Abstract： The potential reactions between natural polysaccharides and iodine and their products have been explored for a long time. Due to the complex factors that can influence these reactions， a clear-cut mechanism has not yet been developed. Starch-iodine complexes， especially the amylose-iodine complex， are the most investigated of the polysaccharide-iodine reactions， and the study of this reaction can be used as a basis for the investigation of other polysaccharide-iodine reactions. In this paper， significant aspects of the reaction were introduced， including the influence of the polysaccharide structure on the properties of the resulting complexes， the relationship between the concentration of CaCl2 and formation of the final products， as well as the form of the polyiodides in these complexes. The interior structure and the surface morphology of the complexes were discussed， along with the progress in research related to this reaction.
　　Keywords： amylose； hemicelluloses； I2-KI； polyiodide； polysaccharides
　　1 Introduction
　　Lignocellulosic biomass is the most abundant raw material， and has potential in the sustainable production of eco-materials， bioethanol， chemicals， food additives， and pulp and paper products[1-5]. Natural polysaccharides are the predominant component of lignocellulosic biomass materials， and have versatile properties. Most natural polysaccharides have been found to react with iodine under some conditions. The first iodine-organic complex， starch-iodine， was discovered in 1814 by two French chemists， and has been widely used[6]. Since then， many natural polysaccharides have been found to react with iodine to form a colored complex. Hemicellulosic polysaccharides， which are a class of natural polysaccharides， can also react with iodine， but the forming mechanism of hemicellulosic polsaccharides-iodine complexes is not the same as that of the starch-iodine complex[7-9]. Practically， the mechanism of the reaction and structure of the complexes have remained unclear due to many factors that influence this reaction. To our dismay， only a limited number of researchers have investigated this issue， and the range of available references is relatively narrow. In this paper， several representative examples of natural polysaccharide-iodine complexes were chosen in order to introduce some proposed theories for this reaction based on experimental results， and we sincerely hope that this review will be helpful in the further study and utilization of lignocellulosic biomass. 　　2 Main factors influencing the reaction
　　2.1 The structure of the polysaccharides
　　The dark-blue color of polysaccharide-iodine complexes was observed in 1886 by Swiss chemist， Hans Heinrich Landolt， in the reaction known as the “iodine clock reaction” or “Landolt reaction”[10]. The final color of the complex is predominantly determined by the structure of the polysaccharide. Based on spectrophotometry results， Swanson proposed that the length of the polysaccharide chain influenced the hue of the complex[11]. For acid-hydrolyzed or enzyme-degraded amylose-iodine complexes， no color was observed for chains consisting of 4～6 glucose units， while chains with 8～12 units formed a red complex with iodine. As the number of glucose units in the polysaccharide chain increased to 30～35， a blue amylose-iodine complex was obtained. The results of Mould and Synge[12-13]， which were based on the continuous electrophoretic fractionation of α-amylase-hydrolyzed amylose in the presence of iodine-iodine， revealed that hydrolyzed amylose with DP （degree of polymerization） of 40～130， 25～40， 19～25， and less than 10 produced amyloses-iodine complexes that were blue， red， orange， and colorless， respectively. Another similar theory regarding the relationship between the molecular weight of the polysaccharides and the color of the resulting complexes was developed by Gaillard[14]. Dark purple and brown complexes were obtained when xylodextrin and xylotetraose were reacted with iodine， respectively， while xylobiose and xylotriose produced only light brown complexes under the same conditions. Further evidence was provided by Handa and Yajima[15]， who used circular dichroism spectroscopy （CD） to characterize the amylose-triiodide reaction. The product was observed to develop a blue color when the DP of the amylose chain reached 30， and deepened as the DP was increased to 100 or more.
　　In addition to chain length， the structures of polysaccharides play important roles in their reactions with iodine. The “amyloid reaction” has frequently been used in the staining and identification of plant materials. This method was modified by Gaillard as a test to distinguish between linear and branched polysaccharides[14]. Based on Gaillard’s research， Morak and Thompson conducted a series of experiments to evaluate its potential[16]. Salep mannan， amylose， and glucomannan were characterized as linear polysaccharides because they reacted with iodine to form dark blue or blue-black complexes under the test conditions， which involved adding I2-KI solution to a mixture containing 1% polysaccharide solution and 30% aqueous CaCl2. However， other polysaccharides with a linear structure， such as pectin and pectin acids， did not react. Branched polysaccharides， such as amylopectin， cherry gum， gum arabic， and some polysaccharides with short terminal branches， including guar gum and locust bean gum， did not react with iodine at all under these conditions. Other polysaccharides with short terminal branches， such as 4-O-methylglucoronoxylans from deciduous woods， reacted to form colored complexes under the test conditions. Based on the results of Morak and Thompson， Gaillard and Bailey carried out similar experiments and provided their analysis[17]. According to the results they obtained in the reaction， the polysaccharides used in the experiment were divided into two groups （Table 1）. One group consisted of the positive polysaccharides， which could form complexes with iodine in concentrated CaCl2 solution， and the other one was composed of negative polysaccharides， which could not react with iodine under the same conditions. Five kinds of polysaccharide structures were listed in Table 1， and it was useful to analyze these in terms of these two groups. Most positive polysaccharides have linear or slightly branched structures， but not all polysaccharides with this structure can form complexes with iodine， under the test conditions described above. Pectin and isolichenin are exceptions. Although plant hemicellulose A， which has a relatively low degree of branching， amylopectin， and filiform B heterxylans are branched polysaccharides， they are positive polysaccharides because they can form a complex with iodine. The phenomenon suggests that a linear polysaccharide structure is not a prerequisite for the formation of the precipitate， while the presence of the 1→4 link segments between glucose， mannose， and xylose units is a vital factor. The results of Gaillard indicate the apparent importance of the 1→4 link in the reaction， and the ratio of 1→4 link to other links is usually 1：3 for positive polysaccharides. In the positive polysaccharides group， lichenin[18-19]， rhodymenan[20]， and oat and barley β-glucans[19-21] contain many segments of three or more 1→4 linked monosaccharides. Comparatively， the 1→4 link ratios of negative polysaccharides are much lower. Isolichenin has a ratio of 0.66， and contains few sequences with more than two 1→4 links between sugar units[22-23]. Therefore， the presence of 1→4 links in the structure of the polysaccharides plays a more important role than the linearity of the structure in the reaction. Although the 1→4 links seem crucial to the formation of colored complexes， they are not the only influencing factor. Some other structural characteristics may also be partly responsible for the formation of such complexes， such as the presence of successive hydroxyl groups； this aspect will be discussed later. 　　2.2 The CaCl2 solution
　　In general， the presence of CaCl2 is essential to the completation of hemicellulosic polysaccharides with iodine， and the concentration of CaCl2 solution greatly affects the formation of the precipitate. In some cases， Ca2+ can be replaced by another multivalent metal cation， and Cl- can also be replaced by Br-. A qualitative study showed that commercial xylan could react with iodine in aqueous CaCl2， Ca（NO3）2， CaI2， MgCl2， BaCl2， SrCl2， Al（NO3）3， and ZnCl2， but not in solutions of AlCl3， HgCl2， PbCl2， CuCl2， CdCl2， CoCl2， FeCl3， KCl， or NaI[24]. Coniferous glucomannan could form a complex in 30% aqueous AlCl3， MgCl2， and BaCl2 solutions， but could not be precipitated with iodine in aqueous FeCl3， CuCl2， NiCl2， or CoCl2[16]. A bright orange complex was obtained when bromine was added to a solution of elm 4-O-methyglucuronomannan in 30% aqueous CaCl2. The variation of multivalent salt solvents indicated that the effect of the cation is predominant in the polysaccharide-iodine reaction. The optical density of the elm xylan-iodine precipitate at 610 nm varies depending on the salt solution used， with the highest density being obtained in CaCl2 solution[24].
　　Amylose can be precipitated by iodine without concentrated CaCl2 solution. The amylose-iodine complex is most favored in pure water， with the favorability of the reaction decreasing as the proportion of nonaqueous solvent （such as acetone， ethanol， isopropanol， and dimethyl sulfoxide） is increased[25]. This is due to the fact that when amylose is exposed to nonaqueous solvents prior to the addition of iodine， the solvent molecules are likely to form rigid complexes， and once the nonaqueous solvent-amylose complex is formed， the iodine molecules are unable to displace the nonaqueous solvent molecules.
　　Yu and Atalla found that the concentration of the CaCl2 solution significantly affected the formation of the xylan-iodine complex[26]. The xylan-iodine complex exhibits UV-vis absorption with a lmax of 610 nm[24]. Fig.1 shows the relationship between the concentration of CaCl2 and the absorption at 610 nm. The UV-vis absorption of the complex at 610 nm decreases dramatically as the concentration of CaCl2 solution is decreased， and no signal is observed when the concentration of the CaCl2 solution is less than 25%. This plot demonstrates that concentrated CaCl2 solution strongly promotes polysaccharide-iodine complex formation. Williams and Atalla proposed that Ca2+ had a strong tendency to chelate saccharides and saccharide analogs through their vicinal hydroxyl groups[27-28]. On the basis of experiments and previous proposals， they found that CaCl2 combined with the hydroxyl groups of polysaccharides chains when the concentration of water was too low to completely hydrate the calcium cations. The conclusion was mainly inspired by the sharp increase in the mercerizability of cellulose when there was insufficient water to hydrate the sodium cation. A sharp increase in absorbance is observed as the CaCl2 concentration increases from 35% to 40% in Fig.1. Under these conditions， the hydroxyl groups of xylan began to chelate with the incompletely hydrated Ca2+. Therefore， CaCl2 promoted the formation of colored complexes of polysaccharides with iodine. 　　2.3 Polyiodide in the reaction
　　The iodine complexes of biopolymers are important， and have been used to fabricate materials. The inclusion of polyiodides in the colored complexes is a controversial point of the polysaccharide-iodine reaction. In the late 20th century， various researchers proposed that polyiodides such as I4 （I2·I2）， I5- （I3-·I2）， I6- （I3-·I3-） led to the appearance of the colored complex[29]， with the predominant form of polyiodide in the complex being I5-， and the I6 unit being mainly responsible for the color in the reaction[30]. These polyiodides originated from the I2-KI system， and thus polyiodides were often associated with KI. A typical example of the influence of the KI concentration on the complex formation described by Galliard and coworkers[24] is mentionescribed here. It was found that regardless of the polysaccharide concentration， the optical density at 610 nm became stronger as the concentration of KI increased. Thus， KI probably participates in or promotes the complex formation reaction. Theories proposing that KI enhances the formation of necessary iodide-iodine species， or that KI is vital for starch helix formation for starch helix， still require further confirmation[31-33].
　　Early potentiometric titrations were carried out using greater iodine concentrations to ensure the reaction of iodine with polysaccharides in the presence of CaCl2[24]. The results were in good agreement with those of Adskins and Greenwood[34]. For example， by plotting the amount of bound iodine against free iodine， as described by Anderson and Greenwood[35]， xylan was found to bind approximately 9 mg of iodine per 100 mg of polysaccharide， which was approximately half the amount measured for undecomposed amylose under slightly different titration conditions. Knutson et al[36] reported that the amount of iodine bound to 100 mg amylose was 30 mg on the basis of the potentiometric excess-iodine titration method. Later， more accurate instruments were used to investigate the reaction precipitated. Raman spectroscopy is a useful measurement in analyzing the structure of molecules. The most sensitive region of the Raman spectrum for the characterization of the iodine-polyiodide system is in the low-frequency spectral region of 25～250 cm-1. The Raman spectra of different polyiodides in this range have been widely investigated， and their characteristic signals have been identified： I2 is observed above 180 cm-1， I3- appears at approximately 110 cm-1， 　　and I5- is found at about 160 cm-1[37-42]. The weak signal that appears at approximately 110 cm-1 in the I5- sample should be assigned to the impurity I3- according to theoretical calculations and infrared spectral evidence[38]. The xylan-iodine reaction was investigated by Yu and Atalla using Raman spectroscopy[26]； the results indicated that the two peaks at 110 cm-1 and 160 cm-1 were sensitive to the concentration of KI （Fig.2）. The intensity of the peak near 160 cm-1 declined， while that of the peak at approximately 110 cm-1 was enhanced dramatically with increasing KI concentration. The peak at approximately 160 cm-1 was only a shoulder at a KI concentration of 1.0%； further increasing the KI concentration resulted in this shoulder becoming so weak that it was barely identifiable. The Raman spectra of amylose-iodine precipitates formed under different conditions with various concentrations of KI （0， 0.01%， and 0.1%） were investigated by Yu and his coworkers[43]. As shown in Fig.3， the two fundamental peaks at 110 cm-1 and 160 cm-1 varied with the iodine concentrations. Compared to that of the peak at 160 cm-1， the intensity of the peak at 110 cm-1 declined significantly with decreasing iodide concentration， and the precipitate formed in a saturated iodine solution without iodide showed only a shoulder at 110 cm-1. Furthermore， they found this intensity change was reversible. The peak at 110 cm-1 was enhanced when the iodide concentration was increased， and declined when the ratio was reduced. This indicates the existence of a dynamic equilibrium between the polyiodide chains and iodine. Interestingly， the peak at 160 cm-1 remained at full intensity regardless of the ratio used. The authors’ interpretation was that the scattering coefficient of I5- was higher than that of I3-， since it has more π electrons and its electronic absorption bands extend further into the visible range. Because the concentrations of I3- to I5- were related， the concentration of each could not be determined independently. However， it appears that only part of I5- can be converted into I3-， while a residual amount of I5- remains unaffected[44].
　　Another valuable result from Raman spectroscopy was reported by Yuguchi et al[45]. Fig.4 shows the Raman spectra of iodine with and without xyloglucan， which were obtained from the complex of xyloglucan （Cp=3.4 mg/mL） and iodine （[I2]=0.76 mg/mL） and a reference iodine solution. In the spectrum of reference iodine solution without xyloglucan， I3- species produced the peak at 110 cm-1， while in the spectrum of the xyloglucan-iodine complex， I5- species were responsible for the peak at 160 cm-1. The amylose-iodine complex also showed an I5- component[46-48]. These results indicated that I5- was the dominant component in the formation of the colored xyloglucan-iodine complex. Additionally， the color of the complex was affected by the polyiodide species. In the amylose-iodine complex， I5- produced a dark blue color， while I3- caused a reddish-purple color in amylopectin[49]. In addition to I3- and I5-， other forms of polyiodides are present in polysaccharide-iodine-polyiodide systems， and the structures of hypervalent polyiodides cannot be explained by simple covalent bond models[50]. These polyiodides decompose into iodine or iodide to some extent at room temperature or higher. The amount of iodine affects the color of the polyiodides， with more iodine producing a darker color， and higher polyiodides often having a metallic luster[51]. 　　A kinetic investigation of the amylose-iodine reaction was proposed by Thompson and Hamori[52]. The fast reaction among amylose， iodine， and iodide ions was investigated in aqueous solution by the rapid mixing technique. The free and bound iodine concentrations were determined by the changes at 6283 ? using a calibration procedure based on simultaneous potentiometric and spectrophotometric titrations. A rate law of -d[I2]/dt =k[P][I2]4-c[I-]3-c[I3]c was used during the initial course of the precipitation， and the value of the exponent c could only be established to the extent that 0≤c≤3， which was due to the equilibrium among the iodine， iodide， and triiodide ions. The mechanism proposed to explain these kinetic results suggests that the rate-determining step during the initial phase of the complexation is the slow formation of stable nuclei （apparently I113-） inside the amylose helices. In an early investigation， Gilbert and Marriott[31] proposed that at low iodine and iodide concentrations， the complex was composed of I82- units； Schneider et al[53] confirmed the conclusion of Gilbert and Marriott that the ratio of [I-] to [I2] bound in the helix is about 0.7 in this iodide range. Interestingly， this ratio is similar to the value of 0.75 characterizing the I113- nucleus reported by Thompson and Hamori[52].
　　A discovery regarding the formation of the amylose-iodine complex without KI is worth mentioning here. Molecular iodine （especially exists in amylose-iodine complex） was highly effective against breast lesions and other related diseases[54-56]. The addition of KI did promote the dissolution of I2， but the iodide ions had undesirable side effects on humans. For these reasons， investigation into the formation of amylose-iodine complexes without KI was carried out. Calabrese and Khan[57] found that amylose-iodine complexes could be formed without KI. The optimum temperature for the formation of the complex was 35℃. Ion-selective electrode experiments showed that both the amylose-iodine complex solution and iodine solution without KI contained only about 5% iodide ions （relative to the total iodine in solution）. They concluded that the iodide ions were not consumed in the reaction， and thus only the neutral iodine molecules were involved in the complex.
　　3 Interior structure of the complex
　　Amylose was established to have a left-handed helix organized structure on the basis of X-ray diffraction studies[58-62]. In the amylose-iodine complex， the iodine components appeared to be linearly arranged in the 5 ? wide inner cavity of the helix with an I-I distance of approximately 3.1 ?. Amari and Nakamura[63] found that the helical structure of the amylose-iodine complex appeared to be more rigid than that of other helical polymers such as poly（y-benzyl-L-glutamate）. However， the flexibility of the helix appeared to increase with the addition of urea. Although this structural model has been widely accepted， some questions remain regarding the polyiodide entities. 　　3.1 Small-angle-X-ray
　　Yuguchi et al[45] developed a creative thermo-reversible gelation system by combining xyloglucan and iodine solution， and small-angle-X-ray （SAXR） measurements were used to determine its structure. As Fig.5 showed， when the RG，c increased to 6.6 ?， a linear profile was observed for the xyloglucan gel with iodine. This size corresponded to the cross section of a rod-like aggregate consisting of two xyloglucan chains associated in parallel and incorporated iodine/iodide ions. The side chains of the xyloglucan constructed an inclusion space， and the polyiodides were determined to be involved in a uni-axial manner along the polysaccharide chains， as Fig. 6 showed.
　　3.2 Molecular lipophilicity profiles （MLPs） and analogous models
　　Another innovative method for the analysis of polysaccharide-iodine structures was proposed by Immel and Lichtenthaler[64]. The computed molecular lipophilicity profiles （MLPs）[65] were projected and color-coded based on the “solvent-accessible” contact surface， which was generated from the X-ray diffraction-derived structures of the polysaccharides. α-cyclodextrin-iodine was chosen as a model for the amylose-iodine complex because α-cyclodextrin is typically formed by enzymatic excision and recombination of a single turn of the VH-amylose helix[64]. It resembles a projection down the helix axis of amylose， and it may have similar ability to include iodine or iodide. Thus， α-cyclodextrin-iodine is a good model for amylose-iodine-iodide complexes[66-69]. Specifically， a bis-α-cyclodextrin lithium-pentaiodide octahydrate complex was used. This complex is a head-to-head dimer formed through an intense hydrogen bonding network between the secondary hydroxyl groups of the two α-cyclodextrin molecules， as shown in Fig.7. The generated MLPs of this complex and their color-coded projections onto the contacts are provided in Fig.8. The blue area corresponds to the hydrophilic 2-OH/3-OH-sides， which are responsible for the assembly of the two macrocycles； the two end sides bearing the primary hydroxyl groups are indicated by the yellow hydrophobic area. The MLPs carefully took the packing of these dimeric units in the crystal lattice into accouce， in which an uneven “nanotube” is formed by well-ordered infinite stacks of the dimer. The distribution of the hydrophilic and hydrophobic zones alternate on the outer surface of the model， but its inner channels successfully trap polyiodides or iodine and provide suitable hosting ability for polymeric supramolecular assemblies. 　　3.3 Zimm and Bragg matrix model
　　A simple model incorporating some commonly accepted characteristics of the amylose-iodine-triiodide complex was introduced by Cesáro et al[70]. The model is in good agreement with some experimental features and molecular parameters. The complex molecular model involved a generalization of the Zimm and Bragg matrix model for the helix-coil transition in polypeptides[71]. The authors assumed that the cooperative interactions between neighboring sites had a short effective range， and this assumption allowed a theoretical description of several features of the complex. Five statistical weights were defined for a generic site based on the formalism mentioned above， but only three of them were free parameters， which were obtained by fitting the experimental data of the bending isotherm for the complex. Based on the analysis of these free parameters， it was suggested that the random coie state was roughly equally preferred to the ordered helical one when the amount of I3- and I2 in solution was insufficient. In the framework of their model， the complex is initiated by binding an I3-， and then grows readily by the addition of I2 to an average length that is always smaller than or equal to 5. The length depends upon DP of the polysaccharides. In their hypothesis， the propagation of the complex occurs due to electrostatic ion-induced dipole interactions.
　　4 Surface morphology of the complex
　　AFM is an ideal instrument for imaging various molecules at the nanoscale， and has the great advantage of not requiring dyes or chemical or physical pretreatments[72]. An et al[73] placed a drop of gelatinized starch on a mica plate， and imaged it by AFM after drying. Subsequently， a drop of I2-KI was added on the surface of the sample， and the sample was imaged again after 30 s. The AFM images showed that branches cannot be clearly distinguished in the starch chains at high starch concentrations； instead， a membrane layer with a thickness of 0.5 nm was observed on the mica plate. The amylose-iodine precipitate formed a net-like structure with a height ranging from 0.6 nm to 5 nm， and consisted partially of thick， helical chains formed from intertwining thin chains. In addition， precipitates formed by the reaction of iodine with other starches， such as maize starch， potato starch， and sweet potato starch， were also detected by AFM. Both assemblies and net structures were observed on their surfaces.
　　5 Conclusions 　　Since the polysaccharide-iodine reaction was discovered， a great deal of important research on this topic has been reported. Based on the conclusions， the intrinsic reason for the formation of complexes in this reaction is the configuration of the polysaccharides. Most of the hemicellulose polysaccharides that can react with iodine in concentrated CaCl2 have a 1：3 ratio of 1→4 links to other links， whether the polysaccharide is linear or branched. Hemicellulosic polysaccharides will be precipitated by iodine in the presence of a sufficiently high concentration of CaCl2 （≥25%）. The results of Raman spectra analysis indicate that in both amylose-iodine and hemicellulosic polysaccharide-iodine complexes， the main polyiodides are I3- and I5-. On the basis of these inchoate theories， the structures of the polysaccharide-iodine complexes have been further revealed with the help of some newer methods including SAXR， MLPs， and the Zimm and Bragg matrix model. The side chains of xyloglucan form an inclusion space that absorbs the polyiodides between the polysaccharide chains. A proposed model also indicates that the bound species sequences are principally initiated by I3- and propagated by I2， with the length of the complex depending upon DP of the polysaccharides. Information regarding the outer surface of the amylose-iodine complexes has only been obtained by AFM so far， and both assembly and net structures were found. On the basis of the above results， the mechanism of the reaction between polysaccharides and iodine is very complicated. Further investigation of this reaction is required in order to achieve better utilization of the abundant polysaccharide resources， through polysaccharides fractionation techniques and the preparation of biocompatible materials and eco-friendly products.
　　Research into polysaccharide-iodine complexes began long ago. The conditions used in many of these past experiments do not meet modern standards， and the test methods are very simplistic when compared with those of the contemporary era. The deficiencies in previous works should be corrected or calibrated in order to better understand this topic. Additionally， the polysaccharide-iodine complexes can be seen as potential materials due to their special properties. The study of the formation mechanism of polysaccharide-iodine complexes is significant for fabricating and developing functional materials for use in the agricultural， pharmaceutical， food， and other industries. Therefore， more research into this topic is urgently needed. 　　Acknowledgments
　　This work was supported by the Fundamental Research Funds for the Central Universities （JC2015-03）， National Natural Science Foundation of China （31470417）， Beijing Municipal Natural Science Foundation （6182031）， and Author of National Excellent Doctoral Dissertations of China （201458）.
　　References
　　[1] Islam M R， Beg M D H， Jamari S S. Development of vegetable-oil-based polymers[J]. Journal of Applied Polymer Science， 2014， 131（18）， https：//onlinelibrary.wiley.com/doi/full/10.1002/app.40787.
　　[2] Sanchez-Vazquez S A， Hailes H C， Evans J R G. Hydrophobic polymers from food waste： resources and synthesis[J]. Polymer Reviews， 2013， 53（4）： 627-694.
　　[3] Song Y， Zheng Q. Eco materials based on food proteins and polysaccharides[J]. Polymer Reviews， 2014， 54（3）： 514-571.
　　[4] Fang D， Deng Z， Jung J， et al. Mushroom polysaccharides-incorporated cellulose nanofiber films with improved mechanical， moisture barrier， and antioxidant properties[J]. Journal of Applied Polymer Science， 2018， 135（15）， https：//onlinelibrary.wiley.com/doi/abs/10.1002/app.46166.
　　[5] Zhao X， Tong T， Li H L， et al. Characterization of hemicelluloses from Neolamarckia cadamba （Rubiaceae） during xylogenesis[J]. Carbohydrate Polymers， 2017， 156： 333-339.
　　[6] Kyle R A. Amyloidosis： a convoluted story[J]. British Journal of Haematology， 2001， 114（3）： 529-538.
　　[7] Binder J B， Blank J J， Cefali A V， et al. Synthesis of furfural from xylose and xylan[J]. Chem Sus Chem， 2010， 3（11）： 1268-1272.
　　[8] Li B G， Zhang L M. Research progress in inclusion complexation of amylose[J]. Progress in Chemistry， 2010， 22： 1161-1168.
　　[9] Peng P， Peng F， Bian J， et al. Isolation and structural characterization of hemicelluloses from the bamboo species Phyllostachys incarnata Wen[J]. Carbohydrate Polymers， 2011， 86（2）： 883-890.
　　[10] Landolt H. Ueber die zeitdauer der reaction zwischen Jods?ure und schwefliger S?ure[J]. European Journal of Inorganic Chemistry， 1886， 19（1）： 1317-1365.
　　[11] Swanson M A. Studies on the structure of polysaccharides IV. Relation of the iodine color to the structure[J]. Journal of Biological Chemistry， 1948， 172（2）： 825-837.
　　[12] Mould D L， Synge R L M. The electrophoretic mobility and fractionation of complexes of hydrolysis products of amylose with iodine and potassium iodide[J]. Biochemical Journal， 1954， 58（4）： 585.
　　[13] Mould D L. Potentiometric and spectrophotometric studies of complexes of hydrolysis products of amylose with iodine and potassium iodide[J]. Biochemical Journal， 1954， 58（4）： 593. 　　[14] Gaillard B D E. Separation of linear from branched polysaccharides by precipitation as iodine complexes[J]. Nature， 1961， 191（4795）： 1295-1296.
　　[15] Handa T， Yajima H. On the blue color of triiodide ions in starch and starch fractions. II. Characterization of the changes in absorption and circular dichroism spectra of triiodide ions in amylose with the DP[J]. Biopolymers， 1980， 19（4）： 723-740.
　　[16] Morak A J， Thompson N S. Factors influencing the formation of some polysaccharide-halogen complexes[J]. Nature， 1965， 205（4966）： 69.
　　[17] Gaillard B D E， Bailey R W. Reaction with iodine of polysaccharides dissolved in strong calcium chloride solution[J]. Nature， 1966， 212（5058）： 202.
　　[18] Peat S， Whelan W J， Roberts J G. The structure of lichenin[J]. Journal of the Chemical Society， 1957， 778： 3916-3924.
　　[19] Perlin A S， Suzuki S. The structure of lichenin： selective enzymolysis studies[J]. Canadian Journal of Chemistry， 1962， 40（1）： 50-56.
　　[20] Manners D J， Mitchell J P. Fine-structure of Rhodymenia palmate xylan[C]//Biochemical Journal. England， 1963， 89（3）： P92-&.
　　[21] Parrish F W， Perlin A S， Reese E T. Selective enzymolysis of poly-β-D-glucans， and the structure of the polymers[J]. Canadian Journal of Chemistry， 1960， 38（11）： 2094-2104.
　　[22] Chanda N B， Hirst E L， Manners D J. A comparison of isolichenin and lichenin from Iceland moss （Cetraria islandica）[J]. Journal of the Chemical Society （Resumed）， 1957： 1951-1958.
　　[23] Peat S， Whelan W J， Turvey J R， et al. The structure of isolichenin[J]. Journal of the Chemical Society （Resumed）， 1961： 623-629.
　　[24] Gaillard B D E， Thompson N S， Morak A J. The interaction of polysaccharides with iodine： Part I. Investigation of the general nature of the reaction[J]. Carbohydrate Research， 1969， 11（4）： 509-519.
　　[25] Barrett A J， Barrett K L， KHAN A. Effects of acetone， ethanol， isopropanol， and dimethyl sulfoxide on amylose-iodine complex[J]. Journal of Macromolecular Science， Part A： Pure and Applied Chemistry， 1998， 35（5）： 711-722.
　　[26] Yu X， Atalla R H. The complex of xylan and iodine： the induction and detection of nanoscale order[J]. Carbohydrate Research， 2005， 340（5）： 981-988.
　　[27] Williams R M， Atalla R H. Raman spectral investigation of the interactions between group II cations and ethylene glycol in aqueous solutions[J]. Journal of the Chemical Society， Perkin Transactions 2， 1975（11）： 1155-1161. 　　[28] Williams R M， Atalla R H. Interactions of Group II Cations and Borate Anions with Nonionic Saccharides： Studies on Model Poltols[M]. Hamilto： American Chemical Society， 1981.
　　[29] Yajima H， Nishimura T， Ishii T， et al. Effect of concentration of iodide on the bound species of I2/I3- in the amylose-iodine complex[J]. Carbohydrate Research， 1987， 163（2）： 155-167.
　　[30] Saenger W. The structure of the blue starch-iodine complex[J]. Naturwissenschaften， 1984， 71（1）： 31-36.
　　[31] Gilbert G A， Marriott J V R. Starch-iodine complexes. Part I [J]. Transactions of the Faraday Society， 1948， 44： 84-93.
　　[32] West C D. Structure-optical studies I. X- ray diffraction by addition compounds of halogens with hydrophilic organic polymers[J]. The Journal of Chemical Physics， 1947， 15（9）： 689-689.
　　[33] Kuge T， Ono S. Amylose-iodine complex. III. Potentiometric and spectrophotometric studies[J]. Bulletin of the Chemical Society of Japan， 1960， 33（9）： 1273-1278.
　　[34] Adkins G K， Greenwood C T. Studies on starches of high amylose-content. Part VIII. The effect of low temperature on the interaction of amylomaize starch with iodine： a unique characterization[J]. Carbohydrate Research， 1966， 3（2）： 152-156.
　　[35] Anderson D M W， Greenwood C T. An investigation of the polysaccharide content of oats， Avena sativa L[J]. Journal of the Science of Food and Agriculture， 1955， 6（10）： 587-592.
　　[36] Knutson C A， Cluskey J E， Dintzis F R. Properties of amylose-iodine complexes prepared in the presence of excess iodine[J]. Carbohydrate Research， 1982， 101（1）： 117-128.
　　[37] Deplano P， Devillanova F A， Ferraro J R， et al. FT-Raman study on charge-transfer polyiodide complexes and comparison with resonance Raman results[J]. Applied Spectroscopy， 1994， 48（10）： 1236-1241.
　　[38] Deplano P， Devillanova F A， Ferraro J R， et al. On the use of Raman spectroscopy in the characterization of iodine in charge-transfer complexes[J]. Applied Spectroscopy， 1992， 46（11）： 1625-1629.
　　[39] Mittag H， Stegemann H， Füllbier H， et al. Raman spectroscopic investigation of N-alkylurotropinium polyiodides[J]. Journal of Raman Spectroscopy， 1989， 20（4）： 251-255.
　　[40] Mizuno M， Tanaka J， Harada I. Electronic spectra and structures of polyiodide chain complexes[J]. The Journal of Physical Chemistry， 1981， 85（13）： 1789-1794.
　　[41] Nour E M， Chen L H， Laane J. Far-infrared and raman spectroscopic studies of polyiodides[J]. The Journal of Physical Chemistry， 1986， 90（13）： 2841-2846. 　　[42] Parrett F W， Taylor N J. Spectroscopic studies on some polyhalide ions[J]. Journal of Inorganic and Nuclear Chemistry， 1970， 32（7）： 2458-2461.
　　[43] Yu X， Houtman C， Atalla R H. The complex of amylose and iodine[J]. Carbohydrate Research， 1996， 292： 129-141.
　　[44] Yu X， Atalla R H. Production of cellulose II by Acetobacter xylinum in the presence of 2， 6-dichlorobenzonitrile[J]. International Journal of Biological Macromolecules， 1996， 19（2）： 145-146.
　　[45] Yuguchi Y， Fujiwara T， Miwa H， et al. Color formation and gelation of xyloglucan upon addition of iodine solutions[J]. Macromolecular Rapid Communications， 2005， 26（16）： 1315-1319.
　　[46] Handa T， Yajima H. On the blue color of triiodide ions in starch and starch fractions. II. Characterization of the changes in absorption and circular dichroism spectra of triiodide ions in amylose with the DP[J]. Biopolymers， 1980， 19（4）： 723-740.
　　[47] Kim H. Normal modes and structure of iodine in the starch–iodine complex[J]. Biopolymers， 1982， 21（11）： 2083-2096.
　　[48] Oishi Y， Yamamoto H， Miyasaka K. Structure of poly （vinyl alcohol）-iodine complex in water swollen film[J]. Polymer Journal， 1987， 19（11）： 1261.
　　[49] Moulay S. Molecular iodine/polymer complexes[J]. Journal of Polymer Engineering， 2013， 33（5）： 389-443.
　　[50] Landrum G A， Goldberg N， Hoffmann R. Bonding in the trihalides （X3-）， mixed trihalides （X2Y-） and hydrogen bihalides （X2H-）. The connection between hypervalent， electron-rich three-center， donor-acceptor and strong hydrogen bonding[J]. Journal of the Chemical Society， Dalton Transactions， 1997 （19）： 3605-3613.
　　[51] Svensson P H， Kloo L. Synthesis， structure， and bonding in polyiodide and metal iodide-iodine systems[J]. Chemical Reviews， 2003， 103（5）： 1649-1684.
　　[52] Thompson J C， Hamori E. Kinetic investigation of the amylose-iodine reaction[J]. The Journal of Physical Chemistry， 1971， 75（2）： 272-280.
　　[53] Schneider F W， Cronan C L， Podder S K. Cooperative binding to a one-dimensional lattice. The amylose-iodine-iodide complex[J]. The Journal of Physical Chemistry， 1968， 72（13）： 4563-4568.
　　[54] Ghent W R， Eskin B A. Elemental iodine supplementation in clinical breast dysplasia[C]//Proceedings of the American Association for Cancer Research. PUBLIC LEDGER BLDG， SUITE 816， 150 S. Philadelphia， 1986： 189-189.
　　[55] Ghent W R， Eskin B A， Low D A， et al. Iodine replacement in fibrocystic disease of the breast[J]. Canadian Journal of Surgery， 1993， 36（5）： 453-460. 　　[56] Eskin B A， Grotkowski C E， Connolly C P， et al. Different tissue responses for iodine and iodide in rat thyroid and mammary glands[J]. Biological Trace Element Research， 1995， 49（1）： 9.
　　[57] Calabrese V T， Khan A. Amylose-iodine complex formation without KI： Evidence for absence of iodide ions within the complex[J]. Journal of Polymer Science Part A： Polymer Chemistry， 1999， 37（15）： 2711-2717.
　　[58] Rundle R E. The configuration of starch in the Starchiodine complex. V. fourier projections from X-ray diagrams[J]. Journal of the American Chemical Society， 1947， 69（7）： 1769-1772.
　　[59] Rundle R E， Baldwin R R. The configuration of starch and the starch-iodine complex. I. The dichroism of flow of starch-iodine solutions[J]. Journal of the American Chemical Society， 1943， 65（4）： 554-558.
　　[60] Rundle R E， Edwards F C. The configuration of starch in the starch-iodine complex. IV. An X-ray diffraction investigation of butanol-precipitated amylose[J]. Journal of the American Chemical Society， 1943， 65（11）： 2200-2203.
　　[61] Rundle R E， Foster J F， Baldwin R R. On the nature of the starch-iodine complex[J]. Journal of the American Chemical Society， 1944， 66（12）： 2116-2120.
　　[62] Rundle R E， French D. The Configuration of starch and the starch-iodine complex. II. Optical properties of crystalline starch fractions[J]. Journal of the American Chemical Society， 1943， 65（4）： 558-561.
　　[63] Amari T， Nakamura M. Viscoelastic properties of aqueous solutions of amylose–iodine complex at ultrasonic frequencies[J]. Journal of Applied Polymer Science， 1976， 20（8）： 2031-2043.
　　[64] Immel S， Lichtenthaler F W. The hydrophobic topographies of amylose and its blue iodine[J]. Flexible and Rigid Non-glucose Cyclooligosaccharides： Synthesis， Structure， and Properties， 2000， 52（1）： 27.
　　[65] Heiden W， Moeckel G， Brickmann J. A new approach to analysis and display of local lipophilicity/hydrophilicity mapped on molecular surfaces[J]. Journal of Computer-aided Molecular Design， 1993， 7（5）： 503-514.
　　[66] Betzel C， Hingerty B， Noltemeyer M， et al. （β-Cyclodextrin）2·KI7·9H2O. Spatial fitting of a polyiodide chain to a given matrix[J]. Journal of Inclusion Phenomena， 1983， 1（2）： 181-191.
　　[67] Cramer F， Bergmann U， Manor P C， et al. Topographie der cyclodextrin-einschlu?verbindungen， IX lineare polyjodidketten in kanalf?rmiger cyclodextrinmatrix； kristallographische daten und vergleiche mit dem blauen st?rke-polyjodid-komplex[J]. European Journal of Organic Chemistry， 1976， 1976（7/8）： 1169-1179. 　　[68] McMullan R K， Saenger W， Fayos J， et al. Topography of cyclodextrin inclusion complexes： part I. classification of crystallographic data of α-cyclodextrin inclusion complexes[J]. Carbohydrate Research， 1973， 31（1）： 37-46.
　　[69] Noltemeyer M， Saenger W. Topography of cyclodextrin inclusion complexes. 12. Structural chemistry of linear α-cyclodextrin-polyiodide complexes. X-ray crystal structures of （α-cyclodextrin）2·LiI3·I2·8H2O and （α-cyclodextrin）2·Cd0.5·I5·27H2O. Models for the blue amylose-iodine complex[J]. Journal of the American Chemical Society， 1980， 102（8）： 2710-2722.
　　[70] Cesaro A， Benegas J C， Ripoll D R. Molecular model of the cooperative amylose-iodine-triiodide complex[J]. The Journal of Physical Chemistry， 1986， 90（12）： 2787-2791.
　　[71] Zimm B H， Bragg J K. Theory of the phase transition between helix and random coil in polypeptide chains[J]. The Journal of Chemical Physics， 1959， 31（2）： 526-535.
　　[72] Wang F F， Hang F. The action of trichoderma reesei cellulases on crystalline cellulose with high speed and high resolution atomic force microscopy[J]. Chemistry and Industry of Forest Products， 2014， 34（3）： 130-134.
　　[73] An H J， Liu Z D， Wang H B， et al. Study on the inclusion of starch and iodine by atomic force microscopy[J]. Journal of Henan University Technology， 2006， 27（5）： 13-17.