55-65
J.C. Courtenay, S.M. Ramalhete, W.J. Skuze, R. Soni, Y.Z. Khimyak, K.J. Edler, et al.
Unravelling cationic cellulose nanofibril hydrogel structure: NMR spectroscopy and small angle neutron scattering analyses.
Soft Matter, 14 (2018), pp. 255-263
G. Kaufman, S. Mukhopadhyay, Y. Rokhlenko, S. Nejati, R. Boltyanskiy, Y. Choo, et al.
Highly stiff yet elastic microcapsules incorporating cellulose nanofibrils.
Soft Matter, 13 (2017), pp. 2733-2737
L.S. Aguilera, R.L.S.B. Marçal, J.B. Campos, M.H.P. Silva, AB-H.S. Figueiredo.
Magnetic filter produced by ZnFe2O4 nanoparticles using freeze casting.
J Mater Res Technol, (2018), pp. 350-355
P.A. Sanchez, T. Gundermann, A. Dobroserdova, S.S. Kantorovich, S. Odenbach.
Importance of matrix inelastic deformations in the initial response of magnetic elastomers.
Soft Matter, 14 (2018), pp. 2170-2183
S. Liu, J. Zhou, L. Zhang, J. Guan, J. Wang.
Synthesis and alignment of iron oxide nanoparticles in a regenerated cellulose film.
Macromol Rap Commun, 27 (2006), pp. 2084-2089
J. Zhou, R. Li, S. Liu, Q. Li, L. Zhang, L. Zhang, et al.
Structure and magnetic properties of regenerated cellulose/Fe3O4 nanocomposite films.
J Appl Polym Sci, 111 (2008), pp. 2477-2484
S. Liu, J. Zhou, L. Zhang.
In situ synthesis of plate-like Fe2O3 nanoparticles in porous cellulose films with obvious magnetic anisotropy.
Cellulose, 18 (2011), pp. 663-673
F. Wang, Y. Yang, Y. Ling, J. Liu, X. Cai, X. Zhou, et al.
Injectable and thermally contractible hydroxypropyl methyl cellulose/Fe3O4 for magnetic hyperthermia ablation of tumors.
V.V. Spiridonov, I.G. Panova, L.A. Makarova, M.I. Afanasov, S.B. Zezin, A.V. Sybachin, et al.
The one-step synthesis of polymer-based magnetic γ-Fe2O3/carboxymethyl cellulose nanocomposites.
Carbohydr Polym, 177 (2017), pp. 269-274
R.J.B. Pinto, M.C. Neves, C.P. Neto, T. Trindade.
Composites of cellulose and metal nanoparticles.
Nanocomposites – new trends and developments, pp. 73-96
K. Liu, J. Nasrallah, L. Chen, L. Huang, Y. Ni.
Preparation of CNC-dispersed Fe3O4 nanoparticles and their application in conductive paper.
Carbohydr Polym, 126 (2015), pp. 175-178
Y. Qin, Z. Qin, Y. Liu, M. Cheng, P. Qian, Q. Wang, et al.
Superparamagnetic iron oxide coated on the surface of cellulose nanospheres for the rapid removal of textile dye under mild condition.
Appl Surf Sci, 357 (2015), pp. 2103-2111
R. Xiong, C. Lu, Y. Wang, Z. Zhou, X. Zhang.
Nanofibrillated cellulose as the support and reductant for the facile synthesis of Fe3O4/Ag nanocomposites with catalytic and antibacterial activity.
J Mater Chem A, 1 (2013), pp. 14910-14918
Y. Li, H. Zhu, H. Gu, H. Dai, Z. Fang, N.J. Weadock, et al.
Strong transparent magnetic nanopaper prepared by immobilization of Fe3O4 nanoparticles in a nanofibrillated cellulose network.
J Mater Chem A, 1 (2013), pp. 15278
H. Zhang, J.Y. Zheng.
Immobilization of magnetic magnetite nanoparticle film on polyamide fabric.
J Appl Polym Sci, 125 (2012), pp. 3770-3777
A. Figuerola, R. Di Corato, L. Manna, T. Pellegrino.
From iron oxide nanoparticles towards advanced iron-based inorganic materials designed for biomedical applications.
Pharmacol Res, 62 (2010), pp. 126-143
I.W. Shim, S. Choi, W.T. Noh, J. Kwon, J.Y. Cho, D.Y. Chae, et al.
Preparation of iron nanoparticles in cellulose acetate polymer and their reaction chemistry in the polymer.
Bull Korean Chem Soc, 22 (2001), pp. 772-774
T.M. Lacerda, M.D. Zambon, E. Frollini.
Effect of acid concentration and pulp properties on hydrolysis reactions of mercerized sisal.
Carbohydr Polym, 93 (2013), pp. 347-356
D.L. Morgado, B.V.M. Rodrigues, E.V.R. Almeida, O.A. El Seoud, E. Frollini.
Bio-based films from linter cellulose and its acetates: formation and properties.
Materials, 6 (2013), pp. 2410-2435
L.A. Ramos, D.L. Morgado, O.A. El Seoud, V.C. Silva, E. Frollini.
Acetylation of cellulose in LiCl-N,N-dimethylacetamide: first report on the correlation between the reaction efficiency and the aggregation number of dissolved cellulose.
Cellulose, 18 (2011), pp. 385-392
C.G.S. Souza, W. Beck, L.C. Varanda.
Multifunctional luminomagnetic FePt@Fe3O4/SiO2/Rhodamine B/SiO2 nanoparticles with high magnetic emanation for biomedical applications.
J Nanopart Res, 15 (2013), pp. 1545
L.C. Varanda, M. Imaizumi, F.J. Santos, M. Jafelicci.
Iron oxide versus Fe55Pt45/Fe3O4: improved magnetic properties of core/shell nanoparticles for biomedical applications.
IEEE Trans Magnet, 44 (2008), pp. 4448-4451
R. Hunter.
Foundations of colloid science.
2nd ed., Oxford University Press, (2001),
A.H. Lu, E.L. Salabas, F. Schüth.
Magnetic nanoparticles: synthesis, protection, functionalization, and application.
Angew Chem Int Ed, 46 (2007), pp. 1222-1244
S. Behera, S. Patel, B.K. Mishra.
Effect of blending of sisal on pulp properties of waste papers in handmade papermaking.
J Scient Ind Res, 74 (2015), pp. 416-422
B.A.P. Ass, G.T. Ciacco, E. Frollini.
Cellulose acetates from linters and sisal: correlation between synthesis conditions in DMAc/LiCl and product properties.
Bioresource Technol, 97 (2006), pp. 1696-1702
Y. Li, Y.W. Mai, L. Ye.
Sisal fibre and its composites: a review of recent developments.
Compos Sci Technol, 60 (2000), pp. 2037-2055
R.M. Cornell, U. Schwertmann.
The iron oxides: structure, properties, reactions occurences and uses.
Wiley-VCH, (2003),
M.P. Morales, S. Veintemillas-Verdaguer, M.I. Montero, C.J. Serna, A. Roig, L. Casas, et al.
Surface and internal spin canting in γ-Fe2O3 nanoparticles.
Chem Mater, 11 (1999), pp. 3058-3064
J.R. Correa, E. Bordallo, D. Canetti, V. León, L.C. Otero-Díaz, C. Negro, et al.
Structure and superparamagnetic behaviour of magnetite nanoparticles in cellulose beads.
Mater Res Bull, 45 (2010), pp. 946-953
J.B. Souza, L.C. Varanda.
Magneto-plasmonic Au-coated Co nanoparticles synthesized via hot-injection method.
Nanotechnology, 29 (2018), pp. 065604
B.D. Cullity, C.D. Graham.
Introduction to magnetic materials.
2nd ed., Wiley-IEEE Press, (2009),
C. Altavilla, E. Ciliberto.
Inorganic nanoparticles: synthesis applications and perspectives.
1st ed., CRC Press, (2011),
Copyright © 2019. Brazilian Metallurgical, Materials and Mining Association
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.