Preparation, characterization and utilization of starch nanoparticles

G Model

ARTICLE IN PRESS

COLSUB-6734; No. of Pages 14

Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Review

Preparation, characterization and utilization of starch nanoparticles Hee-Young Kim, Sung Soo Park, Seung-Taik Lim ∗ Graduate School of Life Sciences and Biotechnology, Korea University, Seoul 136-701, Republic of Korea

a r t i c l e

i n f o

Article history: Received 10 August 2014 Received in revised form 10 November 2014 Accepted 11 November 2014 Available online xxx Keywords: Starch nanoparticles Nanocrystals Nanocomposites Application

a b s t r a c t Starch is one of the most abundant biopolymers in nature and is typically isolated from plants in the form of micro-scale granules. Recent studies reported that nano-scale starch particles could be readily prepared from starch granules, which have unique physical properties. Because starch is environmentally friendly, starch nanoparticles are suggested as one of the promising biomaterials for novel utilization in foods, cosmetics, medicines as well as various composites. An overview of the most up-to-date information regarding the starch nanoparticles including the preparation processes and physicochemical characterization will be presented in this review. Additionally, the prospects and outlooks for the industrial utilization of starch nanoparticles will be discussed. © 2014 Elsevier B.V. All rights reserved.

Contents 1. 2. 3.

4.

5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Granule structure of starch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preparation of starch nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Acid hydrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Enzymatic treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Physical treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1. High-pressure homogenization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.2. Ultrasonication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.3. Reactive extrusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.4. Gamma irradiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Combination of hydrolysis and ultrasonication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Nanoprecipitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6. Complex formation followed by enzymatic hydrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7. Enzymatic hydrolysis and recrystallization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.8. Emulsion-crosslinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Characterization of SNPs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Crystallinity of starch nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Thermal transition properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Molecular composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Rheological properties of SNPs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Application of starch nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Application in composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Emulsion stabilizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Fat replacers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Packaging component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

∗ Corresponding author. Tel.: +82 2 3290 3435; fax: +82 2 921 0557. E-mail address: [email protected] (S.-T. Lim). http://dx.doi.org/10.1016/j.colsurfb.2014.11.011 0927-7765/© 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

2

6.

5.5. Drug carrier and implant material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6. Adsorbents for wastewater treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.7. Thermo-responsive conducting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.8. Binder in papermaking and paper coating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Starch is one of the most abundant biopolymers which is typically synthesized in plant amyloplasts to reserve the energy obtained from photosynthesis. Starch molecules are the polymers of anhydrous glucose units which are typically accumulated in the unique and independent granules. Studies by Gallant et al. [1] and Tang et al. [2] revealed that the starch granules consisted of numerous nano-sized semicrystalline blockets. Through a mild hydrolysis using acids and/or enzymes, the nano-blockets could be isolated from starch [3–8]. Physical treatments may also disintegrate the starch granules and thus release the nano-blocklets [9–11]. These starch nanoparticles have crystalline moiety with the advantages inherently from starch granules, which include renewability and biodegradability. Along with recent interest on nanomaterials, some researchers [12,13] utilized starch nanoparticles (SNPs) as a filler in composites, and found that the incorporation of SNPs improved not only the mechanical properties but also the biodegradability of the composites. In addition, the SNPs have been reported to be applicable in other areas, such as foods, cosmetics, and pharmaceuticals [14–16]. However, the industrial utilization of SNPs is limited until now, and only two applications (BioTRED and Eco-sphere TM) have been reported. One of the applications is called BioTRED (Novamont, Italy), a tire developed in collaboration with an American company (Goodyear, USA). The nanoparticles that had been derived from corn starch were used to replace a part of the carbon black and silica in tire. The incorporation of the SNPs provided an improvement in relining resistance of the tire with environmental advantage. Another example is Eco-sphere TM, a starch-based biolatex that substitutes for the oil-based latex. It has been used as a replacement for petroleum-based coating and binders for paper and paperboard. In this review, various processes for the preparation of SNPs using chemical, enzymatic, and physical treatments were introduced, and granular and molecular structures of SNPs were discussed. The nanoparticle preparation from starch can be classified into “top-down” and “bottom-up” methods according to the preparation scheme. Characterization of the SNPs on the size distribution, crystalline structure, and physical properties was discussed in relation to the starch sources and preparation methods. Finally, potential utilization of SNPs was discussed based on their physical characteristics. In most published researches, the SNPs were focused on their use as reinforcing polymers in composites. Therefore, potential applications of SNPs other than the composites were mainly suggested in this paper.

00 00 00 00 00 00 00

(center), and is extended by apposition (Fig. 1). The shells become thinner toward the granule exterior (due to increasing surface area to be added to at a constant growth rate). The inner architecture of the granule is characterized by growth rings that correspond to concentric semicrystalline shells with thickness that ranges from 100 to 400 nm [18,21]. The shells are separated by amorphous regions. In other words, the granules exhibit an onion-like structure of alternating amorphous and semi-crystalline growth rings. At a high level of molecular order, X-ray diffraction investigations [22–24] indicate a periodicity of approximately 9–10 nm within the granule. The periodicity is interpreted as being due to crystalline and amorphous lamellae that are found within the semi-crystalline shells. These lamellae are believed to represent the crystalline chain clusters and amorphous branching regions of amylopectin molecules according to the models of Robin et al. (1974) and French (1984) [25,26]. It has been estimated by Manners (1989) [27] that 80–90% of the total number of chains in an amylopectin molecule are involved in forming the side chain clusters, whereas the remaining 10–20% of the chains form the inter-cluster and amorphous connections. According to the arrangement of starch chains in crystalline regions, two allomorphs are known, namely, A and B, which correspond respectively to a monoclinic and hexagonal packing of parallel-stranded left-handed double helices [28,29]. The A-type structures are closely packed with water molecules between each double helical structure (e.g., cereal starches), whereas the B-type structures are more open, and water molecules formed by six double helices are located in the central cavity (e.g., tuber and high amylose starches). The C-type structures are believed to contain both types of polymorph: the B-type at the center of the granule and the A-type in the surrounding area (e.g., legume, root, some fruit and stem starches) [30]. Furthermore, recent researches revealed that the semicrystalline lamellae were densely packed with a number of ‘blocklets’ which have a diameter from 20 to 500 nm depending on the botanical source of starch and location in granule [1]. The blocklets have an asymmetric structure with an axial ratio of 2 or 3:1 [2], and its

2. Granule structure of starch Starch is mainly composed of amylose and amylopectin which are different in chain structure. Amylose is defined as linear molecules of anhydrous glucose units that are linked mainly by ␣(1–4)-d-glycoside bonds, with average molecular weights less than a million [17–19]. Amylopectin, however, extensively branched with ␣-(1–6) linkages, has average molar mass up to hundreds of millions [17–20]. The starch molecules are biosynthesized, and are densely packed in granules with dimensions ranging from 1 to 100 ␮m [18,21]. Growth of the granules is initiated at the hilum

Fig. 1. Schematic drawing of hierarchical structure of starch granule [1].

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model

ARTICLE IN PRESS

COLSUB-6734; No. of Pages 14

H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

size is smaller in the semicrystalline shells than in the crystalline shells, as shown in Fig. 1. The starch nanoparticles isolated from starch by disintegration of the granules are often referred to these blocklets. 3. Preparation of starch nanoparticles Acid hydrolysis has been widely used for the preparation of nanoparticles from polysaccharides, because of its simple and ready to control. The crystalline regions in starch granules are more resistant to the acid hydrolysis than the amorphous regions, and thus crystalline moieties can be isolated by the mild acid hydrolysis, typically using hydrochloric or sulfuric acid. Because the mild acid hydrolysis may selectively erode the amorphous regions, the starch nanoparticles in the reaction solution have high crystallinity. However, this selective hydrolysis using acids generally requires long period and thus, the recovery yield is relatively low. Another problem also arises from the presence of a large number of hydroxyl groups, which tend to reform the supramolecular interactions that are characters of starch, yielding aggregates [31]. This tendency hinders the industrial application of SNPs. For these reasons, many researchers have been attempting to find other procedures with physical treatments or a combination of different methods. The preparation of starch nanoparticles may be classified into “top-down” and “bottom-up” processes. In “top-down” process, nanoparticles can be produced from structure and size refinement through a breakdown of larger particles. In “bottom-up” process, nanoparticles can be prepared from a buildup of atoms or molecules in a controlled manner that is regulated by thermodynamic means such as self-assembly [32]. Although a few studies have been reported on the basis of the “bottom-up” approach such as nanoprecipitation [31,32], most studies reported the preparation of SNPs employed “top-down” methods. Therefore, the preparation methods for starch nanoparticles in this review mainly focus on the “top-down” approaches. Starch nanoparticles are often referred to as starch nanocrystals. LeCorre et al. (2010) [33], however, distinguished starch nanocrystals from starch nanoparticles because SNPs may include amorphous matrices. However, it becomes almost impossible to clarify the terms “starch nanocrystals” and “starch nanoparticles”. Both terms have been used to refer to the crystalline parts of starch remaining after hydrolysis or other physical treatments. In this review, the terms that were presented by the original authors of specific studies are used, but elsewhere the general term “starch nanoparticles” is applied to describe the elements that have at least one dimension in the nanoscale. 3.1. Acid hydrolysis The hydrolysis of starch granules is commonly monitored in two ways: the content of soluble sugar in solution (phenol–sulfuric acid) and the recovery of insoluble starch residues. Most of the starches exhibit a two-stage hydrolysis pattern: fast initial hydrolysis followed by slow subsequent hydrolysis [34–36]. However, some authors distinguished three stages of the acid hydrolysis: rapid, slow, and very slow [37]. The initial stage of acid hydrolysis is thought to be the hydrolysis of the amorphous parts within starch granules, whereas the slow stage is attributed to the erosion of the crystalline regions [25,38,39]. Two hypotheses have been proposed to account for the slow hydrolysis of the crystalline domains in starch granules. The first one is that the dense packing in the crystalline regions retards the penetration of H3 O+ [40]. The second one is that the hydrolysis of the glucosidic bonds in crystalline domains requires a change from chair to half chair conformation [38]. If the crystalline structure immobilizes the sugar conformation due to its

3

rigidity, then this transition (chair → half chair) would be sterically impossible. Dufresne et al. (1996) [41] showed that the crystalline residues remaining after a prolonged acid hydrolysis consisted of agglomerated particles of a few tens of nanometers in diameter. When observed at the microscopic scale with a transmission electron microscopy (TEM), internal information regarding these microcrystals could be obtained. In 2003, Putaux et al. [6] observed an edge-on view of the lamellae under the TEM, which was formed by the association of amylopectin side-branches. By subjecting native waxy maize starch to a hydrochloric acid hydrolysis (2.2 N HCl, 2 weeks), it was found that a lamellar arrangement had been disrupted to some extent. In other words, a certain number of ␣(1 → 6) bonds that located in the amorphous regions between the crystalline lamellae were hydrolyzed. Because the branching points located in the interlamellar areas were more readily hydrolyzed, the insoluble residues were fairly well individualized. Indeed, the individual residues could be observed after 6 weeks of hydrolysis under a microscopy. The hydrolysis residues were crystalline nanoplatelets of approximately 6–8 nm thickness, 20–40 nm length, 15–30 nm width [6]. However, many studies reported that the SNPs had spherical or polygonal shapes which often existed as microscale aggregates [7,8,42]. The preparation of these starch nanoplatelets required a long duration of hydrolysis (40 days of treatment) and the recovery yield from the starch granules were relatively low (0.5 wt%). Moreover, the nanoplatelets tended to aggregate to form microscale particles. Angellier et al. [43] employed the hydrolysis using H2 SO4 for the preparation of starch nanoparticles, instead of the hydrolysis using HCl. Compared with the HCl hydrolysis, the H2 SO4 hydrolysis shortened the preparation time, and increased the yield of SNPs. They claimed that the formation of sulfate–ester linkages on the surface of nanoparticles during the hydrolysis should limit the flocculation of nanoparticles and thus produce a nanosuspension with increased stability [37]. This effect was confirmed by comparing the sedimentation rate of the suspensions obtained by using each acid. The stability of nanosuspensions also depends on the dimension of the dispersed particles and polydispersity [9]. As the particles are obtained in smaller size and higher uniformity, the suspension of the particles remains as greater stability. The sulfate–ester bonds on the starch surface may decrease the thermal stability of the nanoparticles, which could negatively affect their use in composites. By adding a small amount of ammonia, the thermal stability of the SNPs suspensions could be enhanced [44]. It is obvious that acid hydrolysis is the procedure most commonly used in preparing nanoparticles from starch. Most of the acid hydrolysis works were carried out with waxy maize starch. However, other sources of starch having different amylose contents and crystallinity were also tested in a recent study [7]. The process optimized by Angellier et al. (2004) [43] has been most frequently used for the production of SNPs [3–6]. Nevertheless, the acid hydrolysis is an energy-intensive process for the production of SNPs and recovery of SNPs is low. The salts produced by neutralizing, and soluble sugars should be carefully removed from the reaction medium, with no or little loss of SNPs. Obtaining uniform SNPs with a high yield is still a task in the process using acid hydrolysis. 3.2. Enzymatic treatment Not much research has been performed on the enzymatic process for the preparation of SNPs, compared to the acid hydrolysis. An enzymatic hydrolysis of waxy rice starch using ␣-amylase for 3 h induced the fragmentation of starch granules with selective dissolution of the amorphous regions [8]. After the enzymatic hydrolysis, the volume-based size distribution of the starch exhibited two major peaks at 0.5 and 3.6 ␮m and a shoulder at 0.1 ␮m. However,

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

4

the average diameter of the particles remained fairly large (approximately 500 nm), which was greater than the value of the blocklets (20–500 nm) reported by Gallant et al. [1] The authors assumed that the enzymatic hydrolysis of the amorphous regions induced the fragmentation of the granules, and a portion of the blocklets could be separated. Combination of both enzymatic and acid hydrolysis has been used for the preparation of SNPs. It was reported that the SNP preparation could be done within a reduced time by using the combined procedure [45]. A pretreatment of starch with glucoamylase for 2 h could effectively decrease the time of acid hydrolysis. This pretreatment created the pathways for the acids to diffuse inside the granules, which made the amorphous regions more readily hydrolyzable. At a same extent of hydrolysis (∼70%), the SNPs from the pretreated starch could be obtained at a higher yield but in a larger size (∼145 nm) compared to those obtained from the nontreated starch (usually between 50 and 100 nm). The extent of acid hydrolysis that was normally required 5 days was obtained after only 45 h (yields of ∼15%). Atomic force microscopy (AFM) and Xray diffraction (XRD) confirmed that the obtained starch particles were nano-sized. 3.3. Physical treatments 3.3.1. High-pressure homogenization Liu et al. [9] developed a simple, versatile and environmentally friendly approach for the isolation of SNPs using a high-pressure homogenization technique. High-pressure homogenization through a microfluidizer is based on the manipulation of a continuous flow of liquid through microfabricated channels. Actuation of the liquid flow is implemented by external pressure sources, external mechanical pumps, integrated mechanical micropumps, or electrokinetic mechanisms [46,47]. In a system that has an electronic-hydraulic intensifier pump, the product stream accelerates to high velocities, creating high shearing within the product stream. When 5% starch slurry (high amylose corn starch, Amylogel TM 03001) was passed through a specially designed microfluidizer 20 times under a pressure of 207 MPa, the particle size of starch granules reduced from 3–6 ␮m to 10–20 nm [48]. This size reduction was attributed to a result of the breakage of the hydrogen bonding inside the large particles by the mechanical shear forces. However, concurrently, it was also observed that partial or complete destruction of the crystalline structure occurred. In addition, because only low concentration of starch slurry could be processed for the homogenization, the recovery yield for each process is low [48]. 3.3.2. Ultrasonication Ultrasound defines the sound waves at a frequency that is above the normal human hearing range (>15–20 kHz). It is generated with either piezoelectric or magnetostrictive transducers that create high-energy vibrations. These vibrations are amplified and transferred to a sonotrode or probe, which is in direct contact with the fluid. The effect of ultrasound on starch depends on many parameters such as sonication power and frequency, temperature and time of the treatment, and properties of starch dispersion (e.g. solid concentration and botanical origin) [49]. Another important factor is the formation of bubbles of gases in the suspension medium, which bombard starch granules before they collapse [50]. This process is called cavitation. Rapid collapsing bubbles could also cause the arising of shear force that may break the polymer chains. Recently, ultrasonication of starch suspension was introduced as a process for the preparation of SNPs [10]. In this study, an aqueous suspension of waxy maize starch (1.5% solids) was treated by ultrasonication in a water bath (8 ◦ C) for 75 min. Based on the SEM observations, the authors claimed that mechanical collision along

with high shear force brought about a progressive erosion of the starch particles starting from the surface, which underwent further fragmentation until a limiting size was reached (between 30 and 100 nm). Furthermore, at the low temperature during the ultrasonication, water molecules could not diffuse inside the amylopectin chains, and no plasticization of the amylopectin phase was likely to take place. It was the first approach to produce nano-sized starch particles by using the physical method of high-intensity ultrasonication. Compared with the common acid hydrolysis, this process offers the advantages of being rapid and easy to implement without the need to undertake repeated washing treatment because no chemical reagent is used. Because no further purification is needed after the SNPs are produced in solution, the recovery yield may be 100%. However, based on wide angle X-ray diffraction analysis, the ultrasonication seriously disrupted the crystalline structure in the starch and appeared to lead to nanoparticles that had a low crystallinity or an amorphous structure. 3.3.3. Reactive extrusion Nanoparticle formation via reactive extrusion has also been investigated [11]. The reactive extrusion has been defined as a concurrent reaction in extrusion processing of polymers [51]. Basically, the premixed starch and plasticizer were loaded in a twin screw extruder, and reversible crosslinkers, such as glyoxal, were added. The SNPs could be produced through cross-linking during reactive extrusion, and is considered to be “regenerated” starch nanoparticles [33]. During the extrusion, starch is subjected to relatively high pressure (up to 103 psi), heat and mechanical shear forces [52]. The hydrogen bonds between starch chains which provide integrity of starch granules can be broken under high shear force and temperature [52]. As a result, significant structural changes, including gelatinization, melting, and fragmentation might occur. Giezen et al. [11] confirmed that starch particles with a size of less than 400 nm could be produced by using reaction extrusion. Due to the limited water content, complete gelatinization did not occur during the extrusion. However, at a high extrusion temperature, starch granules soften and partially melt, and become mobile. The softened and melted starch granules were physically torn apart by the shear force to allow water to transfer into the interior of starch granules. Similar to other physical treatments, a high energy level of extrusion caused mechanical damage of starch crystals. The SNPs prepared by extrusion had a very low viscosity but was stable for more than 6 months [11]. 3.3.4. Gamma irradiation More recently, Lamanna et al. [53] reported that SNPs were prepared by gamma irradiation and were compared with the nanoparticles obtained from acid hydrolysis. Gamma irradiation has been suggested as a rapid and convenient modification technique that fragmenting large molecules by cleaving the glycosidic linkages [54]. Gamma irradiation can generate free radicals which are capable of hydrolyzing chemical bonds, thereby producing smaller fragments of starch called dextrin [55]. According to the study by Lamanna et al. (2013) [53], SNPs that have a size of approximately 20 and 30 nm were obtained by applying a dose of 20 kGy from cassava and waxy maize starches (with an irradiation rate of 14 kGy/h), respectively. Thermal characterization by a simultaneous instrumental analysis revealed that the SNPs obtained from irradiation were more susceptible to thermal degradation than the parent native starch. It suggests that SNPs have a large number of hydroxyl groups on their surface where the thermal degradation starts, similarly to the thermal properties for SNPs obtained by acid hydrolysis [56]. However, similar to the SNPs obtained from other physical treatments, the nanoparticles from gamma irradiation with starch displayed amorphous XRD pattern. The authors

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

explained that the transformation to amorphous structure was not related to the gamma irradiation. Instead, they suggested that it was caused by the heat treatment (85 ◦ C) prior to the gamma irradiation, which was necessary to obtain stable dispersions. 3.4. Combination of hydrolysis and ultrasonication Kim et al. [57] reported the preparation of SNPs from waxy maize starches by a combined process of acid hydrolysis (2 days at 40 ◦ C) and subsequent facile ultrasonication (3 min). By the acid hydrolysis alone, the SNP preparation time took several days. However, the acid hydrolysis for 2 days was sufficient to remove most of the amorphous regions in starch (50–70% of starch). Considering that, the relative crystallinity of waxy maize starch was reported to range from 30% to 50% [36,58]. Subsequently, ultrasonic treatment (60% vibration amplitude, 3 min) was applied to the re-dispersed suspension of the large microparticles of starch hydrolyzates, which had been recovered by a mild centrifugation (500 rpm, 10 min). By this facile process, the microparticles could be completely transformed to nanoparticles. However, the ultrasonication often disrupts the crystalline structure of starch, and recovery yield of the crystalline SNPs becomes low (less than 30%). Kim et al. recently reported a procedure for SNP preparation without disrupting the crystallinity by using acid hydrolysis and ultrasonication [59]. In their study, a low temperature (4 ◦ C) was applied to facilitate the association of the starch chains, minimizing the crystallinity disruption and formation of new crystalline structure in the SNPs by chain association. The starch hydrolyzates, obtained after 6 days of acid hydrolysis, were more resistant to the subsequent ultrasonication than those obtained after 2 or 4 days, regardless of the hydrolysis temperature. Compared to the ordinary process with acid hydrolysis alone [43], the combination of acid hydrolysis and ultrasonication increased the recovery yield of SNPs (15% vs 78%). In addition, the low temperature acid hydrolysis in the combined process was an effective process for the mass production of crystalline SNPs. As previously described [8], small starch particles (∼500 nm) could be prepared by hydrolyzing waxy rice starch using ␣amylase. However, ultrasonication after this enzymatic treatment increased the mean diameter of the starch hydrolyzates. It indicates that the ultrasonication induced the aggregation of the starch hydrolyzates. Under the size distribution profile, the population density of the two main fractions (peaks of 0.5 and 3.6 ␮m) was decreased by the ultrasonication. Additionally, when the degree of the enzymatic hydrolysis was greater, the size increase by the ultrasonication was more significant. This finding indicates that the starch hydrolyzates became more susceptible to the sonication treatment when the hydrolysis level was higher. Ultrasonication may change the X-ray diffraction pattern of starch as well. No change was observed in up to 180 s of treatment, but after treatment for 420 s, the diffractogram revealed that there was a substantial decrease in the crystallinity. Based on the above results, ␣-amylase hydrolyzed the ␣-(1–4) glycosidic linkages in amylopectin and amylose chains. Consequently, some cracks and pores were observed on the surface of the hydrolyzed starch granules, causing the fragmentation of granules. However, these particles were unstable and thus readily swollen by post-treatment like ultrasonication, inducing the formation of starch aggregates. Therefore, to use the present process (combination of both enzymatic hydrolysis and ultrasonication) for producing SNPs, more precise control in the hydrolysis and ultrasonication are needed.

5

precipitation on nanoscale. This method is essentially based on the interfacial deposition of polymers following the displacement of a semipolar solvent that is miscible with water from a lipophilic solution. This approach has many advantages: large amounts of toxic solvents and external energy sources are avoided, and submicron particle sizes with narrow size distributions can be obtained [60]. Tan et al. prepared nanospheres using acetylated waxy maize starch by a nanoprecipitation process [32]. A dispersion of waxy maize starch acetate in acetone was added by distilled water to form nanospheres, and acetone was vaporized from the aqueous suspension. The size of the nanospheres depends on the polymer concentration in the acetone. The mean diameter increases from 249 to 720 nm as the concentration of the starch in the acetone increases from 1 to 20 mg/mL. Similar studies were conducted by the group of Ma et al. (2008) [31]. They prepared SNPs by precipitating starch paste solution with ethanol as the precipitant and then modified the SNPs using citric acid (CASN). When ethanol was delivered dropwise to a starch solution, gelatinized SNPs were gradually precipitated. Therefore, the gelatinization destroyed the A-type crystalline structure of corn starch, and the SNPs exhibited the VH -type crystalline structure. However, most of the crystals of CASN had disappeared. The CASN ranged in size from approximately 50–100 nm and is not gelatinized during the processing of the nano-composites because of the cross-linking by citric acid. 3.6. Complex formation followed by enzymatic hydrolysis In 2009, Kim et al. developed a new process to obtain SNPs based on complex formation between amylomaize starch and nbutanol and subsequent enzymatic hydrolysis [61]. First, a dilute starch solution was placed in the upper compartment of a glass filtration apparatus (Millipore, Billerica, MA, USA) and allowed to gravimetrically pass through a membrane filter (PTFE, 10 ␮m pore size, 47 mm diameter, Millipore, MA, USA) into the bottom compartment, which was filled with n-butanol. The filtration apparatus was kept at 70 ◦ C in a convection oven for 6 days. The starch–butanol complex was isolated as the precipitate and subjected to ␣-amylolysis (Porcine pancreatic alpha-amylase, EC 3.2.1.1, activity 1122 units/mg). Because the amylose-butanol complex that formed in the butanol layer contained a large portion of amorphous matrices, selective removal of these was required to isolate the nanoparticles. The subsequent enzymatic treatment made the SNPs spherical or oval shape with diameters of 10–20 nm. However, there was a significant loss by hydrolysis (85–90%) of the starch, and thus the overall yield of the nanoparticles was very low. 3.7. Enzymatic hydrolysis and recrystallization Sun et al. suggested a green and facile method for obtaining nanoparticles from proso millet starch [62]. Cooked starch was treated by pullulanase and then the solution was stored at 4 ◦ C for starch to recrystallize. The recrystallized starch can be isolated by centrifugation as SNPs with a size range between 20 and 100 nm. The study showed that debranching with pullulanase facilitated the recrystallizing of starch chains. Compared to the conventional acid hydrolysis process, this approach has the advantage of being quite rapid and presenting a higher yield (approximately 55%). This process also offers the advantage of that no chemical reagent is added during the preparation process. 3.8. Emulsion-crosslinking

3.5. Nanoprecipitation Nanoprecipitation process involves the successive addition of a dilute solution of polymer to a solvent which leads to the polymer

More recently, synthesis of SNPs by emulsion-crosslinking technique was reported. It involves the dispersion of aqueous phase containing hydrophilic natural materials such as starch

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14 6

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Fig. 2. Morphology of starch nanoparticles (SNPs) obtained by various preparation methods. (a) TEM micrograph of waxy maize SNPs obtained by acid hydrolysis [43]. (b) SEM micrograph of waxy rice SNPs obtained by enzymatic hydrolysis [8]. (c) SEM micrograph of high amylose corn SNPs obtained by high-pressure homogenization [9]. (d) SEM micrograph of waxy maize SNPs obtained by ultrasonication [10]. (e) TEM micrographs of waxy maize SNPs obtained by gamma radiation [53]. (f) AFM micrograph of waxy maize SNPs obtained by enzymatic hydrolysis and acid hydrolysis [45]. (g) SEM micrograph of waxy maize SNPs obtained by combination of acid hydrolysis and ultrasonication [57]. (h) SEM micrograph of waxy rice SNPs obtained by enzymatic hydrolysis followed by ultrasonication [8]. (i and j) SEM micrograph of waxy maize starch acetate (SA) nanospheres and corn SNPs obtained by nanoprecipitation, respectively [31,32]. (k) TEM micrograph of amylomaize starch–butanol complexes isolated by enzymatic hydrolysis [61]. (l) SEM micrograph of proso millet SNPs obtained by combination of enzymatic hydrolysis and recrystallization [62]. (m) TEM micrograph of soluble SNPs obtained by high-pressure homogenization of water-in-oil (W/O) microemulsion followed by cros-linking through sodium trimetaphosphate (STMP) [65]. (n) SEM micrograph of normal maize SNPS obtained by combination of ionic-in-oil (IL/O) microemulsion preparation and its cross-linking through epichlorohydrin [66]. (o) SEM micrograph of SNPs obtained by combination of water-in-ionic liquid (W/IL) microemulsion preparation using a acid treated starch followed by cross-linking through epichlorohydrin [67].

and cross-linkers in oil phase with the presence of emulsifiers. The emulsion can generate small particles through cross-linking reaction. However, these particles obtained from this emulsioncrosslinking approach were relatively big in micro-scale. Fang et al. (2008) [63], and Franssen and Hennink (1998) [64] observed starch microspheres with the average diameter of 19 ␮m and ranging from 2.5 to 25 ␮m through this emulsion-crosslinking method. To reduce the size of these particles, emulsion should contain nano-scale droplets, which is generally referred to miniemulsion, submicron emulsion or nano emulsion. It is because that the emulsion droplets maintain their shape and size within the dispersed phase [65]. Zhou et al. (2014) [66] attempted to reduce the particle size using ionic liquid-in-oil (IL/O) microemulsion system instead of traditional water-in-oil (W/O) emulsion approach. By substituting the water phase using a 1-octyl-3-methylimidazolium acetate, starch nanoparticles could be obtained, which had an average diameter of 96.9 nm. Another study of Zhou et al. (2014) [67] suggested the possiblity of production of SNPs through in a water-in-ionic liquid (W/IL) microemulsion system. It was an efficient and environmentally friendly approach because no toxic organic reagents such as cyclohexane was used as oil phase [66]. Miniemulsion may be produced by specially designed devices which provide sufficient mechanical energy (e.g., rotor–stator, sonicators, high-pressure homogenizers, and membrane systems) [65,68–71]. Using the mechanical devices increases emulsification efficiency and reduces the cost with less amount of surfactant. Shi et al. (2011) [65] employed a high-pressure homogenizer at pressures ranging from 10 MPa to 60 MPa producing a miniemulsion in which nanoparticles displayed a good sphericity in shape and a comparatively uniform size distribution. The size of the particles varied depending on the process parameters including surfactant content, water/oil ratio, starch concentration, and homogenization pressure and cycles. With applying more energy,

the size of the emulsion might be increased due to the recoalescence of droplets [72]. 4. Characterization of SNPs 4.1. Morphology Figs. 2 and 3 illustrate the micrographs of SNPs which were different in preparation method and in botanic origins, respectively. Morphological characteristics of SNPs depend on botanical source and preparation method. LeCorre et al. prepared SNPs from five starches of different origins (e.g., normal maize, high amylose maize, waxy maize, potato, and wheat) and compared their morphologies (Fig. 3a) [42]. They concluded that the morphology of SNPs appeared to be in relation to the crystalline structure of original starches. Overall, the nanocrystals produced from A-type starches (e.g., waxy maize, normal maize, wheat starch) rendered square-like particles, whereas those from B-type starches (e.g., high amylose maize, potato) produced round-shaped particles. Assuming that the nanoparticles represent the blocklets in starch granule, this difference indicates that the blocklets between the A- and Btype starches have different morphologies. The differences in the arrangement of amylopectin double helices in crystal lattices might induce the formation of different blocklets. More recently, Kim et al. prepared SNPs from various starch sources, and characterized their morphology [7]. According to their observation by TEM (Fig. 3b), all the SNPs had round or oval shapes, regardless of the starch origin. Round-shaped nanoparticles from high-amylose maize starch were in agreement with the result of LeCorre et al. [42]. The nanoparticles from the B-type starch (e.g., 69.7 nm for high amylose maize) and C-type starch (e.g., 53.7 nm for mungbean) were larger than those from the A-type starches (e.g., 41.4 nm for waxy maize, and 41.0 nm for normal

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

7

Fig. 3. Morphology of starch nanoparticles (SNPs) obtained by different botanic origins. (a) SEM micrographs after acid hydrolysis using diluted sulfuric acid (3.16 M, 5 days) [42]. (b) TEM micrographs after acid hydrolysis using diluted sulfuric acid (3.16 M, 5 days) [7].

maize, respectively). The larger nanoparticles from B- and C-type starches might be somewhat related to their greater resistance to acid hydrolysis compared to A-type starches [42]. It was also notable that the size of SNPs appeared to be influenced by amylose content: the more amylose existed, the larger particles were produced. The starches of similar amylose contents such as maize, potato, and wheat starches produced the SNPs in similar sizes. The morphological data reported in these studies [7,42] were somewhat different from the study by Putaux et al. [6], who first reported the SNPs prepared by acid hydrolysis. They reported that the SNPs were obtained in plate-like structure. The morphology and size of the SNPs might not be exactly same as that in starch granules because those could be changed according to the physical and chemical parameters used for the preparation. Although there are some differences in the morphology of SNPs, it is obvious that the SNPs obtained from the hydrolysis of starch granules tend to self-aggregate forming microscale agglomerates. The aggregation behavior of SNPs can be explained by the presence of a large number of hydroxyl groups on the surface of SNPs, which readily participate in the formation of hydrogen bonding or van der Waals attraction between SNPs [53,73]. This aggregation behavior of SNPs greatly limits their industrial applications. In the fabrication of nanocomposites with SNPs, blending aqueous SNP suspension with a matrix solution is required [74]. The homogeneity with nano-sized SNPs is critical to achieve desirable mechanical performance of the nanocomposites. To avoid the aggregations, facile physical treatment such as sonication or homogenization were performed before mixing with a polymeric matrix. More recently, Wei et al. found that aggregated parallelepiped nanoplatelets (1.5 ␮m) changed to monodispersed spherical-like nanoparticles (50 nm) while increasing the dispersion pH from 2.07 to 11.96 [73]. This finding was attributed to the retardation of the aggregation tendency induced by a result from the increased repulsion forces among the SNPs which had become electronegative by alkali. 4.2. Crystallinity of starch nanoparticles Most of the studies on SNPs focused on the changes in the crystalline structure of starch during the production process of SNPs.

The relative crystallinity (RC) of the SNPs to the original crystallinity of starch was assumed to be positively related to the recovery yield of the SNPs. Not many studies calculated the RC of SNPs [42,69,75]. LeCorre et al. investigated the influence of the botanic origin of starch on the crystallinity following the method of Nara and Komiya [76] using an X-ray diffraction analysis [42]. Compared to native starches, the RC of the SNPs increased because the amorphous regions in starch were selectively removed by the acid hydrolysis. Among the maize starches containing different amounts of amylose, the RC of the SNPs from maize starches increased as the amylose content decreased (35% and 48% for high amylose maize and waxy maize starch, respectively). In the comparison between the starches having similar contents of amylose (21–28%) but different botanic origins, no significant differences were observed in the degree of crystallinity of their SNPs (e.g., 42% and 43% for normal maize and potato starch, respectively). This study demonstrated that the most important parameter in determining the degree of crystallinity of SNPs was the amylose content in starch [42]. Duan et al. also reported similar results [75]. The RC of waxy maize SNPs were 48% and 63% after 4 and 6 days of acid hydrolysis, respectively. After 10 days of hydrolysis, the RC of the SNPs could be increased up to 79%, but its yield became very low (10%). Although the SNPs could be obtained by the classic acid hydrolysis with the inherent crystallinity, some drawbacks remain to be solved: an extended period (longer than 5 days) and low yield (less than 20%) for practical utilization. Therefore, recent researches have been focusing on the production of SNPs in alternative processes such as physical treatments (e.g., high pressure homogenization, extrusion, and ultrasonication). Compared to the acid hydrolysis, these physical treatments generate SNPs in shorter periods with higher yields. However, it is difficult to control the physical parameters to minimize the destruction of starch crystals. An attempt was made to isolate crystalline SNPs, which were resistant to the subsequent ultrasonic treatment, by using different temperature profiles: isothermally at 4 ◦ C or 40 ◦ C and under cycled temperatures of 4/40 ◦ C [69]. The starch hydrolyzates obtained from the low temperature hydrolysis (4 ◦ C) exhibited exceptionally high crystallinity (33%) even after the ultrasonic treatment, although its RC was slightly lower than that of native waxy

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

8

Table 1 Examples of preparation methods of starch nanoparticles and their characteristics (size, crystallinity and yield). Methods Top-down approach Chemical treatment

Enzymatic treatment Physical treatment

Sourcesa

Procedures

Dimensions (nm)b

Crystallinity

Yields (%)

References

WM WM NM HM Pea Potato WR HM

2.2 N HCl hydrolysis at 36 ◦ C for 40 days 3.16 M H2 SO4 hydrolysis at 40 ◦ C for 5 days 2.87 M H2 SO4 hydrolysis at 45 ◦ C for 7 days 3.16 M H2 SO4 hydrolysis at 40 ◦ C for 5 days 3.16 M H2 SO4 hydrolysis at 40 ◦ C for 5 days 3.16 M H2 SO4 hydrolysis at 40 ◦ C for 5 days ␣-Amylase hydrolysis at 37 ◦ C for 3 h Passing 20 times through microfluidizer under a pressure of 207 MPa Ultrasonication (80% power, 8 ◦ C, 75 min) Starch premix (starch, water, glycerol) and subsequently extrusion in a reactive extruder (55–110 ◦ C) Gamma radiation at a dose of 20 kGy (14 kG/h) Glucoamylase-enzymatic pretreatment (2 h), followed by 3.16 M H2 SO4 hydrolysis at 40 ◦ C for 45 h 3.16 M H2 SO4 hydrolysis at 40 ◦ C for 2 days, followed by ultrasonication (60% amplitude, 3 min) 3.16 M H2 SO4 hydrolysis at 4 ◦ C for 6 days, followed by ultrasonication (60% amplitude, 3 min) ␣-Amylase hydrolysis at 37 ◦ C for 3 h, followed by ultrasonication (300 W, 420 sec)

L: 20–40 W: 15–30 L: 20–40 W: 15–30 L: 50 W: 69.7 L: 60–150 W: 15–30 30–80 W: 500 10–20

No change No change No change Decrease No change No change No change Decrease

0.5 15 – 38.6 – – – ∼100

[6] [41] [77] [7] [78] [79] [8] [9]

W: 30–100 ∼160

Decrease Decrease

∼100 –

[10] [48]

W: 20–30

Decrease

–

[53]

–

No change

15

[45]

W: 70–120

Decrease

27

[57]

W: 50–90

No change

78

[59]

More than 500 (increase)

Decrease

–

[8]

Preparation by precipitating starch paste solution with ethanol as the precipitant Starch–butanol complex formation, followed by enzymatic hydrolysis Debranching by pullulanase, followed by retrogradation at 4 ◦ C up to 24 h High-pressure homogenization of W/O microemulsion, followed by cross-linking using a STMP cross-linking of IL/O microemulsion using a epichlorohydrin Cross-linking of W/IL microemulsion using a epichlorohydrin

W: 50–300

Transformation to VH -type Transformation to V6-I type Transformation to B-type –

–

[31]

10–15

[61]

∼54.66

[62]

–

[65]

–

[66]

WM NM

Cassava, WM Combined treatment

WM

WM

WM

WR

Bottom-up approach Nanoprecipitation

NM

Combined treatment

HM PM SS

NM ATS a b

W: 10–20 W: 20–100 W: 50–250

W: ∼100 W: 100–300

[67]

WM, waxy maize; NM, normal maize; HM, high amylose maize; WR, waxy rice; PM, proso millet; SS, soluble starch; ATS, acid treated starch. L: length, W: width.

maize starch (36.37% vs 33%). In contrast, the starch hydrolyzates, obtained by the same acid hydrolysis at a higher temperature, had less crystallinity either under the isothermal (27.68%) or cycled conditions (26.42%), indicating the disruptions of starch crystals depended on hydrolysis temperature. Keeping a lower temperature during hydrolysis helped the starch granules remain more rigid and intact during the hydrolysis although the hydrolysis occurred slower. A mild hydrolysis prior to the physical fragmentation is needed for protecting the starch crystals against the physical treatment and assisting the fragmentation. In contrast to cellulose nanoparticles which are recovered in almost pure crystal form, SNPs may be produced in crystalline shape. The crystallinity of SNPs is influenced from the inherent properties of starch such as botanical source, amylose content of starch, crystalline structure, as well as the processing parameter for SNP preparation such as hydrolysis temperature and ultrasonic power and level. The crystallinity of SNPs is a merit because the SNPs remain intact and are readily isolated from the suspension. Some characteristics (size, crystallinity and yield) of the SNPs are listed in Table 1. 4.3. Thermal transition properties The thermal transition behavior of SNPs has been characterized by differential scanning calorimetry (DSC). Compared with

native starch, the endothermic melting transition peak of the SNPs, prepared by acid hydrolysis, was observed in broadened ranges. This trend was observed regardless of the processing methods for SNPs [7,53,59,62,80]. However, melting temperatures and enthalpy depended on the parameters for the SNP preparation and the botanical source of starch. The broad melting range of SNPs could be attributed to the heterogeneity of the SNPs in which amorphous and crystalline phases were in mixture [53]. It was reported that some A-type starches such as waxy and normal maize starches displayed broad melting endotherms in wider temperature ranges than the corresponding native starches [7,80]. In contrast, no endothermic peak was found for the SNPs prepared from B-type potato and high amylose maize starches and C-type mungbean starch [7]. It was hypothesized that the crystals in the SNPs from the B-and C-type starches might have melting temperature below the room temperature, and thus were readily disrupted in water added for DSC analysis [7]. These cold-water-soluble SNPs could have similar properties to pregelatinized starch [81]. As shown in the X-ray diffraction patterns, the B-type crystals were more susceptible to acid hydrolysis than the A-type crystals [7]. In native starch, the Btype crystalline packing is less dense than the A-type crystalline packing and hence, more mobile and more prone to disruption [28,29,82].

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Although similar processes were used for the SNP preparation in both studies, LeCorre et al. [80], however, reported thermal transition results different from those by Kim et al. [7]. The melting temperatures for the SNPs reported by Kim et al. were higher than those reported by LeCorre et al. The minor differences in preparation and analysis processes might cause the differences in the melting behavior of SNPs. It was observed that the SNPs from B-type starches (e.g., high amylose maize, potato) showed slight higher melting temperatures, compared to those of A-type SNPs [7]. The higher thermal stability of the B-type SNPs could be attributed to the longer chains in the amylopectin helices of the B-type starches which could form more stable crystals [83]. By changing the preparation conditions such as hydrolysis temperature and time, starch nanoparticles could be obtained with different melting characteristics [59]. After 6 days of hydrolysis followed by ultrasonic treatment, the temperature range for melting was 11.4 ◦ C for the hydrolysis at 4 ◦ C, 35.7 ◦ C for the 4/40 ◦ C hydrolyzates, and 34.9 ◦ C for the 40 ◦ C hydrolyzates. The melting temperatures for the SNPs were also slightly increased by either removal of the amorphous regions or minor rearrangement of the starch chains that may occur during the ultrasonic treatment. The physical treatments applying high levels of energy could disrupt the crystallinity of SNPs. The absence of the endothermic peak was observed when SNPs were obtained by applying a gamma radiation (20 kGy) [53]. The authors suggested that this result could be from the pre-heating (85 ◦ C, 30 s) that was employed to prepare the stable suspensions probably caused starch gelatinization. The thermal transition characteristics of the SNPs are important in the applications of the SNPs. As an example, the endothermic transition of SNPs should be carefully considered when the SNPs are used in the production of composites with other polymers where thermal melting of the polymers is typically accompanied [53]. When the SNPs should remain as an reinforcing filler, thermal resistance in the polymeric blends against the composite preparation conditions is required. The crystals in SNPs are originated from those in native starches, which are generally stable when moisture content is limited. By changing the physical parameters and starch sources for the SNP preparation, thermal transition characteristics and resistance of the SNPs might be somewhat controlled. Therefore, when SNPs are used in composites, the preparation process of the SNPs should be carefully designed to render the proper thermal properties. Based on the reported data, in a case where thermal resistance is highly required for the utilization of SNPs such as composites, the acid hydrolysis at a high temperature such as 40 ◦ C and A-type starches might be favorable.

9

Fig. 4. (a) A hypothetical model of branched building blocks (encircled in light gray) of clusters, as found in the amorphous and crystalline lamellae of starch granule. Branches are symbolized by arrows and chains of (1→4)-linked glucosyl residues by solid lines. The borders between the lamellae are indicated by irregular light gray lines. (b) The molecular composition of nanocrystals formed after acid treatment of the granule. Examples of possible structures found in populations A and B are shown. Ø: reducing-ends and the filled symbol symbolizes reducing ends involved in a (1→6) linkage being resistant to debranching enzymes, 䊉: single residues in a branch also resistant to debranching. (c) Dextrins of different types (1–8) formed from the structures in (b) by ␤-amylolysis [84].

4.4. Molecular composition The molecular features of SNPs have been rarely studied [7,84]. The molecular composition of the nanoparticles prepared from waxy maize starch was reported by Angellier et al. (Fig. 4) [84]. Waxy maize SNPs, which consisted of double helices containing ␣(1→6) linkages between the parallel strands, were progressively degraded into low molecular weight oligosaccharides when they were subjected to debranching enzymes (isoamylase and pullulanase) followed by ␤-amylase. Eight different dextrins were found in the SNPs (Fig. 4c) prepared by the enzymatic treatment, and classified into two major groups: A and B populations. The population A consisted of multi-branched dextrins with an average degree of polymerization (DP) of 31.7. Population B, which was approximately 1/3 of the total dextrins, however, contained at least one branch and its average DP was 14.2. However, the mole percentages of branched chains of population A and B were similar (27 vs. 24 mol%). It indicates that the SNPs could be considered to possess principally homogeneous and regular molecular structures.

The chain length and chain distribution of starch depend on the botanic origin, and thus chain conformation after debranching the SNPs was influenced by the botanical source of starch [7]. The amylopectin chains isolated by debranching were designated as long B (B ≥ 2), short B (B1), and very short A-chains [85]. The chain length distribution of the waxy maize SNPs showed that the proportion of short chains (B1 and A-chains) were higher than that of native waxy maize starch. This is attributed that the long amylopectin chains such as B ≥2 were more susceptible to the acid hydrolysis. However, other starches that contained amylose (normal maize, high AM maize, potato and mungbean starches) exhibited increased proportions of both B ≥2 and B1 plus A-chains because the amylose was readily disintegrated by the acid hydrolysis. The chain length distributions of the SNPs revealed that the amylose and long amylopectin chains were highly susceptible to the acid hydrolysis, possibly because those chains are located mostly in the amorphous regions in starch granule [86].

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14 10

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Fig. 5. (a) Pasting curves of corn starch, corn SNPs, and citric acid (CA) crosslinked SNPs. The top curve (dash dot line) is the temperature profile [48]. (b) Viscosity of corn starch particle suspension and cooked starch at room temperature. MCR 300 rheometer was used for viscosity measurement [31].

4.5. Rheological properties of SNPs Ma et al. showed that the pasting behavior of SNPs was different from that of native starch (Fig. 5a) [31]. When heated in excess water, native corn starch granules swelled, and the ordered granule structure was disrupted at the gelatinization temperature range, which resulted in an increase in viscosity. The SNPs, however, easily dissolved in water, even at a lower temperature than the gelatinization temperature of native starch. The pasting viscosity of SNPs was much lower than that of native starch, because the SNPs had been produced by acid hydrolysis. However, when the amount of the SNPs increased, the pasting viscosity was increased and the pasting pattern appeared somewhat different from that of native starch (Fig. 5a). The viscosity of the SNPs continuously increased with the increase in temperature (e.g., from 50 ◦ C to 95 ◦ C) and then reached a plateau at 95 ◦ C. There was no breakdown during the stage of hot shearing. When the temperature decreased from

95 to 50 ◦ C, the viscosity of the SNPs paste increased again. Overall viscograms revealed that the SNPs does not show the rapid viscosity increase induced by the granule swelling as shown by native starch, possibly because of the reduced size. However, the heating induced the melting of crystals in SNPs forming a paste with viscosity increase. As nanofillers, the SNPs should not be destroyed during the processing of composite preparation. In most composites, water is rarely used and polymers and SNPs were typically blended in organic solvents. Thus, the SNPs remain as crystalline solids during the processing such as hot pressing or casting-evaporation. At a given shear rate, the viscosity of starch particle suspensions significantly increased with the decrease in particle size, especially when the size was reduced to 600 nm or lower (Fig. 5b) [48]. This relationship could be attributed to the fact that, at a given mass concentration, the smaller the particle resulted in the more particles in the solution. Thus the interaction between the particles was increased so that the viscosity of the suspension

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model

ARTICLE IN PRESS

COLSUB-6734; No. of Pages 14

H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

11

Table 2 Use of starch nanoparticles in nanocomposites and their mechanical properties. Fillers

Matrixes

E (MPa)

 b (MPa)

εb (%)

References

Waxy maize SNP

Natural Rubber

Potato SNP Pea SNP

↑ (∼20%) ↓ ↑ ↑ ↑ (1%) ↑ (∼25%) ↑ (∼5%) ↑ (∼5%) ↓ ↑ (∼10%) ↓ ↑ (2%) ↓ (40%)

↓ ND ↓ ↓ – (1%) ↓ (∼25%) ↑ (∼2%) ↑ (∼5%) ↓ ↑ (∼10%) ↓ ↓ ND

[105]

Thermoplastic starch Sorbitol-plasticized pullulan Waterborne polyurethane Thermoplastic starch Waterborne polyurethane Polylactic acid

↑ ND ↑ ↑ ↑ (1%) ↑ (∼25%) ↑ (∼5%) ↓ ND ↓ (∼10%) ↑ ↑ (2%) ↓ (40%)

Polyvinyl alcohol Soy protein isolate

[106] [12] [107] [108] [79] [109] [110] [78]

E: Young’s modulus (Elastic modulus),  b : tensile strength, εb : Elongation at break, (): % filler content.

increased. Compared to the traditionally cooked starch paste at a low concentration (3%), the SNP suspension at a much higher concentration (9%) showed much lower viscosity at the same shear rate of 1000s−1 . It indicates that the use of SNPs allows the suspensions of much higher solid contents with lower viscosity, which could be desirable in paper making and coating. The high solid content may increase the bonding strength when the suspension is used as an adhesive [87]. The low viscosity of SNP suspension could increase the absorption of starch on cellulose substrate. The SNPs were also suggested as a candidate for industrial application such as thickening agents.

5. Application of starch nanoparticles 5.1. Application in composites “Composites” are materials that are composed of two types of components: the matrix whose role is to support and protect the filler materials, transmit and distribute the applied load to them and the mentioned fillers, which are the stronger and stiffer components that reinforce the matrix. Furthermore, “Nanocomposites” mean polymeric composite materials that are filled with nanosized rigid particles [88]. The advantages of these nanocomposite materials, when compared with conventional composites, are their superior mechanical, barrier, and thermal properties at low levels (e.g., ≤5 wt%) as well as their recyclability, transparency and low weight [89,90]. As a biodegradable and nontoxic polymer, starch has been widely used in non-food applications including papers, textiles, plastics, cosmetics and pharmaceuticals. However, SNPs has been suggested as a reinforcement filler in the polymeric composites in many researches. A wide range of polymeric matrixes including both natural and synthetic polymers have been suggested for the composites with SNPs [13,88,91–102]. Synthetic polymers are versatile materials for many industrial applications because of their excellent physical properties and chemical resistance. However, most of the synthetic polymers are not biodegradable and biocompatible. For these reasons, recent researches focuses on the use of environmentally-friendly polymers including a variety of natural polymers such as starch, pullulan, polylactide, and soybean protein. By incorporating the SNPs into the synthetic polymer matrix, not only the physical properties but also the biodegradability of the composite enhances [12,13]. Several studies [60,78,103] emphasized the importance of having a uniform mixing of the reinforcing filler within the polymeric matrix of a composite formulation. Especially when the content of fillers became increased (more than 40%), the authors believed that the nanocrystals tended to self-aggregate, which decreased the

surface area for mutual interactions with polymers resulting in the decreases in strength and modulus. Composite films can be prepared from the blends of SNPs and polymers in different methods using various techniques such as casting and solvent evaporation, hot-pressing, compression molding, and extruding [104]. Except the casting-evaporation method, the composite films are subjected to a relatively high temperature and/or pressure [37]. Thus, a substantial melting of the SNPs may occur during the composite preparation, especially when water is present in the blends. For this reason, the casting and evaporation method was suggested as a better choice than other thermal methods for film preparation [80]. The mechanical properties such as tensile strength ( b , MPa), elasticity or Young’s modulus (E, MPa), and elongation at break (εb , %) of the nanocomposites containing SNPs are summarized in Table 2. The reinforcing effect of SNPs was evident from the tensile tests of their composite products. For the most cases, the incorporation of SNPs resulted in increases in both tensile strength and elastic modulus of the composites but a decrease in elongation at break [12,37,79,106,109,110] as shown in Table 2. For instance, nanocomposites of natural rubber (NR) with waxy maize SNPs exhibited reinforcing effects from the SNPs [105]. Up to the SNPs content around 20 wt%, εb slightly decreased from 1960 to 1500%, and then decreased more rapidly down to 920% for the NR film filled with 30 wt% SNPs. Whereas, E increased nearly exponentially with SNPs content, from 0.64 MPa for the unfilled NR matrix up to 77.8 MPa for the composite films filled with 30 wt%. The SNPs appeared to function as a good substitute for carbon black since the addition of only 10 wt% of SNPs to NR induced the reinforcing effect similar to that observed with 26.6 wt% addition of carbon black [111]. However, SNPs may be not so competitive as cellulose nanoparticles, so that a higher amount of the SNPs is necessary to achieve similar reinforcing effect. One of the major reasons for this is the difference in geometry of the nanoparticles. The cellulose nanocrystals (CNC) have a higher aspect ratio (50–500 nm in length, and 3–5 nm in diameter) [104] than SNPs which have a length of 20–40 nm and a width of 15–30 nm [84]. It has been reported that geometrical aspect ratio, which is defined as the ratio of length-to-diameter (L/d), is an important factor of the composite fillers that controls the mechanical properties of the nanocomposites. The nanoparticles that have a higher aspect ratio give the better reinforcing effect [91,112–114]. 5.2. Emulsion stabilizer Various kinds of particles, e.g., hydrophobic silica [115–119], carbon nanotubes [120], latex [121,122], microfibrillated cellulose [123], and bacterial cellulose nanocrystals [124] have been used as emulsion stabilizers. In a recent study, SNPs were suggested to

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14 12

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

be used as a stabilizer in oil-in-water emulsions [125]. An emulsion prepared with equivalent volume ratio of water and paraffin could be stabilized by adding an aliquot of SNP dispersion. Li et al. claimed that the addition of SNPs more than 0.02 wt% stabilized the emulsion longer than 2 month of storage without coalescence of oil droplets [125]. However, the emulsion became phase-separated when the SNPs melted by heating at 80 ◦ C for 2 h. The SNPs may be used in various emulsions not only for foods but cosmetics and pharmaceuticals. However, additional study is needed for the utilization of SNPs in this respect. 5.3. Fat replacers Another specific application of SNPs is the use as fat replacers in foods. Fat replacers are the substances that imitate the organoleptic or physical properties of triglycerides but cannot replace fat on a gram-for-gram basis [126]. Native and modified starches can sometimes be used to replace fat [127]. The particle size of the starches plays an important role in determining both the fat-like taste and mouthfeel [128]. Most common modifications of starch are used as fat replacer includes chemical depolymerization (e.g., acid or enzymatic hydrolysis) or mechanical disintegration of starch granules. The SNPs may be a promising candidate for a fat replacer because of the small size. It is expected that when SNPs are blended with other components, the mixture will form a smooth cream-like substance that has properties similar to fat. Additionally, the use of SNPs may result in calorie decrease by replacing high calorie fat with the carbohydrate. However, there has been no study on the use of SNPs as a fat replacer. 5.4. Packaging component The SNPs have attracted interest as a material in barrier films for food packaging. Barrier property of the packaging films focuses primarily on water vapor transmission and oxygen permeability. Several studies reported a decrease in water vapor permeability when the maize SNPs were incorporated. Kristo and Biliaderis reported that adding 30–40% waxy maize SNPs led to a significant decrease in water vapor permeability of sorbitol-plasticized pullulan film [12]. However, García et al. reported a decrease in the permeability by 40% for a cassava starch film by reinforcing with 2.5% SNPs [129]. The SNPs may also improve the barrier properties with respect to oxygen permeability. Angellier et al. demonstrated that the SNPs from waxy maize starch could reduce the oxygen diffusion and permeability of a nanocomposite film prepared with natural rubber (NR) [105]. This finding was attributed to the platelet-like SNPs, which might block the migration of oxygen molecules through the film. 5.5. Drug carrier and implant material Starch is nontoxic, biodegradable, and biocompatible polymer [130,131]. Thus it is an excellent carrier for drug delivery. More recently, SNPs also have received a great attention in this application. Nanoparticles have the ability to deliver an ample range of molecules to different locations in the body for sustained periods of time. A higher intracellular uptake of nanoparticles has been reported compared to micron-sized particles. Nanoparticles can access a variety of biological tissues because of their tiny size and mobility [132]. Indeed, the nanoparticles larger than 230 nm in size could congregate in the organ especially in the spleen due to the capillary size of this organ [132]. Starch has been used as a delivery carrier for tumor-targeted drugs [133] and transdermal drugs [134]. It was used as a carrier for phenethylamines [135], acetylsalicylic acid (Aspirin) [136], and estrone [137]. Chemically modified starches or SNPs were also reported to be used in

sustained drug delivery systems. For example, a cross-linked high amylose starch was used as a matrix for the controlled release of contramid [138]. Propylated SNPs that had been loaded with different types of drug (flufenamic acid, testosterone, and caffeine) showed enhanced effectiveness in the permeation through human skin [134]. Dialdehyde starch nanoparticles (DASNPs) that were conjugated with 5-fluorouracil were found to have enhanced inhibition in vitro against breast cancer cell (MF-7) compared to free 5-Fu [139]. However, some studies [140,141] have mentioned that the nanoparticles used in drug delivery may cause a toxic effect in human body. However, there is no demand of research on the safety of SNPs, especially when those are injected directly in veins. In addition, biodegradability and biocompatibility of SNPs make them an excellent candidate for implant materials. Biodegradable implants offer clear advantages over traditional metal implants which often cause stress shielding [142,143]. However, SNPs can retain their strength long enough to support a healing bone, and then they can gradually and harmlessly disintegrate in the patient’s body [144]. More research should be followed for the practical application in this area. 5.6. Adsorbents for wastewater treatment Chemically modified SNPs can be used as adsorbents for the removal of aromatic organic pollutants from water [145]. Currently, activated carbon is the most commonly used adsorbent for the removal of toxic organic substances from water [146–148]. However, in spite of its prolific use, activated carbon is expensive [149,150]. A synthetic polymer may also be used, but there is a growing interest in developing renewable and low-cost alternatives. Biopolymer-based materials may be the most attractive adsorbent for wastewater treatments. A recent study by Morandi et al. showed the large absorption potential of polystyrene-modified cellulose nanowhiskers for the removal of aromatic organic molecules from water [151]. Alila et al. first reported the potential use of starch nanoparticles as adsorbent after modified by grafting with stearate [145]. Compared to native biopolymers, the nanoparticles derived from those have increased efficiency and capacity due to the increased surface area, as proved by the cellulose nanowhiskers [151]. The chemical modification of the nanoparticles (e.g., grafting, cross-linking, etc.) further enhanced the adsorption capacity [145]. For instance, adsorption capacity of the SNP-g-stearate ranged between 150 and 900 ␮ mol g−1 for 2-naphthol and nitrobenzene, respectively, whereas unmodified SNPs exhibited much lower adsorption of 50 and 40 ␮ mol g−1 [145]. The driving force for the adsorption was assumed to be largely governed by Van der Waals interactions between the grafted chains and the organic solute. The authors explained that any factor that increased these interactions would contribute to an enhancement of the adsorption capacity. In particular, the planar structure of the aromatic organic solutes favor the intercalation of the molecules inside the domain that is formed by the grafted chains. This finding may account for the significant enhancement of the adsorption capacity after the grafting of the linear alkyl chains onto the nanocrystals surface. 5.7. Thermo-responsive conducting Valodkar et al. synthesized a crosslinked SNPs with HMDI (1,4-hexamethylene diisocyanate) which were insoluble and sufficiently reactive for the use in synthesis of nanocomposites [152]. It is well-known that the aliphatic chain of six methylene groups in HMDI is highly flexible among various isocyanates [152]. The SNPs behave as nanofillers and crosslinkers in the synthesis of PPG-based polyurethanes (PU). The highly crosslinked and nanofilled polymeric systems are likely to exhibit a good electrical

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

13

Table 3 Applications of starch nanoparticles. Application fields Food application Emulsion stabilizer Fat replacer Other industries application Composites filler Packaging component Biomedical materials Water treatment agent Thermo-responsive conducting application Binder in papermaking and paper coating

Roles of starch nanoparticles

References

Stabilization of oil-in-water emulsion against coalescence Improvement of the taste and mouthfeel due to the small particle size

[125] [127,128]

Reinforcement of polymeric matrix Oxygen and water vapor barrier (When platelet-like SNPs were incorporated into matrix, oxygen or water vapor permeability was decreased.) Drug release regulator for sustained drug delivery system Biodegradable implant material Absorbents for the removal of dissolved organic polluants from water Improvement of thermo-responsive electrical conductivity by acting as crosslinker and nanofiller Decrease of paste viscosity, increase of binding capability

[105,37,106,12,108,110] [37,129,31]

conductivity with a high rigidity. This effect is further enhanced by an internal plasticizing effect of the nanoparticles. The PPG-HMDImodified SNP PU film exhibited interesting conducting behavior which was observed to be temperature and frequency dependent. Notably, a considerable effect of the temperature on the conductivity indicated that the PU nanocomposites had a thermoresponsive behavior. Ionic conductivity was found to increase the temperature, an activation energy of 0.15 eV, which indicated that there was a thermally activated conduction mechanism in the nanocomposite. This behavior is attributed to an increase in the charge carrier energy with a rise in the temperature. Based on this result, the polyurethane nanocomposites containing the modified SNPs are potentially applicable in temperature sensors. 5.8. Binder in papermaking and paper coating The SNPs can be used as a binder in papermaking and paper coating. The cooked starch has been widely used as paper-making additives. The retention of the cooked starch on the paper matrix is based on the absorption of starch. Thus, the absorbed amount of starch is limited by the saturation of absorption on cellulose substrates. Another problem is the high viscosity of the starch paste after cooking of raw starch, which might cause operational problems. Bloembergen et al. reported an improvement of the performance as a binder for paper by using SNPs instead of cooked starches [153]. With the addition of SNPs, the viscosity of the paste can be substantially reduced, whereas the binding capability can be increased. Other examples of the applications of SNPs in various fields are summarized in Table 3. 6. Conclusions For the past few decades, the nanoparticles originated from polysaccharides (e.g., starch, cellulose, chitin, etc.) have been extensively studied in their preparation and utilization. Despite the great potentials of starch nanoparticles for industrial applications, more research should be done on the efficient nanoparticle production on large industrial scale. A simple and economic process and improvement of recovery yield from the hydrolysis still remain as tasks. Additionally, starch nanoparticles have a strong tendency to aggregate, so the recovery as powder products became difficult. However, the extensive hydrophilic nature of the SNPs provides the significantly high reactivity, which may be one of the advantages of the SNPs. The SNP incorporation resulted in the physical properties superior to the conventional microscale composites. Even when a small amount of nanoparticles were added to a matrix, nanocomposites have a strong reinforcing effect. In addition, positive impact such as improvement in barrier properties was also found. It is obvious that the starch-based nanoparticles are one of

[134,139,144,67] [145] [152] [153]

the potential fillers not only for the mechanical properties but for the renewability and biodegradability. Although most researches for the application of SNPs have been focused on the nanocomposites, utilization as carriers for biofunctional ingredients and drugs should be considered. In food industries, SNPs may be used as a novel ingredient for the control of rheological properties and texture. One of the potential applications would be the replacement of fat. Acknowledgements This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2011-0023595). This research was also supported by a Korea University Grant. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]

D.J. Gallant, B. Bouchet, P.M. Baldwin, Carbohydr. Polym. 32 (1997) 177. H. Tang, T. Mitsunaga, Y. Kawamura, Carbohydr. Polym. 63 (2006) 555. A. Dufresne, Can. J. Chem. 86 (2008) 484. N.L. García, L. Ribba, A. Dufresne, M.I. Aranguren, S. Goyanes, Carbohydr. Polym. 84 (2011) 203. H. Namazi, A. Dadkhah, Carbohydr. Polym. 79 (2010) 731. J.L. Putaux, S. Molina-Boisseau, T. Momaur, A. Dufresne, Biomacromolecules 4 (2003) 1198. H.-Y. Kim, J.H. Lee, J.-Y. Kim, W.-J. Lim, S.-T. Lim, Starch/Stärke 64 (2012) 367. J.-Y. Kim, D.-J. Park, S.-T. Lim, Cereal Chem. 85 (2008) 182. D. Liu, Q. Wu, H. Chen, P.R. Chang, J. Colloid Interface Sci. 339 (2009) 117. S. Bel Haaj, A. Magnin, C. Pétrier, S. Boufi, Carbohydr. Polym. 92 (2013) 1625. F. Giezen, R., Jongboom, K., Gotlieb, A. Boersma, Biopolymer nanoparticles. U. S. Patent WO 00/69916, January 25, 2000. E. Kristo, C.G. Biliaderis, Carbohydr. Polym. 68 (2007) 146. H. Angellier, S. Molina-Boisseau, L. Lebran, A. Dufresne, Macromolecules 38 (2005) 3783. N. Lin, J. Huang, P.R. Chang, L. Feng, J. Yu, Colloids Surf. B 85 (2011) 270. A. Bodin, H. Backdahl, B. Risberg, P. Gatenholm, Tissue Eng. 13 (2007) 885. P. Wongpanit, N. Sanchavanakit, P. Pavasant, T. Bunaprasert, Y. Tabata, R. Rujiravanit, Eur. Polym. J. 43 (2007) 4123. J.P. Mua, D.S. Jackson, J. Agric. Food Chem. 45 (1997) 3840. A. Buléon, P. Colonna, V. Planchot, S. Ball, Int. J. Biol. Macromol. 23 (1998) 85. C.G. Biliaderis, in: H.R. Water (Ed.), Polysaccharide Association Structures in Foods, Marcel Dekker, New York, 1998, p. 57. W.R. Morrison, J. Karkalas, in: P.M. Dey (Ed.), Starch, Methods in Plant Biochemistry, Academic Press, London, 1990, p. 323. R.F. Tester, J. Karkalas, X. Qi, J. Cereal Chem. 39 (2004) 151. J.M.V. Blanshard, D.R. Bates, A.H. Muhr, D.L. Worcester, J.S. Higgins, Carbohydr. Polym. 4 (1984) 427. G.T. Oostergetel, E.F.J. Van Bruggen, Starch/Stärke 41 (1989) 331. C.J. Sterling, J.A. Radley, Starch and Its Derivatives, Chapman & Hall, London, 1968, pp. 139. J.P. Robin, C. Mercier, R. Charbonnière, A. Guilbot, Cereal Chem. 51 (1974) 389. D. French, in: R.L. Whistler, J.N. BeMiller, J.F.L. Paschal (Eds.), Starch: Chemistry and Technology, Academic Press, Orlando, 1984, p. 183. D.J. Manners, Carbohydr. Polym. 11 (1989) 87. A. Imberty, H. Chanzy, S. Perez, A. Buléon, V. Tran, Macromolecules 20 (1987) 2634. A. Imberty, S. Perez, Biopolymers 27 (1988) 1205.

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011

G Model COLSUB-6734; No. of Pages 14 14

ARTICLE IN PRESS H.-Y. Kim et al. / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

[30] T. Ya Bogracheva, V.J. Morris, S.G. Ring, C.L. Hedley, Biopolymers 45 (1998) 323. [31] X. Ma, R. Jian, P.R. Chang, J. Yu, Biomacromolecules 9 (2008) 3314. [32] Y. Tan, K. Xu, L. Li, C. Liu, C. Song, P. Wang, ACS Appl. Mater. Interfaces 1 (2009) 956. [33] D. LeCorre, J. Bras, A. Dufresne, Biomacromolecules 11 (2010) 1139. [34] R. Hoover, Food Rev. Int. 16 (2000) 369. [35] C.G. Biliaderis, D.R. Grant, J.R. Vose, Cereal Chem. 58 (1981) 496. [36] L. Jayakody, R. Hoover, Food Res. Int. 35 (2002) 665. [37] H. Angellier, J.-L. Putaux, S. Molina-Boisseau, D. Dupeyre, A. Dufresne, Macromol. Symp. 221 (2005) 95. [38] K. Kainuma, D. French, Biopolymers 10 (1971) 1673. [39] W.C. Mussulmann, J.A. Wagoner, Cereal Chem. 45 (1968) 162. [40] W. Li, H. Corke, T. Beta, Carbohydr. Polym. 69 (2007) 398. [41] A. Dufresne, J.Y. Cavaille, W. Helbert, Macromolecules 29 (1996) 7624. [42] D. LeCorre, J. Bras, A. Dufresne, J. Nanopart. Res. 13 (2011) 7193. [43] H. Angellier, L. Choisnard, S. Molina-Boisseau, P. Ozil, A. Dufresne, Biomacromolecules 5 (2004) 1545. [44] P.R. Chang, F. Ai, Y. Chen, A. Dufresne, J. Huang, J. Appl. Polym. Sci. 111 (2009) 619. [45] D. LeCorre, E. Vahanian, A. Dufresne, J. Bras, Biomacromolecules 13 (2011) 132. [46] L. Martin-Banderas, M. Flores-Mosquera, P. Riesco-Chueca, A. Rodriguez-Gil, A. Cebolla, S. Chavez, A.M. Ganan-Calvo, Small 1 (2005) 688. [47] M.S. Hale, J.G. Mitchell, Nano Lett. 1 (2001) 617. [48] D. Song, Y.S. Thioc, Y. Deng, Carbohydr. Polym. 85 (2011) 208. [49] J.Y. Zuo, K. Knoerzer, R. Mawson, S. Kentish, M. Ashokkumar, Ultrason. Sonochem. 16 (2009) 462. [50] P. Tomasik, M.F. Zaranyika, Adv. Carbohydr. Chem. Biochem. 51 (1995) 243. [51] S.B. Brown, C.M. Orlando, H.F. Mark, N.M. Bikales, C.G. Overberger, G. Menges, J.I. Kroschwitz, Encyclopedia of Polymer Science and Engineering, 14, Wiley, New York, 1988, pp. 169. [52] L.S. Lai, J.L. Kokini, J. Rheol. 34 (1990) 1245. [53] M. Lamanna, N.J. Morales, N.L. García, S. Goyanes, Carbohydr. Polym. 97 (2013) 90. [54] S. Singh, N. Singh, R. Ezekiel, A. Kaur, Carbohydr. Polym. 83 (2011) 1521. [55] Y. Yu, J. Wang, Food Res. Int. 40 (2007) 297. [56] N.L. Garcia, M. Lamanna, N. D’Accorso, A. Dufresne, M. Aranguren, S. Goyanes, Polym. Degrad. Stabil. 97 (2012) 2021. [57] H.-Y. Kim, J.-A. Han, D.-K. Kweon, J.-D. Park, S.-T. Lim, Carbohydr. Polym. 93 (2013) 582. [58] D. Cooke, M.J. Gidley, Carbohydr. Res. 227 (1992) 103. [59] H.-Y. Kim, D.-J. Park, J.-Y. Kim, S.-T. Lim, Carbohydr. Polym. 98 (2013) 295. [60] A.P. Mathew, A. Dufresne, Biomacromolecules 3 (2002) 609. [61] J.-Y. Kim, S.-T. Lim, Carbohydr. Polym. 76 (2009) 110. [62] Q. Sun, M. Gong, Y. Li, L. Xiong, Carbohydr. Polym. 111 (2014) 133. [63] Y.Y. Fang, L.J. Wang, D. Li, B.Z. Li, B. Bhandari, X.D. Chen, Z.H. Mao, Carbohydr. Polym. 74 (2008) 379. [64] O. Frassen, W.E. Hennink, Int. J. Pharm. 168 (1998) 1. [65] A.-M. Shi, D. Li, L.-J. Wang, B.-Z. Li, B. Adhikari, Carbohydr. Polym. 83 (2011) 1604. [66] G. Zhou, Z. Luo, X. Fu, Ind. Crop. Prod. 52 (2014) 105. [67] G. Zhou, Z. Luo, X. Fu, J. Agric. Food. Chem. 62 (2014) 8214. [68] K. Urban, G. Wagner, D. Schaffner, D. Roglin, J. Ulrich, Chem. Eng. Technol. 29 (2006) 24. [69] J.R. Canselier, H. Delmas, A.M. Wilhelm, B. Abismail, J. Dispers. Sci. Technol. 23 (2002) 333. [70] J.M. Asua, Prog. Polym. Sci. 27 (2002) 1283. [71] S. van der Graaf, C.G.P.H. Schroen, R.M. Boom, J. Membr. Sci. 251 (2005) 7. [72] S.M. Jafari, E. Assadpoor, Y. He, B. Bhandari, Food Hydrocoll. 22 (2008) 1191. [73] B. Wei, X. Hu, H. Li, C. Wu, X. Xu, Z. Jin, Y. Tian, Food Hydrocoll. 36 (2014) 369. [74] L. Ren, M. Jiang, J. Zhou, J. Tong, Carbohydr. Polym. 87 (2012) 1874. [75] B. Duan, P. Sun, X. Wang, C. Yang, Starch/Stärke 63 (2011) 528. [76] S. Nara, T. Komiya, Starch/Stärke 35 (1983) 407. [77] S. Song, C. Wang, Z. Pan, X. Wang, J. Appl. Polym. Sci. 107 (2008) 418. [78] H. Zheng, F. Ai, P.R. Chang, J. Huang, A. Dufresne, Polym. Compos. 30 (2009) 474. [79] G. Chen, M. Wei, J. Chen, J. Huang, A. Dufresne, P.R. Chang, Polymers 49 (2008) 1860. [80] D. LeCorre, J. Bras, A. Dufresne, Carbohydr. Polym. 87 (2012) 658. [81] H.M. Lai, Food Chem. 72 (2001) 455. [82] Z.-G. Luo, X. Fu, Q.-Y. Gao, S.-J. Yu, Int. J. Food Sci. Technol. 46 (2011) 429. [83] T.A. Waigh, M.J. Gidley, B.U. Komanshek, A.M. Donald, Carbohydr. Res. 328 (2000) 165. [84] H. Angellier, J.-L. Putaux, S. Molina-Boisseau, A. Dufresne, E. Bertoft, S. Perez, Carbohydr. Res. 344 (2009) 1558. [85] S. Hizukuri, Carbohydr. Res. 147 (1986) 342. [86] S. Hizukuri, Carbohydrate in Food, Mercel Dekker, New York, 1996, pp. 347. [87] S. Bloembergen, F., Kappen, B. Beelen, Environmentally friendly biopolymer adhesives and applications based thereon. U.S. Patent 6921430 B2, July 26, 2005. [88] K.G. Nair, A. Dufresne, Biomacromolecules 4 (2003) 666.

[89] K. Oksman, A.P. Mathew, D. Bondeson, I. Kvien, Compos. Sci. Technol. 66 (2006) 2776. [90] A. Sorrentino, G. Gorrasi, V. Vittoria, Trends Food Sci. Technol. 18 (2007) 84. [91] D. Dubief, E. Samain, A. Dufresne, Macromolecules 32 (1999) 5765. [92] A. Dufresne, M.B. Kellerhals, B. Witholt, Macromolecules 32 (1999) 7396. [93] M.N. Anglès, A. Dufresne, Macromolecules 33 (2000) 8344. [94] Y.Y. Noishiki, Y. Nishiyama, M. Wada, S. Kuga, J. Magoshi, J. Appl. Polym. Sci. 86 (2002) 3425. [95] M. Grunert, W.T. Winter, J. Polym. Environ. 10 (2002) 27. [96] W. Helbert, J.-Y. Cavaillé, A. Dufresne, Polym. Compos. 17 (1996) 604. [97] L. Chazeau, J.Y. Cavaillé, G. Canova, R. Dendievel, B. Boutherin, J. Appl. Polym. Sci. 71 (1999) 1797. [98] M. Matos Ruiz, J.-Y. Cavaillé, A. Dufresne, C. Graillat, J.-F. Gérard, Macromol. Symp. 169 (2001) 211. [99] K.G. Nair, A. Dufresne, Biomacromolecules 4 (2003) 657. [100] K.G. Nair, A. Dufresne, Biomacromolecules 4 (2003) 1835. [101] M. Paillet, A. Dufresne, Macromoleulces 34 (2001) 6527. [102] A. Morin, A. Dufresne, Macromolecules 35 (2002) 2190. [103] D.W. Schaefer, R.S. Justice, Macromolecules 40 (2007) 8501. [104] R.J. Moon, A. Martini, J. Nairn, J. Simonsen, J. Youngblood, Chem. Soc. Rev. 40 (2011) 3941. [105] H. Angellier, S. Molina-Boisseau, S. Dufresne, Macromolecules 38 (2005) 9161. [106] H. Angellier, S. Molina-Boisseau, P. Dole, A. Dufresne, Biomacromolecules 7 (2006) 531. [107] X. Wang, L. Zhang, J. Nanosci. Nanotechnol. 8 (2008) 5831. [108] J. Viguié, S. Molina-Boisseau, A. Dufresne, Macromol. Biosci. 7 (2007) 1206. [109] J. Yu, F. Ai, A. Dufresne, S. Gao, J. Huang, P.R. Chang, Macomol. Mater. Eng. 293 (2008) 763. [110] Y. Chen, X. Cao, P.R. Chang, M.A. Huneault, Carbohydr. Polym. 73 (2008) 8. [111] M. Arroyo, M.A. Lopez-Manchado, B. Herrero, Polymer 44 (2003) 2447. [112] N.L.G. de Rodriguez, W. Thielemans, A. Dufresne, Cellulose 13 (2006) 261. [113] B. Jiang, C. Liu, C. Zhang, B. Wang, Z. Wang, Composites B 38 (2007) 24. [114] K.K.H. Wong, M. Zinke-Allmang, J.L. Hutter, S. Hrapovic, J.H.T. Luong, W. Wan, Carbon 47 (2009) 2571. [115] B.P. Binks, S.O. Lumsdon, Langmuir 16 (2000) 8622. [116] B.P. Binks, C.P. Whitby, Langmuir 20 (2004) 1130. [117] B.R. Midmore, Colloids Surf. A 132 (1998) 257. [118] N.P. Ashby, B.P. Binks, Phys. Chem. Chem. Phys. 2 (2000) 5640. [119] Y. Nonomura, N. Kobayashi, J. Colloid Interface Sci. 330 (2009) 463. [120] M. Shen, D.E. Resasco, Langmuir 25 (2009) 10843. [121] N.P. Ashby, B.P. Binks, V.N. Paunov, Chem. Commun. 4 (2004) 436. [122] B.P. Binks, S.O. Lumsdon, Langmuir 17 (2001) 4540. [123] M. Andresen, P. Stenius, J. Dispers. Sci. Technol. 28 (2007) 837. [124] I. Kalashnikova, H. Bizot, B. Cathala, I. Capron, Langmuir 27 (2011) 7471. [125] C. Li, P. Sun, C. Yang, Starch/Stärke 64 (2012) 497. [126] C.C. Akoh, Food Technol. 52 (1998) 47. [127] J. Jane, L. Shen, L. Wang, C.C. Maningat, Cereal Chem. 69 (1992) 280. [128] J.R. Daniel, R.L. Whistler, CFW 35 (1990) 825. [129] N.L. García, L. Ribba, A. Dufresne, M.I. Aranguren, S. Goyanes, Macromol. Mater. Eng. 294 (2009) 169. [130] A. El-Hag Ali, A. AlArifi, Carbohydr. Polym. 78 (2009) 725. [131] L. Chen, X. Li, L. Li, S. Guo, Curr. Appl. Phys. 7 (2007) 1, e90. [132] J. Kreuter, J. Control. Release 16 (1991) 169. [133] S.Y. Xiao, C.Y. Tong, X.M. Liu, Chin. Sci. Bull. 51 (2006) 1693. [134] M.J. Santander-Ortega, T. Stauner, B. Loretz, J.L. Ortega-Vinuesa, D. BastosGonzalez, G. Wenz, U.F. Schaefer, C.M. Lehr, J. Control. Release 141 (2010) 85. [135] B.Z. Weiner, M. Tahan, A. Zikha, J. Med. Chem. 15 (1972) 410. [136] P. Stjarnkvist, T. Laakso, I. Sjoholm, J. Pharm. Sci. 78 (1989) 52. [137] C. Won, C. Chu, T. Yu, Carbohydr. Polym. 32 (1997) 239. [138] V. Lenaerts, I. Moussa, Y. Dumoulin, F. Mebsout, F. Chouinard, P. Szabo, M.A. Mateese, L. Cartilier, R. Marchessault, J. Control. Release 53 (1998) 225. [139] S.Y. Xiao, X.M. Liu, C.Y. Tong, Chin. Sci. Bull. 57 (2012) 3226. [140] P.H. Hoet, I. Bruske-Hohlfeld, O.V. Salanta, J. Nanobiotechnol. 2 (2004) 12. [141] A. Nel, T. Xia, L. Madler, N. Li, Science 311 (2006) 622. [142] D.S. Brodke, S. Gollogly, R. Alexander Mohr, B.K. Nguyen, A.T. Daily, A.K. Bachus, Spine 26 (2001) 1324. [143] M.C. Kennady, M.R. Tucker, G.E. Lester, M.J. Buckley, Int. J. Oral Maxillofac. Surg. 18 (1989) 170. [144] J.T. Gourville, Harv. Bus. Rev. 83 (June (6)) (2005) 35–38, 40, 42 passim. [145] S. Alila, F. Aloulou, W. Thielemans, S. Boufi, Ind. Crop Prod. 33 (2011) 350. [146] T.O. Philip, J.J. Gary, R. Geoffery, Hamilton, Water Res. 38 (2004) 4455. [147] J.M. Chern, Y.W. Chien, Water Res. 37 (2003) 2347. [148] K. Kamegawa, K. Nishikubo, M. Kodama, Y. Adachi, H. Yoshida, Colloids Surf. A 254 (2005) 31. [149] Y.C. Sharma, S.S.N. Uma, G.F. Paras, Chem. Eng. J. 132 (2007) 319. [150] Y.C. Sharma, B. Singh, A. Agrawal, C.H. Weng, J. Hazard. Mater. 151 (2008) 789. [151] G. Morandi, L. Heath, W. Thielemans, Langmuir 25 (2009) 8280. [152] M. Valodkar, S. Thakore, Carbohydr. Res. 345 (2010) 2354. [153] S. Bloembergen, I. McLennan, D.I. Lee, J.V. Leeuwen, Paper 360◦ 2008, September: 46.

Please cite this article in press as: H.-Y. Kim, et al., Preparation, characterization and utilization of starch nanoparticles, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.011