Crystalline starch based nanoparticles

142 downloads 0 Views 2MB Size Report
apposition from the hilum (center of the granule). The thickness of the combined repeated crystalline and amorphous lamellae is 9 nm regardless the botanic ...
Current Opinion in Colloid & Interface Science 19 (2014) 397–408

Contents lists available at ScienceDirect

Current Opinion in Colloid & Interface Science journal homepage: www.elsevier.com/locate/cocis

Crystalline starch based nanoparticles Alain Dufresne ⁎ Univ. Grenoble Alpes, LGP2, F-38000 Grenoble, France CNRS, LGP2, F-38000 Grenoble, France

a r t i c l e

i n f o

Article history: Received 10 March 2014 Received in revised form 24 June 2014 Accepted 27 June 2014 Available online 8 July 2014 Keywords: Starch Nanocrystal Nanocomposite Nanoparticle

a b s t r a c t Starch is an abundant, natural, renewable, and biodegradable polymer produced by many plants as a source of stored energy. Because of the multiscale structure of starch granules consisting of alternating crystalline and amorphous concentric layers, the controlled acid hydrolysis treatment of starch disrupts this organization and releases crystalline platelet-like particles with nanoscale dimensions. This paper intends to provide a comprehensive overview of their preparation, characterization, properties, and applications. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Fossil energy depletion and growing environmental concerns have brought up increasing interest in bio-based eco-efficient and high technology materials. In this context, an increasing demand is made for products made from renewable and sustainable non-petroleum based resources. It has brought two scientific fields together, viz. nanotechnologies, which allow the development of innovative and efficient materials, and biomaterial processing, with the use of renewable raw materials for more environmentally-friendly and sustainable solutions. Starch is a natural, renewable, and biodegradable polymer produced by many plants as a source of stored energy. It is therefore a promising material because of its versatility, low price, availability, and numerous industrial applications. It is the major carbohydrate reserve in plant tubers and seed endosperm, and it is found in plant roots, stalks, crop seeds, and staple crops such as rice, corn, wheat, tapioca and potato [1]. In 2000, the world starch market was estimated to be 48.5 million tons, including native and modified starches, but also the large volume of starch that is converted into syrups for direct use as glucose and isoglucose, and as substrates in the form of very high dextrose syrups (known as starch hydrolysates) for fermentation into organic chemicals, including ethanol [2]. The value of the output is worth € 15 billion per year, explaining the interest of the industrialists and researchers seeking new properties of high value application. The major categories to be considered while mapping the starch processing industry are food and non-food products. Use of starch in food products includes

⁎ Univ. Grenoble Alpes, LGP2, F-38000 Grenoble, France. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.cocis.2014.06.001 1359-0294/© 2014 Elsevier Ltd. All rights reserved.

food processing and beverages. Non-food products of starch include paper, glue, thickening agent, and stiffening agent, among others. Starch consists of amorphous and crystalline domains. The amorphous regions are highly susceptible to hydrolysis and, under controlled conditions, may be dissolved leaving the rigid crystalline regions intact. The acid hydrolysis of native starch granules releases platelet-like nanoscale highly crystalline residues. As the size of a particle is decreasing down to the nanometer scale important changes occur. Both specific surface area and total surface energy increase. Moreover, starch nanoparticles display a highly reactive surface with plenty of hydroxyl groups. When blended with a polymeric matrix, ensuing nanocomposites show unique properties, because of the nanometric size effect, compared to conventional composite even at low filler content. Indeed, nanofillers have strong reinforcing effect and studies have also shown their positive impact in barrier effect. However, for decades studies have been conducted with non-renewable inorganic fillers. Increasing environmental concerns have led to investigating the potential uses of renewable resources for such application.

2. Native starch Starch is the major energy reserve of higher plants. The starch industry extracts and refines starches by wet grinding, sieving and drying. After its extraction from plants, starch occurs as a flour-like tasteless and odorless white powder insoluble in cold water. This powder called native starch consists of microscopic granules with diameters depending on the botanic origin, ranging from 2 (for wheat and rice) to 100 μm (for potato), and with a density around 1.5 g.cm− 3 [1].

398

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

2.1. Starch composition The composition of starch was originally determined by studying the residue of its total acid hydrolysis. It mainly consists of a combination of two glucosidic macromolecules, namely amylose and amylopectin. In most common types of starch, the amylopectin content ranges between 72 and 82%, while the amylose content ranges between 18 and 28%. However, some mutant types of starch have very high amylopectin content (99% for waxy maize), and some very high amylose content (up to 70% and more for amylomaize). Other trace elements are lipids, proteins, minerals, phosphorous, enzymes, amino acids, and nucleic acids. Amylose is essentially a linear polymer consisting of glucose units linked by α-(1 → 4) glycosidic bonds, slightly branched by α-(1 → 6) linkages. Amylopectin is a highly branched polymer constituted of relatively short branches of α-D-(1 → 4) glycopyranose that are interlinked by α-D-(1 → 6) glycosidic linkages. 2.2. Multiscale structure of starch granule The starch granule displays a multiscale structure as shown in Fig. 1. It consists of the (a) starch granule (2–100 μm), into which we find (b) growth rings (120–500 nm) composed of (d) blocklets (20–50 nm) made of (c) amorphous and crystalline lamellae (9 nm) [4] containing (g) amylopectin, and (h) amylose chains (0.1–1 nm). When observed under a microscope and polarized light, starch shows birefringence. The refracted characteristic “Maltese cross” corresponding to the crystalline region is typical of a radial orientation of the macromolecules. The so-called onion-like structure of starch granule with more or less concentric growth rings is composed of alternating hard crystalline and soft less ordered shells growing by apposition from the hilum (center of the granule). The thickness of the combined repeated crystalline and amorphous lamellae is 9 nm regardless the botanic origin [4]. Native starches contain between 15% and 45% of crystalline material [5]. Depending on their X-ray diffraction pattern, starches are

categorized in three crystalline types referred to A, B and C. A-type is characteristic of cereal starches (wheat and maize starch). B-type is typical of tuber and amylose-rich cereal starches. C-type is characteristic of leguminous starches and corresponds to a mixture of A and B crystalline types. V-type is observed during the formation of complexes between amylose and a complexing molecule (iodine, alcohols, cyclohexane, fatty acids, …). The appearance of starch X-ray diffraction pattern depends on the water content of granules during the measurement. The more starch is hydrated, the thinner the diffraction pattern rings up to a given limit. Water is therefore one of the components of the crystalline organization of starch. The crystalline to amorphous transition occurs at 60–70 °C in water and this process is called gelatinization. 3. Preparation of starch nanocrystals Nanoparticles can be prepared from starch following different strategies involving regeneration and precipitation and leading to particles with different properties, crystallinities, and shapes [3]. Moreover, the method for producing nanofibrillated cellulose (NFC) has been transferred for producing starch colloids [6]. Diluted slurry of high amylose corn starch was run through a Microfluidizer for several passes (up to 30). The particle size of the sample obtained from more than 10 passes was below 100 nm with a yield close to 100% and the gel-like suspension remained stable for more than one month. However, the ensuing starch colloids were obtained from breaking down both amorphous and crystalline domains, rendering amorphous nanoparticles after 10 passes. Nevertheless, the classical and most investigated procedure for preparing starch nanoparticles is acid hydrolysis yielding highly crystalline nanoparticles or starch nanocrystals. 3.1. Acid hydrolysis of starch Acid hydrolysis is a chemical treatment largely used in industry to prepare glucose syrups from starch. Nägeli reported the preparation of a low molecular weight acid-resistant fraction of potato starch after the hydrolysis with a 15% (w/v) sulfuric acid (H2SO4) suspension at

Fig. 1. Starch multiscale structure: (a) starch granules from normal maize (30 μm), (b) amorphous and semi-crystalline growth rings (120–500 nm), (c) amorphous and crystalline lamellae (9 nm): magnified details of the semi-crystalline growth ring, (d) blocklets (20–50 nm) constituting unit of the growth rings, (e) amylopectin double helixes forming the crystalline lamellae of the blocklets, (f) nanocrystals: other representation of the crystalline lamellae called starch nanocrystals when separated by acid hydrolysis, (g) amylopectin's molecular structure, and (h) amylose's molecular structure (0.1–1 nm). Reproduced with permissions from [3].

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

room temperature during 30 days [7]. The fraction would be known as Nägeli amylodextrin. After that, Lintner reported a 40 days hydrolysis process with a 7.5% (w/v) hydrochloric acid (HCl) suspension of potato starch at 30–40 °C to produce a high molecular weight starch suspension called “lintnerized starch” [8]. In 1996, by analogy with cellulose nanocrystal preparation, Dufresne et al. reported a method for producing what they called at the time “microcrystalline starch” and which was reported to be agglomerated particles of a few ten nanometers in diameter [9]. The procedure consisted of hydrolyzing starch (5 wt.%) in a 2.2 N HCl suspension for 15 days. The degradation of starch from different origins by HCl has been studied in detail [10]. The kinetics of lintnerization showed two main steps. For lower times (typically t b 8–15 days), the hydrolysis kinetics was fast and corresponded to the hydrolysis of amorphous domains. For higher times (~ t N 8–15 days), the hydrolysis kinetics was slow and corresponded to the hydrolysis of crystalline domains. The critical time corresponding to fast/slow hydrolysis conditions depends on the botanical origin of starch [11,12]. It has been also reported that hydrolysis is faster when using HCl rather than H2SO4 [13]. Temperature favors the hydrolysis reaction but it is restricted to the gelatinization temperature of starch in acidic medium. Gelatinization corresponds to an irreversible swelling and solubilization phenomenon when native granules are heated above 60 °C in excess water. As for temperature, the acid concentration favors the hydrolysis kinetics. However, above a given acid concentration, granule gelatinization occurs, around 2.5–3 N for hydrochloric acid [14].

3.2. Morphological investigation Figs. 2 and 3 show the transmission electron micrographs (TEM) obtained from dilute suspensions of waxy maize starch nanocrystals prepared by HCl and H2SO4 hydrolysis, respectively. They consist of 5–7 nm thick platelet-like particles with a length ranging from 20 to 40 nm and a width in the range 15–30 nm. The detailed investigation on the structure of these platelet-like nanoparticles was reported [15,17]. Marked 60–65° acute angles were observed. TEM observations show

399

Fig. 3. TEM micrographs of negatively stained starch nanocrystals obtained by 3.16 M H2SO4 hydrolysis of waxy maize starch granules during 5 days, at 40 °C, 100 rpm and with a starch concentration of 14.69 wt.% (optimized conditions) (scale bar: 50 nm). Reproduced with permissions from [16].

that during acid hydrolysis, branching points are first hydrolyzed in amorphous domains, starch nanocrystals lying parallel to the incident electron beam (Fig. 2a). When the acid hydrolysis is progressing, the amorphous regions between crystalline lamellae become completely hydrolyzed and nanocrystals are seen lying flat on the carbon film (Fig. 2b–d). Such nanocrystals are generally observed in the form of aggregates having an average size around 4.4 μm, as measured by laser granulometry [18]. The influence of the botanic origin and amylose content on the morphology of starch nanocrystals has been investigated [19]. Nanocrystals were prepared from five different starches, viz. normal maize, high amylose maize, waxy maize, potato and wheat, covering three botanic origins, two crystalline types, and three ranges of amylose content (0, 25, and 70%) for maize starch. Only a moderate influence of the botanic origin of starch was reported on properties such as size,

Fig. 2. TEM micrographs of negatively stained waxy maize starch samples: (a–c) fragments of waxy maize starch granules after 2 weeks of 2.2 N HCl hydrolysis at 36 °C. In (a) a lamellar organization is clearly revealed with the platelets lying parallel to the incident electron beam. In (b) and (c) parallelepipedal platelets are seen lying flat on the carbon film. The arrow in (b) indicates a pyramidal stack of crystals. (d) Nearly individual waxy maize starch nanocrystals obtained after 6 weeks of hydrolysis (scale bars: 50 nm). Reproduced with permissions from [15].

400

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

size distribution, and thickness of the nanoparticles, as well as viscosity of the suspension. Differences were more pronounced when comparing shapes and crystallinity. Nanocrystals produced from A-type starches rendered square-like particles, whereas nanocrystals produced from B-type starches rendered round-like particles. This was explained by the different packing configurations of amylopectin chains for A and B-type starches. A detailed characterization of the molecular content of A-type nanocrystals prepared by acid hydrolysis of waxy maize starch granules was reported [20]. Several populations of dextrins were found corresponding to different structural motifs. One of these had a degree of polymerization (DP) of 14.2, which in the double-helical structure corresponds to a length of 5 nm and to the thickness of the crystalline lamellae within the starch granule. This clearly indicated that the nanocrystals correspond to the crystalline lamellae present in native starch granules. As the nanocrystals were described by parallelepipedal blocks with a length of 20–40 nm and a width of 15–30 nm [15], this would indicate that between 150 and 300 double-helical components are making up these crystalline domains. Further analysis indicated that roughly half of the dextrins in the nanocrystals were branched molecules, which was far more than previous investigations suggested. It was also concluded that they were equally distributed between populations of high and low molecular weights, respectively. Taking into account the length of these branches and the thickness of the platelets, it was likely that the majority of the branching points were found at the reducing-end surface of the nanocrystals, whereas the rest were located at the non-reducing side. Theoretically, the degree of crystallinity (ratio between the mass of crystalline domains and the total mass of nanocrystal) of starch nanocrystals should be 100%, but actually incomplete removal of amorphous regions less ordered surface chains may result in a lower degree of crystallinity. The values commonly reported in the literature are within the range 45–50% [21]. However, higher values such as 79% after 10 days hydrolysis have been reported [22].

extraction process of starch anocrystals. Differential centrifugation has been tested as an isolation process for separating these two kinds of particles but did not seem fitted for fractionation due to hydrogen bonding and different densities within starch granules [25]. Filtration of the hydrolyzed residues using a microfiltration unit equipped with ceramic membranes to assess the cross-flow membrane filtration potential of starch nanocrystal suspensions was conducted [26]. The proposed microfiltration process is shown in Fig. 4. Process parameters were monitored and the properties of feed, permeate and retentate were investigated. Cross-flow filtration was proved to be an efficient continuous operation for separating starch nanocrystals from the bulk suspension and non-fully hydrolyzed particles whatever the ceramic membrane pore size (0.2–0.8 μm). Analysis on permeate showed not only that collected nanoparticles were more crystalline than feed, but also that mostly B-type particles were produced during the first day of hydrolysis. Based on this observation and as an attempt to establish a predictive model for the optimal parameter setting for preparing starch nanocrystals in 1 day, a statistical experimental design and a multi-linear regression method analysis were performed [27]. The possibility of developing an enzymatic pretreatment of starch to reduce the acid hydrolysis duration was also investigated [28]. The objective of this pretreatment was to create pit holes at the surface of native granules without damaging the crystalline structure of starch, and therefore to create pathways expected to facilitate and make more homogeneous the acid penetration during the subsequent acid hydrolysis treatment. A screening of three types of enzymes, namely α-amylase, β-amylase, and glucoamylase, was proposed. The latter was the most efficient for producing microporous starch (Fig. 5) while keeping intact the semicrystalline structure of starch. With a 2 h pretreatment of waxy maize starch granules, the extent of acid hydrolysis currently reached in 24 and 120 h (5 days) was reached in only 6 and 45 h, respectively as shown in Fig. 6. 3.4. Thermal properties

3.3. Optimization of the acid hydrolysis treatment The main limitation for the use of starch nanocrystals was the duration (40 days treatment) and the yield (0.5 wt.%) of the HCl hydrolysis step [23]. The process has been improved by performing periodic stirring of the suspension and the duration of the acid hydrolysis treatment has been reduced to 15 days using a 5 wt.% starch suspension and 2.2 M HCl [9]. Response surface methodology was used to investigate the effect of five selected factors on the selective sulfuric acid hydrolysis of waxy maize starch granules in order to optimize the preparation of aqueous suspensions of starch nanocrystals [16]. These predictors were temperature, acid concentration, starch concentration, hydrolysis duration and stirring speed. The preparation of aqueous suspensions of starch nanocrystals was achieved after 5 days of 3.16 M H2SO4 hydrolysis at 40 °C, 100 rpm and with a starch concentration of 14.69 wt.% with a yield of 15.7 wt.%. This procedure continues to serve for most studies as the standard recipe for the preparation of starch nanocrystals. Confirming the effect of acid concentration, it was shown that when hydrolysis was carried out with 2.2 M HCl, nanoparticles with the minimum size were obtained after 50 days whereas this occurred just after 24 days using 3.7 M HCl [24]. It was also observed that the stronger acidity not only shortened the time required to obtain the minimum size for particles but also resulted in tinier crystals. However, it was recently shown that starch nanocrystals are produced from a very early stage of the acid hydrolysis treatment [25]. It was observed that starch nanocrystals were formed, at least, after 24 h of H2SO4 hydrolysis and that consequently, at any time including final suspension, both microscaled and nanoscaled particles can be found and coexist. The earlier formed nanocrystals might turn to sugar by the end of the batch production process explaining the low yields. This study clearly showed the need for a continuous production and

The thermal properties of five types of starches (waxy maize, normal maize, high amylose maize, potato and wheat) and their corresponding starch nanocrystals were characterized by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) [29]. Native starches showed only one thermal transition, whereas nanocrystals showed two transitions. In excess water, the first peak was attributed to the first stage of crystallites melting (unpacking of the double helixes) and the second transition to the second stage of crystallites melting (unwinding of the helixes). B-type crystallinity starch nanocrystals gained more stability than A-type nanocrystals as they consist of more rigid crystallites. In the dry state, the peaks were attributed to crystallite melting, with a direct transition from packed helixes to unwinded helixes. The presence of two peaks was attributed to the heterogeneity in crystallite quality. Limited influence of the amylose content of starch was observed. This study gave important information for the processing conditions of starch nanocrystals based nanocomposites. It showed that starch nanocrystals can be used in wet processes, such as coating, if temperature remains lower than 80–100 °C, and in dry processes at temperatures below 150–200 °C. Crosslinking of starch nanocrystals with sodium hexametaphosphate (SHMP) was found to hinder the phase transitions of the nanoparticles upon heating [24]. 3.5. Sustainability of starch nanocrystals Starch nanocrystal is a new bio-based nanomaterial. To be sustainable, its preparation and processing should have limited impact on the environment. The “environmentally sensitive” steps have been identified using life cycle analysis (LCA) and different scenarios have been proposed and compared according to different environmental impacts [30]. Also, a comparison to its main competitor, i.e. organically modified

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

401

Microfiltration

Current Process Oignon-like structure

H2SO4

Oignon-like structure

H2SO4

Day 1

Day n

H2SO4

H2SO4

FEED (micro & Nano)

Mono & Oligosaccharides

Day 2 H2SO4

H2SO4

Day 3

RETENTATE

H2SO4

H2SO4

Day 4

PERMEATE = SNC

H2SO4

H2SO4

Day 5 = Final batch

Increased yield SNC + Microparticles Fig. 4. Schematic comparison between the current preparation process involving the progressive production of starch nanocrystals as evidenced in [25], and the proposed microfiltration process. Reproduced with permissions from [26].

Fig. 5. Scanning electron micrographs of waxy maize starch granules after 2 h of enzymatic hydrolysis pretreatment with (a) α-amylase, (b) β-amylase, and (c) glucoamylase under optimal conditions. Reproduced with permissions from [28].

402

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

Fig. 6. Kinetics of sulfuric acid hydrolysis of non-pretreated (filled lozenges) and pretreated (gray squares) waxy maize starches. Reproduced with permissions from [28].

nanoclay (OMMT), was proposed. From a LCA point of view, the production of starch nanocrystal requires less energy than the extraction of OMMT but global warming and acidification indicators were higher than for OMMT. However, starch nanocrystals have the added advantages to be renewable and biodegradable contrary to OMMT which contribute to non-renewable energy and mineral depletion. Recommendations for the scaling-up of starch nanocrystal production process were made and the main concern deals with the extensive use of land and water. It was suggested to: 1. Use non-cultivated low-water consuming naturally occurring starch sources, such as amaranth and some tubers, to limit acidification and eutrophication linked to starch cultivates. 2. Include less-water and energy consuming extraction and washing processes such as microfiltration, as described in [26]. 3. Reach higher yield than current process. 4. Aqueous dispersions of starch nanocrystals After the acid hydrolysis treatment, starch nanocrystals are obtained in the form of a colloidal aqueous suspension. The stability of the suspension depends on the dimensions of the dispersed particles, their size polydispersity and surface charge. It was demonstrated that the use of H2SO4 for the preparation of starch nanocrystals leads to more stable aqueous suspensions than for those prepared using HCl [18]. Indeed, H2SO4 reacts with the nanocrystal surface hydroxyl groups via an esterification process inducing the grafting of anionic sulfate ester groups (\OSO− 3 ). The presence of these randomly distributed negatively charged surface ester groups on the surface of the nanoparticle results in the formation of a negative electrostatic layer covering the nanocrystals and promotes their dispersion in water. The high stability of H2SO4-prepared nanoparticles results therefore from an electrostatic repulsion between individual or aggregated nanoparticles. Acid hydrolysis with HCl does not produce as many negative surface charges on starch nanocrystals resulting in less stable suspensions. A comparison between the effects of the two acids was performed with waxy maize starch as shown in Fig. 7 [18]. It was clearly observed that the use of H2SO4 instead of HCl allows reducing the possibility of agglomeration of starch nanoparticles and limited their flocculation in aqueous medium. Small angle light scattering experiments were performed on 3.4 wt.% H2SO4-prepared starch nanocrystal aqueous suspensions in order to evaluate the kinetic of sedimentation of the nanoparticles [31]. It was shown that there was no sedimentation of

Fig. 7. Comparison of the sedimentation properties of HCl- (left tube) and H2SO4- (right tube) hydrolyzed starch nanocrystals suspended in water after (a) 5 min and (b) 60 min. Reproduced with permissions from [18].

the nanocrystals for a period of at least 12 h. However, the intensity of scattered light slightly increased, revealing that starch nanocrystals tend to aggregate in aqueous medium but not sufficiently to induce a sedimentation phenomenon. The effect of the pH of the dispersion on zeta potential, size distribution, and aggregation behavior of H 2 SO 4 -prepared starch nanocrystals was also investigated [32]. It was shown that as the pH of the dispersion increased the zeta potential decreased and aggregated platelet-like nanocrystals (1.5 μm) changed to mono-dispersed spherical-like nanoparticles (50 nm). It was concluded that stable starch nanocrystal aqueous suspensions could be obtained by adjusting the pH of the dispersion in the range 7.44–9.45. Improved dispersibility in water of starch nanocrystals obtained from H2SO4 hydrolysis of waxy maize starch was observed upon crosslinking with sodium hexametaphosphate (SHMP) [33]. The crosslinking reaction was performed in water at temperatures below the gelatinization temperature of starch and did not disrupt the crystalline structure of the nanocrystals. Stable and uniform starch nanocrystal aqueous suspensions were obtained. Starch nanocrystals are surface active particles. Their emulsifying ability (Pickering effect) was investigated for oil-in-water emulsions consisting of 50 vol.% paraffin liquid and equal concentration of water [34]. Stable emulsions were obtained when the starch nanocrystal content relative to water was above 0.02 wt.%. Both size of the droplets and creaming decreased for increasing starch nanocrystal concentrations. The emulsions were very stable to coalescence over months and the creaming was completely inhibited with 6.0 wt.% starch nanocrystals, possibly because of the formation of an emulsion gel structure. When heating the emulsion above the gelatinization temperature of starch nanocrystals it completely phase-separated. 5. Chemical modification of starch nanocrystals Nanoscale dimensions and ample surface hydroxyl groups enable targeted surface chemical modification of starch nanocrystals to introduce virtually any desired surface functionality. The reactive hydroxyl group content present on the surface of platelet-like starch nanocrystals was estimated to be ca. 14% of the total amount available, i.e. 0.0025 mol.g−1 of starch nanocrystal [35]. Different grafting strategies have been investigated as shown in Table 1. The common surface chemical modifications of starch nanocrystals are schematically depicted in Fig. 8.

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

403

Table 1 Surface chemical modification of starch nanocrystals. Source of starch

Reagent

Objective of the modification

Reference

Corn

Polystyrene Acetic anhydride Microwave-assisted ROP of PCL

Amphiphilic Hydrophobization Blending with PLA Blending with PCL Medical applications Hydrophobization Dispersion in dichloromethane

[36] [37] [38] [39] [40] [41] [35]

Hydrophobization

[42,43] [42] [44]

Pea Potato Waxy corn Waxy maize

Sn(Oct)2-catalyzed ROP of PCL Octanoyl, nonanoyl, decanoyl chloride Alkenyl succinic anhydride Phenyl isocyanate Stearic acid chloride Poly(ethylene glycol) methyl ether Poly(tetrahydrofuran) Poly(ethylene glycol) monobutyl ether Polycaprolactone

Compatibilization with polymer matrices

[44,45]

ROP: ring-opening polymerization; PCL: polycaprolactone; PLA: polylactic acid; Sn(Oct)2: tin(II) octoate.

The first report was conducted with alkenyl succinic anhydride (ASA) and phenyl isocyanate (PI) [35]. Reaction was conducted in toluene/(dimethylamino)pyridine medium to avoid hydrolysis of ASA. Modified nanoparticles were characterized by Fourier transform infrared (FTIR) and X-ray photoelectron (XPS) spectroscopies, contact angle measurements, TEM and X-ray diffraction analysis. The lower polarity of the modified nanocrystals was also demonstrated by a simple experiment (Fig. 9). The pristine and the modified nanoparticles were mixed with two immiscible solvents having both different polarities and densities and it was visually observed with which solvent they are best wetted. Distilled water and dichloromethane were chosen for the

test and it was observed that unmodified starch nanocrystals remained in the water medium whereas modified nanoparticles migrated towards the dichloromethane phase. A crystalline brushlike structure from the starch surface outward was observed from the stearate modification performed by the reaction of starch nanocrystals with stearic acid chloride in methyl ethyl ketone (MEK) [42]. It was evidenced from X-ray diffraction experiments (Fig. 10). The hydrolyzed unmodified waxy maize starch nanocrystals (Fig. 10a) showed the expected scattering pattern for the A allomorph. Poly(ethylene glycol) methyl ether (PEGME) modification did not have pronounced effect on the diffraction pattern (Fig. 10b). Hydrogen

Fig. 8. Common chemical modifications of starch nanocrystals. [PTHF: poly(tetrahydrofuran); PPGBE: poly(propylene glycol) monobutyl ether]. Reproduced with permissions from [21].

404

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

signals were clearly superimposed over the starch pattern. This was confirmed by DSC analysis, where a distinct melting endotherm appeared between 35 and 110 °C. These stearate-modified starch nanocrystals were employed as adsorbents for the removal of a wide range of dissolved aromatic compounds from water [43]. It was shown that the grafted stearate long chains enhanced the adsorption capacity of aromatic organic compounds on the nanometric substrate, which ranged between 150 and 900 μmol.g− 1 and the maximum adsorption amount reached 100 mg. g−1. The adsorption isotherms were described accurately by the Langmuir model and the adsorption kinetics followed a two-step process with first pure adsorption of the aromatic compounds onto the surface of the nanoparticles followed by a diffusion of the compounds into the layer of surface chains grafted onto the nanoparticles. Furthermore, the feasibility of using these nanoparticles in continuous flow mode processes was confirmed using a fixed bed column setup. The fixed bed column could also be regenerated by washing with ethanol and was found not to exhibit any loss in adsorption capacity over multiples adsorption–desorption cycles. Polycaprolactone (PCL)-grafted starch nanocrystals were also obtained using a “grafting onto” [44,45] and “grafting from” [37,38,40] approaches. Polystyrene was also grafted on starch nanocrystals using a “grafting from” strategy [34]. It was systematically verified that the crystalline structure of the nanoparticles was not changed after grafting and that it only occurred on the surface. The surface coating of the nanoparticles allowed dispersion in organic solvents and compatibilization with apolar polymeric matrices. Amphiphilic starch nanocrystals prepared by the graft copolymerization of starch nanocrystals with styrene were well dispersed both in polar and nonpolar solvents [34]. Moreover, microscopic observations of modified starch nanocrystals showed the individualization of nanoparticles. The grafting efficiency of PCL chains onto the surface of starch nanocrystals decreased with the length of the polymeric chains, as expected [44]. 6. Starch nanocrystal reinforced polymer nanocomposites

Fig. 9. Wettability tests: (a) a drop of an aqueous suspension of waxy maize starch nanocrystals in dichloromethane, (b) migration of unmodified starch nanocrystals in distilled water, migration of (c) ASA-modified and (d) PI-modified starch nanocrystals modified with (c) ASA and (d) PI in dichloromethane. Reproduced with permissions from [35].

bonding between PEGME ether groups and unreacted starch hydroxyl groups was supposed to provide ample interactions to bend the surface-grafted chains onto the surface of the nanoparticles. On the contrary significant crystallization of the stearate surface modification was evidenced from its diffraction pattern (Fig. 10c). The stearate diffraction

Fig. 10. X-ray diffraction patterns of (a) unmodified, (b) poly(ethylene glycol) methyl ether-, and (c) stearate-modified waxy maize starch nanocrystals. Reproduced with permissions from [42].

Different polymeric matrices have been associated with starch nanocrystals to prepare nanocomposite materials. Some systems reported in literature are collected in Table 2. The first investigation was performed using a copolymer of styrene and butyl acrylate (poly(S-co-BuA)) matrix in latex form [9]. This aqueous dispersion was mixed with the aqueous suspension of starch nanocrystals and the mixture was freeze-dried and hot-pressed. Following works generally involved casting/evaporation method because of its simplicity and good dispersion level of the nanoparticles in water. Therefore, most polymeric matrices consisted of water-soluble polymers such as carboxymethyl chitosan [22], pullulan [54], polyvinyl alcohol (PVA) [55], gelatinized starch [56–59], and soy protein isolate (SPI) [63] or polymers in the latex form (aqueous polymer dispersion) such as natural rubber (NR) [31,46–50], poly(β-hydroxyoctanoate) (PHO) [51], poly(S-co-BuA) [9,18,53], and waterborne polyurethane (WPU) [38, 60–62]. However, care must be taken regarding the processing temperature to avoid gelatinization of starch nanocrystals. For instance for the processing of thermoplastic starch reinforced with starch nanocrystals, the temperature of gelatinized starch was decreased to 40 °C before adding starch nanocrystals [57,59]. In situ one-pot miniemulsion polymerization reaction of butyl methacrylate (BuMA) in the presence of starch nanocrystals was conducted to impart a high degree of binding between the nanoparticles and the polymer particles to improve the nanoparticle–polymer interface and promote individualization [52]. It was shown that starch nanocrystals were not sufficient to stabilize the monomer droplets, but provided a synergetic stabilization effect when used together with a cationic surfactant, reducing the required surfactant amount by a factor of 4. For starch nanocrystal reinforced polycaprolactone (PCL) [45] and polylactic acid (PLA) [37], surface chemical modification of the

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408 Table 2 Polymer nanocomposites obtained from starch nanocrystals and polymeric matrix. Polymer

Source of starch

Processing technique

Reference

Carboxymethyl chitosan NR

Waxy maize Waxy maize Amylomaize Normal maize Potato Wheat Potato Waxy maize Pea Pea Waxy maize Potato Waxy maize Waxy maize Pea Waxy maize Waxy maize Potato Pea Pea

Casting/evaporation Casting/evaporation

[22] [31,46–49] [49]

PCL PHO PLA Poly(BuMA) Poly(S-co-BuA) Pullulan/sorbitol PVA/glycerol Starch/glycerol Starch/sorbitol WPU SPI/glycerol

Casting/evaporation Casting/evaporation Casting/evaporation Casting/evaporation Freeze-drying/ hot-pressing Casting/evaporation Casting/evaporation Casting/evaporation Casting/evaporation Casting/evaporation Freeze-drying/ hot-pressing

[50] [45] [51] [37] [52] [9,53] [54] [55] [56] [57–59] [60] [61] [38,62,63] [64]

NR: natural rubber; PCL: polycaprolactone; PHO: poly(β-hydroxyoctanoate); PLA: polylactic acid; Poly(BuMA): poly(butyl methacrylate); Poly(S-co-BuA): poly(styreneco-butyl acrylate); PVA: polyvinyl alcohol; WPU: waterborne polyurethane; SPI: soy protein isolate.

nanocrystals was performed to allow their dispersion in dichloromethane, a solvent for both PCL and PLA. 6.1. Mechanical properties The potential use of starch nanocrystals as a mechanically reinforcing phase in a polymeric matrix has been evaluated in both linear (dynamic mechanical analysis — DMA) and nonlinear ranges (tensile tests). In the pioneering work on potato starch nanocrystal reinforced

405

poly(S-co-BuA), a high reinforcing effect of the filler was observed, especially above the glass transition temperature (Tg) of the matrix as shown in Fig. 11 [9,53]. This reinforcing effect was later confirmed by most authors for different polymeric systems. Obviously, the magnitude of the relative reinforcing effect depends on the mechanical properties of the neat polymer. In nonlinear testing conditions, the introduction of starch nanocrystals induced an increase of both tensile modulus and strength whereas the strain at break decreased. However, a decrease of the reinforcing capability of starch nanocrystals has been reported in some studies for higher filler contents because of self-aggregation within the polymeric matrix [55,63]. The reinforcing effect of starch nanocrystals to NR was compared to the one provided by other fillers such as clays, organoclays, carbon black, flyash, and chitin nanocrystals [46]. Starch nanocrystals were not as competitive as organoclays. They displayed a lower tensile modulus but higher ultimate properties (strength and elongation at break) than chitin nanocrystals because of the higher aspect ratio of the latter. Starch nanocrystals presented better mechanical properties than flyash and carbon black. It was shown that the addition of only 10 wt.% starch nanocrystals to NR induced a reinforcing effect similar, in terms of stiffness, to the one observed with 26.6 wt.% carbon black while maintaining a higher elongation at break. The effect of moisture content was also investigated for natural rubber (NR) based materials and the stiffness of the material was found to decrease when increasing the water content [46]. For some systems, an increase of the glass transition temperature (Tg) of the matrix when increasing the starch nanocrystal content was observed attributed to the existence of an interphase of immobilized matrix material in contact with particle surface [54,57,59]. The formation of this interphase resulted from favorable filler/matrix interactions. Interestingly, a considerable slowing down of the recrystallization (retrogradation) of the thermoplastic starch matrix upon storage in humid atmosphere was observed when adding starch nanocrystals [57,59]. For this system, the reinforcing effect was more significant than in NR because of strong interactions between the filler and amylopectin chains from the matrix and possible crystallization at the filler/ matrix interface. Waterborne polyurethane (WPU) was reinforced with starch and cellulose nanocrystals obtained by acid hydrolysis of waxy maize starch granules and cotton linter pulp, respectively [61]. A synergistic effect was observed when adding 1 wt.% starch and 0.4 wt.% cellulose nanocrystals with a significant improvement in tensile strength, Young's modulus and tensile energy at break without significant loss for the elongation at break. The mechanical performance was found to be higher than for individual filler, but it is worth noting that the total filler content was different. In the ternary system, the formation of much jammed network consisting of nanoparticles with different geometrical characteristics was suggested to play an important role in the enhancement of the cross-linked network. Moreover, strong hydrogen bonding interactions between the nanoparticles and between the nanoparticles and the hard segments of WPU matrix were suspected to improve the mechanical properties. 6.2. Effect of chemical modification

Fig. 11. Storage tensile modulus E′ versus temperature at 1 Hz for poly(S-co-BuA) based nanocomposites filled with 0 (●), 5 (Δ), 10 (■), 15 (□), 20 (+), 25 (♦), 30 ( ), 35 (×), 40 (○), 45 (Δ), and 60 wt.% ( )starch nanocrystals. Reproduced with permissions from [9].

Surface chemical modification of starch nanocrystals allows broadening the range of polymeric matrices that can be used through processing of composite materials from an organic solvent instead of aqueous dispersions. It consists in transforming the polar hydroxyl groups from the surface of starch nanocrystals into moieties capable of enhancing interactions with non-polar polymers. The mechanical performance of ASA- and PI-modified starch nanocrystals reinforced NR was lower than for unmodified starch nanocrystals reinforced NR [46]. It was ascribed to hindered interactions between chemically modified particles resulting from their coating with the grafting agents. However, it is worth noting that the nanocomposite

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

6.3. Reinforcing effect The understanding of the reinforcing mechanism of starch nanocrystals in a non-vulcanized NR matrix was studied through the development of a phenomenological modeling approach [48]. Nonlinear dynamic mechanical experiments highlighted the significant reinforcing effect of starch nanocrystals and the occurrence of the Mullins and Payne effects. Two models were used to predict the Payne effect assuming the preponderance of either filler–filler (Kraus model) or matrix–filler (Maier and Göritz model) interactions. With the Maier and Göritz model it was demonstrated that phenomena of adsorption and desorption of NR chains on the surface of the nanofiller governed the nonlinear viscoelastic properties, even if the formation of a percolating network for filler contents higher than 6.7 vol.% (i.e. around 10 wt.%) was evidenced by the Kraus model. 6.4. Swelling properties The interaction of polymeric materials with solvents can be an issue from a technological point of view because the dimensions and physical properties of the material may change with the penetration of solvent molecules into the specimen. The effect of starch nanocrystals on the swelling property of the nanocomposite system strongly depends on the nature of the matrix (polar or apolar) and swelling liquid. Incorporation of starch nanocrystals to poly(S-co-Bu-A) [53] or NR [31,49] was shown to increase the swelling of the material by water probably because of the hydrophilic nature of starch and hydrophobic nature of the matrix (Fig. 12). The coefficient of diffusion of water increased as well but showing two well-defined regions (Fig. 12a). Below a critical starch nanocrystal concentration, the evolution of the diffusion coefficient was relatively low, whereas it was more significant above it. It was assumed to be due to the establishment of a starch nanocrystal network through strong hydrogen linkages between nanoparticle clusters but also to favorable interactions between the NR matrix chains and the filler. This critical concentration was determined around 20 wt.% and 10 wt.%, for poly(S-co-BuA) and NR, respectively. For starch nanocrystal reinforced NR the percolation threshold of the nanoparticles was observed around 15 wt.% from electrical conductivity measurements [50]. Moreover, the diffusion coefficient of water for NR based nanocomposite films was found to be correlated with the composition of starch and to decrease when increasing the initial starch amylose content [49]. It was attributed to the presence of amorphous or less organized, and less bonded amylose chains in starch nanocrystals prepared from higher amylose content starch. This amorphous material was supposed to not participate to the formation of a diffusing network but to participate to the water sorption explaining why similar equilibrium water uptake values were reached. A slight increase of the water uptake of soy protein films was reported for increasing starch nanocrystal contents [63]. For sorbitol-plasticized pullulan (a hydrophilic system) a decrease of the water uptake was observed when adding starch nanocrystals particularly at high filler loading

level [54]. Again, it was ascribed to the formation of a three-dimensional network of nanoparticles that was able to restrict the swelling of the matrix. The decrease in the water uptake of PVA [54] and carboxymethyl chitosan [22] upon starch nanocrystal addition was attributed to the crystallinity of the nanoparticles and strong interfacial interactions. On the contrary, for glycerol-plasticized thermoplastic starch the composites reinforced with starch nanocrystals were found to absorb more water than the unfilled matrix [57]. It was ascribed to a relocalization of glycerol around the nanoparticles leading to more hydroxyl groups in the matrix able to interact with water molecules. Swelling of the NR matrix by toluene decreased when adding starch nanocrystals and the toluene diffusion coefficient decreased strongly for low filler contents and more progressively above 10 wt.% [31] (Fig. 13). For low nanocrystal contents, a correlation between the toluene sorption behavior and calculated specific surface area of nanocrystals obtained from different botanical origin starches was observed [49]. The higher the theoretical specific surface area was, the higher the toluene uptake, and the lower the diffusivity were. For higher starch nanocrystal contents this phenomenon was not observed. It was supposed to be due to aggregation at higher filler contents. 6.5. Barrier properties Given their platelet-like morphology and highly crystalline nature, starch nanocrystals were suspected, as nanoclays do, to create a

a

WU at equilibrium (%)

films were obtained by casting/evaporation from an aqueous or toluene mixture, respectively, for unmodified and modified nanoparticles. On the contrary, both a higher modulus and elongation at break were reported when PCL-decorated starch nanocrystals were used in a PCL matrix, compared to unmodified nanoparticles [45]. For both systems, nanocomposite films were obtained by casting/evaporation from dichloromethane. Obviously, strong self-aggregation of the unmodified nanocrystals in dichloromethane was the source of the poor mechanical properties of the nanocomposites prepared from these particles. Addition of 5 wt.% PCL-grafted starch nanocrystals in a PLA matrix induced simultaneous enhancements of the strength and elongation at break [37]. The grafted rubbery PCL component was supposed to improve the flexibility of PLA chains.

b

D Water (cm²/s)

406

35 30 25 20 15 10 5 0

0

10

20

30

40

10

20

30

40

Starch content (wt%)

50

1,6E-09 1,4E-09 1,2E-09 1,0E-09 8,0E-10 6,0E-10 4,0E-10 2,0E-10 0,0E+00

0

50

Starch content (wt%) Fig. 12. Evolution of (a) the water uptake at equilibrium (WU) and (b) the diffusion coefficient of water (DWater) for starch nanocrystals/NR nanocomposite films immersed in distilled water as a function of the filler content. Reproduced with permissions from [31].

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408

TU at equilibrium (%)

a

D Toluene (cm²/s)

b

and the properties of the ensuing coating color and final coated paper were investigated. Coating colors containing starch nanocrystals showed higher viscosity but were still processable. It was shown that the nanoparticles can resist studied drying processes without melting and that their addition to the coating color decreased WVP and compensated some of the loss of mechanical properties due to the use of the water-based coating.

4000 3000 2000

6.6. Drug carriers

1000 0

0

10

20

30

40

Starch content (wt%)

50

5,0E-07 4,0E-07 3,0E-07

A “green” method of ionic crosslinking through Ca2+, forming semiinterpenetrating polymeric networks (semi-IPN), was used to develop starch nanocrystal reinforced pH-sensitive alginate microsphere controlled release system for drug delivery [65]. Rod-like cellulose and chitin nanocrystals were also investigated in the study. The presence of polysaccharide nanocrystals in alginate-based microspheres showed more consistent swelling patterns, higher encapsulation efficiency, and promising sustained release profiles of the drug. It was ascribed to the improvement of the stability of the crosslinked network structure and enhancement of mechanical strength, mainly for higher aspect ratio cellulose and chitin nano-rods, of nanocomposite microspheres. Moreover, because of the restriction effect of rigid nanocrystals, the free diffusing routes of the drug molecules were increased and endowed the microspheres improved drug loading and sustained release profiles. 7. Conclusions

2,0E-07 1,0E-07 0,0E+00

407

0

10

20

30

40

50

Starch content (wt%) Fig. 13. Evolution of (a) the toluene uptake at equilibrium (TU) and (b) the diffusion coefficient of toluene (DToluene) for starch nanocrystals/NR nanocomposite films immersed in toluene as a function of the filler content. Reproduced with permissions from [31].

tortuous diffusion pathway for penetrant molecules. However, few reports investigated the barrier properties of starch nanocrystal reinforced nanocomposites. A continuous and significant reduction of the permeability to water vapor and oxygen was reported for NR films when adding starch nanocrystals up to 30 wt.% [31]. A substantial 40% decrease of the water vapor permeability (WVP) value was also observed for cassava starch films plasticized with glycerol when adding only 2.5 wt.% of waxy maize starch nanocrystals [56]. However, when using a glycerolplasticized waxy maize starch matrix, a close association between starch nanocrystals and glycerol-rich domains was suspected to explain the unexpected increase of the WVP value when adding starch nanocrystals [58]. A similar decrease of the WVP upon starch nanocrystal addition value was reported for carboxymethyl chitosan films [22]. For sorbitol-plasticized pullulan films no significant differences were observed in WVP values when adding up to 20 wt.% starch nanocrystals [54]. Nevertheless, above this critical value, a significant decrease of WVP was reported. On the contrary, a detrimental effect of starch nanocrystals on the WVP of NR films was reported [49]. However, in this study WVP was measured under tropical conditions (38 °C, 90%RH), and it was suggested that the hydrophilic nature of starch nanocrystals was predominant. Starch nanocrystals were also reported to be promising bionanofillers for improving barrier properties of bio-based coated papers [64]. The nanoparticles have been introduced in a starch based coating

Cost effective and environmentally benign alternatives to petroleum for energy and engineering material demands are required in order to realize a future with a sustainable economy and environment. Abundant and renewable semicrystalline polysaccharides such as starch consisting of crystalline and amorphous domains are possible candidates for organic nanofillers because the amorphous domains can be removed under certain conditions such as controlled acid hydrolysis. Ensuing nanosized crystalline particles, usually known as starch nanocrystals, can be used as reinforcing nanofillers in polymer nanocomposites, thereby creating opportunities to develop highperformance and highly-durable engineering materials from a renewable resource that could be suitable for potential consumer, structural and biomedical applications. References [1] Buléon A, Colonna P, Planchot V, Ball S. Starch granules: structure and biosynthesis. Int J Biol Macromol 1998;23:85–112. [2] http://ec.europa.eu/agriculture/eval/reports/amidon/chap1.pdf . [accessed March 1, 2014]. [3] Le Corre D, Bras J, Dufresne A. Starch nanoparticles: a review. Biomacromolecules 2010;11:1139–53. [4] Gallant DJ, Bouchet B, Baldwin PM. Microscopy of starch: evidence of a new level of granule organization. Carbohydr Polym 1997;32:177–91. [5] Zobel HF. Starch crystal transformation and their industrial importance. Starch/ Sträke 1988;40:1–7. [6] Liu D, Wu Q, Chen H, Chang PR. Transitional properties of micro- to nano-sized starch particles manufactured via a green process. J Colloid Interface Sci 2009;339:117–24. [7] Nägeli CW. Beitage zur näheren Kenntniss des starke grupe. Ann Chem 1874;173:218–27. [8] Lintner CJ. Studien über die diastase. J Prakt Chem 1886;34:378–86. [9] Dufresne A, Cavaillé JY, Helbert W. New nanocomposite materials: microcrystalline starch reinforced thermoplastic. Macromolecules 1996;29:7624–6. [10] Robin JP, Mercier C, Duprat RF, Charbonniere R, Guibot A. Amidons lintnérisés. Etudes chromatographique et enzymatique des résidus insolubles provenant de l'hydrolyse chlorhydrique d'amidons de céréales, en particulier de maïs cireux. Die Stärke 1975;27:36–45. [11] Singh V, Ali SZ. Acid degradation of starch. The effect of acid and starch type. Carbohydr Polym 2000;41:191–5. [12] Jayakody L, Hoover R. The effect of lintnerization on cereal starch granules. Food Res Int 2002;35:665–80. [13] Muhr AH, Blanshard JMV, Bates DR. The effect of lintnerisation on wheat and potato starch granules. Carbohydr Polym 1984;4:399–425. [14] Robin JP. Comportement du grain d'amidon à l'hydrolyse acide ménagée. Etude physico-chimique et enzymatique de la fraction insoluble. Contribution à la

408

[15]

[16]

[17] [18] [19] [20]

[21]

[22] [23] [24] [25]

[26] [27] [28] [29] [30] [31]

[32] [33]

[34] [35] [36] [37] [38]

[39]

[40]

A. Dufresne / Current Opinion in Colloid & Interface Science 19 (2014) 397–408 connaissance de la structure de l'amylopectine. [PhD Thesis] Paris: Université Pierre et Marie Curie; 1976. Putaux JL, Molina-Boisseau S, Momaur T, Dufresne A. Platelet nanocrystals resulting from the disruption of waxy maize starch granules by acid hydrolysis. Biomacromolecules 2003;4:1198–202. Angellier H, Choisnard L, Molina-Boisseau S, Ozil P, Dufresne A. Optimization of the preparation of aqueous suspensions of waxy maize starch nanocrystals using a response surface methodology. Biomacromolecules 2004;5:1545–51. Putaux JL. Morphology and structure of crystalline polysaccharides: some recent studies. Macromol Symp 2005;229:66–71. Angellier H, Putaux JL, Molina-Boisseau S, Dupeyre D, Dufresne A. Starch nanocrystals fillers in an acrylic polymer matrix. Macromol Symp 2005;221:95–104. LeCorre D, Bras J, Dufresne A. Influence of botanic origin and amylose content on the morphology of starch nanocrystals. J Nanoparticle Res 2011;86:1565–72. Angellier-Coussy H, Putaux JL, Molina-Boisseau S, Dufresne A, Bertoft E, Perez S. The molecular structure of waxy maize starch nanocrystals. Carbohydr Res 2009;344:1558–66. Lin N, Huang J, Dufresne A. Preparation, properties and applications of polysaccharide nanocrystals in advanced functional nanomaterials: a review. Nanoscale 2012;4:3274–94. Duan B, Sun P, Wang X, Yang C. Preparation and properties of starch nanocrystals/ carboxymethyl chitosan nanocomposite films. Starch-Starke 2011;63:528–35. Battista OA. Microcrystal polymer science. New York: McGraw-Hill Book Company; 1975. Jivan MJ, Madadlou A, Yarmand M. An attempt to cast light into starch nanocrystals preparation. Food Chem 2013;141:1661–6. LeCorre D, Bras J, Dufresne A. Evidence of micro and nano-scaled particles during starch nanocrystals preparation and their isolation. Biomacromolecules 2011;12:3039–46. LeCorre D, Bras J, Dufresne A. Ceramic membrane filtration for isolating starch nanocrystals. Carbohydr Polym 2011;86:1565–72. Le Corre D, Bras J, Choisnard L, Dufresne A. Optimization of the batch preparation of starch nanocrystals to reach daily time-scale. Starch-Starke 2012;64:489–96. LeCorre D, Vahanian E, Dufresne A, Bras J. Enzymatic pretreatment for preparing starch nanocrystals. Biomacromolecules 2012;13:132–7. LeCorre D, Bras J, Dufresne A. Influence of native starch's properties on starch nanocrystals thermal properties. Carbohydr Polym 2012;87:658–66. LeCorre D, Hohenthal C, Dufresne A, Bras J. Comparative sustainability assessment of starch nanocrystals. J Polym Environ 2013;21:71–80. Angellier H, Molina-Boisseau S, Lebrun L, Dufresne A. Processing and structural properties of waxy maize starch nanocrystals reinforced natural rubber. Macromolecules 2005;38:3783–92. Wei B, Hu X, Li H, Wu C, Xu X, Jin Z, et al. Effect of pHs on dispersivity of maize starch nanocrystals in aqueous medium. Food Hydrocoll 2014;36:369–73. Ren L, Jiang M, Wang L, Zhou J, Tong J. A method for improving dispersion of starch nanocrystals in water through crosslinking modification with sodium hexametaphosphate. Carbohydr Polym 2012;87:1874–6. Li C, Sun P, Yang C. Emulsion stabilized by starch nanocrystals. Starch-Starke 2012;64:497–502. Angellier H, Molina-Boisseau S, Belgacem MN, Dufresne A. Surface chemical modification of waxy maize starch nanocrystals. Langmuir 2005;21:2425–33. Song S, Wang C, Pan Z, Wang X. Preparation and characterization of amphiphilic starch nanocrystals. J Appl Polym Sci 2008;107:418–22. Xu Y, Ding W, Liu J, Li Y, Kennedy JF, Gu Q, et al. Preparation and characterization of organic-soluble acetylated starch nanocrystals. Carbohydr Polym 2010;80:1078–84. Yu J, Ai F, Dufresne A, Gao S, Huang J, Chang PR. Structure and mechanical properties of poly(lactic acid) filled with (starch nanocrystal)-graft-poly(ε-caprolactone). Macromol Mater Eng 2008;293:763–70. Chang PR, Ai F, Chen Y, Dufresne A, Huang J. Effects of starch nanocrystal-graftpolycaprolactone on mechanical properties of waterborne polyurethane-based nanocomposites. J Appl Polym Sci 2009;111:619–27. Namazi H, Dadkhah A. Surface modification of starch nanocrystals through ringopening polymerization of ε-caprolactone and investigation of their microstructures. J Appl Polym Sci 2008;110:2405–12.

[41] Namazi H, Dadkhah A. Convenient method for preparation of hydrophobically modified starch nanocrystals with using fatty acids. Carbohydr Polym 2010;79:731–7. [42] Thielemans W, Belgacem MN, Dufresne A. Starch nanocrystals with large chain surface modifications. Langmuir 2006;22:4804–10. [43] Alila S, Aloulou F, Thielemans W, Boufi S. Sorption potential of modified nanocrystals for the removal of aromatic organic pollutant from aqueous solution. Ind Crop Prod 2011;33:350–7. [44] Labet M, Thielemans W, Dufresne A. Polymer grafting onto starch nanocrystals. Biomacromolecules 2007;8:2916–27. [45] Habibi Y, Dufresne A. Highly filled bionanocomposites from functionalized polysaccharide nanocrystals. Biomacromolecules 2008;9:1975–80. [46] Angellier H, Molina-Boisseau S, Dufresne A. Mechanical properties of waxy maize starch nanocrystals reinforced natural rubber. Macromolecules 2005;38:9161–70. [47] Angellier H, Molina-Boisseau S, Dufresne A. Waxy maize starch nanocrystals as filler in natural rubber. Macromol Symp 2006;233:132–6. [48] Mélé P, Angellier-Coussy H, Molina-Boisseau S, Dufresne A. Reinforcing mechanisms of starch nanocrystals in a nonvulcanized natural rubber matrix. Biomacromolecules 2011;12:1487–93. [49] Le Corre D, Bras J, Dufresne A. Influence of starch nanocrystals' botanic origin on morphological and mechanical properties of natural rubber nanocomposites. Macromol Mater Eng 2012;297:969–78. [50] Bouthegourd E, Rajisha KR, Kalarical N, Saiter JM, Thomas S. Natural rubber latex/ potato starch nanocrystal nanocomposites: correlation morphology/electrical properties. Mater Lett 2011;65:3615–7. [51] Dubief D, Samain E, Dufresne A. Polysaccharide microcrystals reinforced amorphous poly(β-hydroxyoctanoate) nanocomposite materials. Macromolecules 1999;32:5765–71. [52] BelHaaj S, Ben Mabrouk A, Thielemans W, Boufi S. A one-step miniemulsion polymerization route towards the synthesis of nanocrystal reinforced acrylic nanocomposites. Soft Matter 2013;9:1975–84. [53] Dufresne A, Cavaillé JY. Clustering and percolation effect in microcrystalline starch reinforced thermoplastic. J Polym Sci B 1998;36:2211–24. [54] Kristo E, Biliaderis CG. Physical properties of starch nanocrystal-reinforced pullulan films. Carbohydr Polym 2007;68:146–58. [55] Chen Y, Cao X, Chang PR, Huneault MA. Comparative study on the films of poly(vinyl alcohol)/pea starch nanocrystals and poly(vinyl alcohol)/native pea starch. Carbohydr Polym 2008;73:8–17. [56] Garcia NL, Ribba L, Dufresne A, Aranguren MI, Goyanes S. Physico-mechanical properties of biodegradable starch nanocomposites. Macromol Mater Eng 2009;294:169–77. [57] Angellier H, Molina-Boisseau S, Dole P, Dufresne A. Thermoplastic starch–waxy maize starch nanocrystals nanocomposites. Biomacromolecules 2006;7:531–9. [58] Garcia NL, Ribba L, Dufresne A, Aranguren MI, Goyanes S. Effect of glycerol on the morphology of nanocomposites made from thermoplastic starch and starch nanocrystals. Carbohydr Polym 2011;84:203–10. [59] Viguié J, Molina-Boisseau S, Dufresne A. Processing and characterization of waxy maize starch films plasticized by sorbitol and reinforced with starch nanocrystals. Macromol Biosci 2007;7:1206–16. [60] Chen G, Wei M, Chen J, Huang J, Dufresne A, Chang PR. Simultaneous reinforcing and toughening: new nanocomposites of waterborne polyurethane filled with low loading level of starch nanocrystals. Polymer 2008;49:1860–70. [61] Wang Y, Tian H, Zhang L. Role of starch nanocrystals and cellulose whiskers in synergistic reinforcement of waterborne polyurethane. Carbohydr Polym 2010;80:665–71. [62] Zou J, Zhang F, Huang J, Chang PR, Su Z, Yu J. Effects of starch nanocrystals on structure and properties of waterborne polyurethane-based composites. Carbohydr Polym 2011;85:824–31. [63] Zheng H, Ai F, Chang PR, Huang J, Dufresne A. Structure and properties of starch nanocrystal-reinforced soy protein plastics. Polym Compos 2009;30:474–80. [64] LeCorre D, Dufresne A, Rueff M, Khelifi B, Bras J. All starch nanocomposite coating for barrier material. J Appl Polym Sci 2014;131(39826) [7 pp.]. [65] Lin N, Huang J, Chang PR, Feng L, Yu J. Effect of polysaccharide nanocrystals on structure, properties, and drug release kinetics of alginate-based microspheres. Colloid Surf B 2011;85:270–9.