Crystals in Antheraea assamensis silkworm cocoon

Materials and Design 88 (2015) 236–244

Contents lists available at ScienceDirect

Materials and Design journal homepage: www.elsevier.com/locate/jmad

Crystals in Antheraea assamensis silkworm cocoon: Their removal, recovery and roles Jasjeet Kaur a, Rangam Rajkhowa a, Takuya Tsuzuki b, Xungai Wang a,c,⁎ a b c

Australian Future Fibres Research and Innovation Centre, Institute for Frontier Materials, Deakin University, VIC 3217, Australia Research School of Engineering, College of Engineering and Computer Science, Australian National University, Canberra, ACT 0200, Australia School of Textile Science and Engineering, Wuhan Textile University, Wuhan 430073, China

a r t i c l e

i n f o

Article history: Received 16 December 2014 Received in revised form 12 August 2015 Accepted 30 August 2015 Available online 4 September 2015 Keywords: Antheraea assamensis Demineralisation Crystals Protection Semi-domestic

a b s t r a c t Brick shaped mineral deposits or crystals are found in the shell of semi-domestic silk cocoon of Antheraea assamensis (A. assamensis). Effective removal and recovery of these crystals are important to understand their roles in the cocoon's protective function towards pupae. In this study, chemical and physical (ultrasonication) demineralisation methods were investigated for A. assamensis. It was found that the physical demineralisation method could effectively separate crystals without changing their shape and size and not effecting other components of the silk cocoon. The efficient recovery of the crystals, without any change in their chemical composition was confirmed based on FTIR, XRD and EDX techniques. Chemical demineralisation method was optimised and performed under milder conditions than reported in the past. It helped reeling of silk without much loss of strength or natural colour of silk fibre. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction A silk cocoon is a protein-fibre composite shell formed by a twin filament (bave) of high molecular weight fibroin proteins surrounded by glue like proteins [1]. The shell is a multilayered structure made of overlaying bave [2]. Silk sericin provides the inter fibre and inter layer adhesion to form the composite cocoon. Cocoon-forming silkworms can be divided into three categories: domestic, semi-domestic and wild. This classification is based upon whether the silkworms are reared under controlled or uncontrolled conditions [3]. Generally, domestic silkworm cocoons have only the fibre and the sericin as components, whilst the semi-domestic and wild cocoons also have the so-called crystals/mineral deposits embedded in their cocoon shells [4–7]. Cocoon formation of a wild silkworm was monitored using the cinefluoroscope television and it was seen that after 10–20 h of formation of outer layer by spinning larva, a dense substance exudes slowly from the larva's anus and impregnates the entire cocoon. The exudes were considered to contain crystals [8]. There is no clear understanding of the relationship between diet and crystal contents in silk cocoons though diet is likely to be the source of crystals. There are certain studies that show a link between the moth diet and the secretion on the cocoon [9,10,11]. Cocoon shells vary in their morphologies and tensile properties depending on the silk species [12]. This difference in their structure and properties is due to their evolution over millions of years by the process ⁎ Corresponding author at: Australian Future Fibres Research and Innovation Centre, Institute for Frontier Materials, Deakin University, VIC 3217, Australia.

http://dx.doi.org/10.1016/j.matdes.2015.08.148 0264-1275/© 2015 Elsevier Ltd. All rights reserved.

of natural selection in response to different threats [12,13]. It is expected that cocoon structures are related to the various protective functions needed for the pupae [14]. Various cocoon components are responsible for providing protection to pupae against radiation, humidity, pathogens and predators. The innermost layer of a silk cocoon has a denser microstructure and better mechanical properties that can increase the cocoon's ability to resist possible attacks from outside [15]. Crystals embedded in the cocoon shell are thought to be associated with regulating the gas diffusion and water vapour exchange to allow metabolic activity of pupae and to maintain physiological temperature by trapping the still air inside [16], while the dense structure of the cocoon provides protection from extreme external environment [12,13,17–19]. Since cocoons are believed to provide protection against predators and the crystals have exhibited high hardness [7], these crystals are also thought to be responsible for providing first point of defence to any predator attack. However, the role of crystals in cocoons is not yet well understood. Therefore, in order to study the specific role of crystals in pupae protection, it is important to examine cocoons with and without crystals. The isolation of crystals (demineralisation) should be carried out without affecting other components. In particular, sericin needs to be retained after demineralisation, which cannot be done using conventional cooking (partial degumming) methods as well as during demineralisation at pH 10 as reported earlier [5][20]. An ideal demineralisation process can also help to obtain pure sericin if demineralisation is followed by degumming. This is important because purified sericin has potential uses in various biomedical applications [21] and can be used also as an antioxidant [22,23] and antibacterial agent [24,25].

J. Kaur et al. / Materials and Design 88 (2015) 236–244

Antheraea assamensis (A. assamensis) silkmoths are normally fed on host plants Machilus bombycina, Litsaea polyantha in the outdoor environment and later transferred to the indoor space for spinning cocoons. Hence A. assamensis silk falls in the category of semi-domesticated silk [26]. This variety of silk contains large amount of crystals and therefore was selected in this study. A. assamensis silk is a highly priced fibre material for the textile industry [27]. The current price of 1000 A. assamensis cocoons is about 50 USD. In the production of yarn and fabric from semi-domestic and wild silk varieties such as A. assamensis and Antheraeamylitta, often degumming is not desired so that the natural colour and handle of the native silk thread can be retained. This is because degumming has an effect on the mechanical properties, handle and colour of the fibre [28]. In such a case, a method of mild demineralisation that can facilitate easy reeling of silkworm cocoons while keeping the inherent mechanical properties, handle and colour of silk fibre is an important requirement [5]. Currently commercial reeling of A. assamensis is still performed after cooking the cocoons at boil in alkali to remove part of silk gum and the crystals [29]. A better understanding of the crystals and an effective demineralisation process is vital for more efficient processing and better quality of A. assamensis raw silk yarn with silk gum and natural colour retained. This is also important for other types of silk where crystal deposition causes reeling problems. A study by Gheysens et al. [5] suggested reelability of demineralised cocoons at pH 10 which has provided inspiration to further investigate the demineralisation studies on A. assamensis to improve its reelability. We believe that pH 10 is a highly alkaline condition in which the silk gum, sericin is partially disintegrated and thus helped reeling. Our interest was to use a much mild condition to retain the sericin during the demineralisation process and perform reeling at the same time. Later the effect of demineralisation on the reeling of these cocoons was studied using commercial type A. assamensis reeling systems. Gheysens et al. [5] and Chen et al. [13] found that the crystals are present on the outer surface of the cocoons except for Antheraea roylei that had crystals on both the outer and inner surface of the cocoon. Calcium crystals have been shown to exist in A. assamensis by Freddi et al. [30]. In the past, demineralisation was performed on Gonometa postica cocoons using 1 M EDTA and concentrated NaOH at pH of 10 for 72 h [5]. 3 M HCl and sodium–EDTA solutions of 0.4 M were also used for A. mylitta[17]. We have used EDTA and sodium–EDTA solutions for demineralisation as reported for G. postica by Gheysens et al. [5] previously and optimised processing conditions for A. assamensis in such a way that a milder process can facilitate reeling as well as maintain fibre mechanical properties and natural colour. Up to date, only chemical demineralisation methods have been reported for silk cocoons in which crystals are likely to change in composition and it is also difficult to recover as they dissolve in the process. In the present study, demineralisation was carried out a simple method to isolate crystals without changing their chemical composition of cocoons. Since the crystals are known to be attached loosely to the silk fibre [13], a physical method was used to remove crystals for further analysing their morphological and chemical nature without changing their properties during demineralisation. Ultrasonication was used previously to extract nanofibres from reeled silk fibres [31]. Also ultrasonication has been used to remove silk sericin in combination with the conventional chemical methods [32]. In this study a mild ultrasonication treatment was used to remove the crystals from cocoons without causing fibrillation of silk. The effect of demineralisation on the mechanical strength of the fibre was studied. 2. Experimental 2.1. Materials and preparations Ethylenediaminetetraacetic acid (EDTA), EDTA disodium salt, sodium hydroxide, sodium carbonate, sodium hydrogen carbonate and calcium oxalate monohydrate were purchased from Sigma Aldrich

237

Australia. All reagents were used as received. A. assamensis cocoons were purchased from Fabric Plus Ltd. India. 2.2. Demineralisation methods 2.2.1. Chemical demineralisation Aqueous solutions of EDTA (pH 6–7) and EDTA disodium salt (pH 7– 8) were prepared separately by adding concentrated NaOH until a clear solution of a definite molarity was obtained. The solutions were made into various concentrations of 0.01 M, 0.03 M, 0.06 M, 0.12 M and 0.25 M from the stock solution of 0.5 M. Temperatures used were ambient, 40 °C and 60 °C. The demineralisation was performed in a laboratory dyeing machine (Ahiba, Japan). 2.2.2. Physical demineralisation To avoid the use of hazardous chemicals, ultrasonication was employed as a physical technique to isolate crystals from the cocoons. Ultrasonic processor UlP1000hd (Hielscher Technology, Germany) with a maximum power input of 1000 W, operational frequency 20 kHz, sonotrode amplitude of up to 170 μm, variable amplitudes (30%, 50% and 100%), time intervals of 15 min, 30 min, 45 min and 1 h and energy from 388 to 2592 kJ were used. The pulse rate was 2 s on and 1 s off. Cocoon-to-liquor ratio and treatment time were the same as the chemical method with EDTA solutions. After ultrasonication, the cocoon was removed and the remaining deionised water was dried at 60 °C to obtain the crystals. 2.3. Degumming method A. assamensis silk cocoon was degummed using sodium carbonate (10 g/500 ml) plus sodium bicarbonate (10 g/500 ml) in 1:1 ratio, making 2 % base at 98 °C for 1 h in laboratory dyeing machine (Ahiba). 2.4. Morphological and structural characterisation The structure of A. assamensis cocoons was studied layer-by-layer using a scanning electron microscope (Zeiss Supra 55VP, Germany) at 2–3 kV accelerated voltage and 5–6 mm working distance. The cocoons were washed with deionised water and the layers were carefully peeled by hand. The cocoon morphology before and after chemical and physical demineralisation was observed by scanning electron microscopy (SEM). SEM images were also taken before and after demineralisation performed by method suggested by Gheysens et al. [5] at pH 10 and our optimised method at pH 6–7 to observe the effect of pH on the cocoon elements (crystals and sericin). Fourier transform infrared spectra (FTIR) of the desiccated demineralised cocoons after physical demineralisation were obtained using attenuated total reflection (ATR) mode in a Bruker VERTEX 70 spectrometer (Etllingen, Germany) and associated software OPUS 5.5 with a resolution of 4 cm−1 and 128 scans per sample. The crystal phase of the obtained crystals was analysed by X-ray diffraction (XRD) study, using Panalytical X'Pert PRO diffractometer (Netherlands) with Cu-Kα radiation. Voltage and current of the X-ray source were 40 kV and 30 mA respectively. The sample powder was placed with the help of vacuum grease on a glass slide and then was fixed inside the X-ray diffractometer. X-ray spectra were recorded at a scan step size of 0.02 and a time per step of 4 s. The thermal properties of the crystals were studied by differential scanning calorimetery (DSC) and thermal gravimetric analysis (TGA). DSC measurements were made using a TA DSC Q200 (USA) with a nitrogen gas flow rate of 25 ml/min and a heating rate of 10 °C/min. 5–10 mg of the sample was used in the aluminum pan. The thermal behaviour of three different layers of A. assamensis cocoon was evaluated from room temperature to 110 °C and then the sample was reheated from 50 °C to 400 °C. The first cycle from room temperature to 110 °C helped to remove water molecules [30]. TGA was performed using a Netzsch STA 407 DSC (Germany).

238


13–15 mg of the sample was heated up to 800 °C at a rate of 10 °C/min under a nitrogen atmosphere. To analyse the chemical composition of the crystals, energy dispersive spectroscopy (EDX) was performed using a Supra 55VP (Zeiss, Germany) SEM equipped with an EDX detector. The samples were carbon coated prior to measurements. The measurements were done at 20 kV. The FTIR, EDX, XRD, DSC and TGA data of the crystals were compared with those of commercial calcium oxalate monohydrate powder. Tensile properties of raw, demineralised and degummed fibres were measured using a Favimat Airobot 2 (Textechno, Monchengladbach, Germany) fitted with a 210 cN load cell. In this instrument, the test specimens were picked up and loaded by a robot which reduced the fibre handling and ensured minimum variations in mounting tensions. Fibre specimens were loaded into a magazine using 100 mg pretensions. Pre-tension applied by the instrument was 0.05 cN/dtex and specimens were strained at 60% min−1. The gauge length was 25 mm. The average cross sectional area of the fibres was measured to calculate the fibre stress. To obtain the silk fibre cross sections, 20–25 fibres were embedded in a resin (Spurr replacement kit from TAAB Laboratories) and then cured at 60 °C for 48 h. Later these fibre-embedded resin blocks were cut perpendicular to the fibre axis using ultramicrotome EMUC6 (Leica, Germany). The fibre-embedded resin blocks were gold coated and viewed under SEM at 3 kV accelerated voltage and 6–7 mm working distance. Imaging of the silk fibres was done and the cross sectional area of 20 fibres was measured using ImageJ software. A. assamensis cocoons were reeled using an commercial motor operated reeling machine after demineralisation at a linear speed of 12 m/ min as well as using a custom built reeling machine. 3. Results and discussion 3.1. Distribution of cocoon crystals to prevent predator attack The cocoon layers of A. assamensis silk were easier to peel than other Antheraea varieties (e.g. A. mylitta, Antheraea pernyi), because of relatively weaker interlayer adhesion [13]. It was found that the peeling strength of A. assamensis is about half compared to other Antheraea

cocoons [7]. This indicated that the crystals were not the major contributor to the interlayer adhesion in A. assamensis silk cocoons considering significant deposition of crystals in this variety of silk compared to others. The interlayer adhesion therefore should come from other factors such as silk sericin. It was not easy to prepare multiple layers from cocoons due to interlayer adhesion. There was also possibility of removing some of crystals during peeling a number of very thin layers. Hence we restricted this study to three layers. Fig. 1 shows SEM images of the outer, middle and inner layers of an A. assamensis cocoon. Each cocoon layer consists of fibroin, sericin and crystals. The outermost layer was found to have a very rough, coarse surface due to the presence of large amount of sericin (Fig. 1(a, a’)). The density of fibre packing in the outer layer is low and hence large gaps can be noticed between the silk strands (Fig. 1(a)). The middle layer (Fig. 1(b’) has less sericin than the outermost layer and the gaps between silk strands are reduced. The innermost layer shows striated characteristics on the fibre surface, which indicates that the sericin coating is not thick enough to cover the fibre (Fig. 1(c’)). The innermost layer has a tight fibre network with very small gaps between the strands (Fig. 1 (c)). The results are in agreement with the previous report that the sericin content reduces along the length of the filament in mulberry and non-mulberry cocoons [33,34]. It is evident in Fig. 1(c) that the density of crystal deposition is reduced in the order of outer N middle N inner layers (panels a”, b”, c”). There are three important strategies that can be used by the predator to break the cocoon barrier or harm the pupae: 1) peeling the silk cocoon shells, 2) punching through, and 3) applying an impact force. In a study by Jin et al. it was shown that the crystals have hardness of around 2 GPa and these remain intact after compression tests. The cocoons with crystals showed higher compressive modulus than the one without crystals. Jin et al. also showed that the delamination resistance of the outer layers was higher than the inner layers in wild silk cocoons [7]. This satisfies the requirement for the wild cocoon to protect the pupa against physical attack from natural predators. Chen et al. study also supports that the graded layer structure improves the impact resistance of the cocoon [4]. The decrease of crystals from outer to inner cocoon surface may act as a protection barrier to the pupa against any external attacks. It is likely that even if the crystals in the outer layers are removed

Fig. 1. SEM images of A. assamensis cocoon layers, outer layer (a, a’, a”), middle layer (b, b’, b”) and inner layer (c, c’, c”).


due to cocoon being exposed to water, wind etc. (environmental) and mechanical (external attack) action, any possible deterrent will still be present inside. Deterrent action of crystals may be expected as oxalate ingestion can result in corrosion of the mouth and gastrointestinal tract in animals [35] though studies are warranted to test the hypothesis. Calcium based crystals (known as raphides) are also secreted by specialised plant cells called idioblasts and such raphides save plants from micro predators. These crystals exhibit variable morphology depending on various plant species. The cells and genetics of particular species forming these crystals control their morphology [36]. Since these crystals are found in the semi-domestic and wild silk cocoon varieties which are at a greater risk of predator attack, their presence may act as a protection barrier to the pupa against any external attacks [35]. 3.2. Optimisation of chemical method using EDTA to remove crystals in A. assamensis Fig. 2 shows SEM images of cocoons treated with EDTA solutions of different concentrations at 60 °C. The untreated (raw) A. assamensis cocoon has brick-like crystals embedded in the cocoon shell. When treated with a 0.01 M EDTA solution, the majority of crystals still remained in the shell. With a 0.03 M EDTA, the shape of the crystals became round and the number of remaining crystals decreased, indicating that many of the crystals were dissolved in the solution. At 0.06 M concentration, all the crystals were completely removed. The images suggest that 0.06 M EDTA concentration is sufficiently high for removing the crystals. We have considered this as an optimised EDTA concentration for A. assamensis silk cocoon for treatment at 60 °C, pH 6–7. When the temperature was lowered from 60 °C to 40 °C to room temperature (R.T), complete removal of the crystals was not possible at 0.06 M concentration (pH 6–7) in 1 h. The cocoons treated with EDTA disodium salt solutions showed similar trends. However, EDTA disodium salt was slightly less effective than EDTA. The crystals were completely removed using EDTA disodium salt solution at 0.12 M concentration (pH 7–8) and 1 h reaction time (data not shown). Fig. 2 also shows the stages of crystal breakage during chemical demineralisation process. Using different concentrations of EDTA, the mechanism of crystal removal was also studied. When an EDTA concentration of 0.01 M was used, the crystals started to disintegrate (Fig. 2d). At a slightly higher concentration of 0.03 M (Fig. 2f), the crystals were fragmented and when the concentration was further increased to

239

0.06 M (Fig. 2h), the crystals were completely removed, leaving brick shaped marks on the sericin layer over the silk fibres. 3.3. Optimisation of physical method of ultrasonication for effective crystal removal Demineralisation was also performed using ultrasonic energy. Fig. 3(a–d) shows SEM images of the cocoon subjected to ultrasonication with various input energy. The extent of demineralisation increased with the increase in ultrasonic energy. 2592 kJ was sufficient to remove all crystals (Fig. 3(d)). The crystals were collected after separation. The crystals isolated from silk cocoons after ultrasonication showed bricklike shapes and varied in their size (Fig. 3(e)). It was observed that, even after high energy ultrasonication, the brick like shape of the isolated crystals was retained. Demineralisation was performed at amplitudes of 30%, 50% and 100% and for treatment time of 15 min, 30 min, 45 min and 60 min at 1000 W power and operational frequency 20 kHz. We concluded that demineralisation was complete at 100% amplitude for 1 h. All the results are not shown here. At high input energy of 2592 kJ, small particles of sericin protein were also formed. However, there was no evidence of the splitting of fibres during the ultrasonication process (Fig. 3(d)). Change in surface morphology of the same treatment on degummed fibres was also assessed. After degumming with base, A. assamensis silk fibres were subjected to the same ultrasonication treatment as that of cocoons and no splitting or fibrillation was observed (Supplementary Fig. 2). In order to check if ultrasonication damaged fibres in the presence of alkali, the ultrasonication was done on these fibres in a bath containing base (2%) instead of water. Significant fibrillation was visible on the fibre surface as seen in the SEM images (Supplementary Fig. 2). The results suggest that ultrasonic treatment required for demineralisation of A. assamensis does not damage fibres easily but may harm silk fibres if sonication is performed on relatively weak degummed fibres and when the pH of the sonication bath is high. As shown previously by Hong ping et al. ultrasonication was used to obtain nanofibres from Bombyx mori silk fibres [31]. We confirmed this by ultrasonicating degummed B. mori fibres in water for 1 h. Some fibrillation was observed in B. mori fibres (Supplementary Fig. 3) in 1 h when no fibrillation was seen in A. assamensis fibre (Supplementary Fig. 2). When the treatment time was increased the fibrillation in B. mori was significant (Supplementary Fig. 3). This shows that the effect of ultrasonication is different on different silk varieties.

Fig. 2. SEM images of cocoons subjected to chemical demineralisation using EDTA solutions of different concentrations and various stages of crystal breakage.

240


Fig. 3. SEM images of cocoons subjected to ultrasonication at various input energy levels (a–d) and obtained ultrasonicated crystals (e).

Although ultrasonication is a chemical free technique and scalable for industrial scale operations, it would be interesting to calculate the energy saving made using this physical method relative to heating water for treating cocoons. Fig. 4(a) shows FTIR spectra of A. assamensis cocoons ultrasonicated at various energy levels from 388 kJ to 2592 kJ. All the spectra showed peaks associated with silk proteins. In the spectrum of raw cocoon, additional peaks at 1317 cm−1 and 781 cm−1 were present. Their intensities reduced as the energy supplied for ultrasonication increased and, at 2592 kJ, these peaks disappeared. Both of the additional peaks correspond to calcium oxalate monohydrate [4]. The results indicated that all the crystals were removed at 2592 kJ of energy as suggested by SEM images in Fig. 3. Fig. 4(b) shows the EDX spectra of raw and demineralised cocoons. From the figure it is seen that Ca peaks are located at 3.691(Kα) and 4.012(Kβ). It was found that the spectra of raw A. assamensis cocoon had prominent calcium peaks, while these peaks were absent in the spectrum of cocoon demineralised with physical method (ultrasonication).

identical to those of commercial calcium oxalate monohydrate. The results indicate that the crystals in A. assamensis cocoons are calcium oxalate monohydrate.

3.4. Confirmation of chemical composition of recovered cocoon crystals after ultrasonication There are reports that the crystals in silk cocoons are made of calcium oxalate [30]. The calcium oxalate monohydrate in malpighian tubes of tent caterpillar was first reported by Takahashi et al. [9]. We have shown earlier by comparing the FTIR of commercial calcium oxalate monohydrate and the A. assamensis cocoon shell that the crystals are calcium oxalate monohydrate crystals [7]. Freddi et al. [30] used commercial calcium oxalate in XRD experiments to confirm that the crystals in silk were of similar nature. As physical separation process maintains crystal size and shape and there is no chemical agents used for reactions, analysis of separated crystals can further validate the composition of crystals. We have extended the investigation on cocoon crystals using other experimental tools such as EDX, DSC/TGA and FTIR by comparing commercial calcium oxalate monohydrate with the separated crystals obtained from A. assamensis silk cocoons after physical treatment (ultrasonication) to confirm the crystal composition. 3.4.1. FTIR Fig. 5(a) shows the FTIR spectra of commercial calcium oxalate monohydrate and silk cocoon crystals obtained after physical demineralisation. In the spectrum of commercial calcium oxalate monohydrate, the peaks were found at 781 cm−1 and 1317 cm−1. The peak at 781 cm− 1 represents O–C–O out-of-phase bonding and 1317 cm−1 represents asymmetric C–O stretching vibrations [37]. It is evident that the spectrum of isolated crystals has absorption peaks

Fig. 4. (a) FTIR spectra of A. assamensis cocoons subjected to ultrasonication at various energy levels (b) EDX spectra of raw A. assamensis and A. assamensis demineralised with physical method (ultrasonication).


241

Fig. 5. a) FTIR spectra of the crystals isolated by physical demineralisation and commercial calcium oxalate monohydrate, (b) XRD spectra of comparison of commercial calcium oxalate monohydrate crystals and cocoon crystals isolated by ultrasonication (c) EDX spectra of commercial crystals and ultrasonicated crystals.

3.4.2. XRD The XRD patterns of commercial calcium oxalate monohydrate powder and the crystals obtained after ultrasonication of A. assamensis cocoons are shown in Fig. 5(b). The peaks at 14.95°(101), 24.39°(020), 30.12°(202), 38.13°(130), 39.909° and 43.630° are indexed based on

JCPDS: 00-003-0087 pattern that corresponded to commercial calcium oxalate monohydrate powder. Both crystals show similar XRD patterns, which supports the results of FTIR study. There are additional peaks present at 28.37° and 40.48° in the crystals isolated from A. assamensis cocoon. It was found that these peaks were due

Fig. 6. (a) TGA curves of commercial calcium oxalate monohydrate crystals and cocoon crystals isolated by ultrasonication. (b) Differential scanning calorimetry of isolated crystals from cocoons (red dotted line indicates the on-set temperature of ~150 °C), and (c) commercial calcium oxalate monohydrate crystals different layers of A. assamensis cocoon.

242


Fig. 7. SEM images of raw A. assamensis silk filament (a,a'), A. assamensis demineralised at pH 10 using Gheysens et al. method (b,b') and A. assamensis demineralised at pH 6–7 (c,c').

to the contamination stemming from the titanium alloy probe used for ultrasonication. 3.4.3. EDX Fig. 5(c) shows the EDX spectra of crystals obtained after ultrasonication and commercial calcium oxalate monohydrate crystals. As shown in Fig. 5(c), Ca peaks are located at 3.691(Kα) and 4.012(Kβ). It was found that the spectra of commercial calcium oxalate monohydrate powder and ultrasonicated crystals had prominent calcium peaks. The results further confirmed that the crystals in A. assamensis cocoons contained calcium. 3.4.4. TGA/DSC The TGA curves of commercial calcium oxalate monohydrate powder and the crystals obtained after ultrasonication of A. assamensis cocoons are presented in Fig. 6(a). In the spectrum of commercial calcium oxalate monohydrate, the weight loss starting at ~150 °C indicates dehydration of calcium oxalate monohydrate crystals. The second weight loss around 500 °C is associated with the decomposition of anhydrous calcium oxalate into calcium carbonate and carbon monoxide [38]. The decomposition above 700 °C is due to the conversion of calcium carbonate into calcium oxide and carbon dioxide [38]. The TGA curve of the crystals obtained after ultrasonication of A. assamensis cocoons follows the same trend as the TGA curve of commercial calcium oxalate monohydrate. However, the weight loss at each step was not as sharp as that of commercial calcium oxalate monohydrate, indicative of perhaps the different impurities present in the cocoon crystals. Fig. 6(b) shows the DSC curves of commercial calcium oxalate monohydrate powder and the crystals obtained after ultrasonication

of A. assamensis cocoons. As mentioned earlier, in the DSC curve of commercial calcium oxalate monohydrate, the endothermic peak with an on-set temperature of ~150 °C is associated with dehydration. The on-set temperature of the dehydration of cocoon crystals is slightly lower than that of commercial powder. The difference in the on-set temperature and endothermal energy (i.e. peak area) between the two samples may be due to the difference in purity. There may be a trace of sericin or other materials present with the crystals obtained by the process of ultrasonication. Fig. 6(c) shows the DSC curves of inner, middle and outer layers of an A. assamensis cocoon shell. The endothermic peak between 150 °C–200 °C, which is associated with calcium oxalate monohydrate crystals, decreased from outer layer to inner layer. The peak for crystal was the strongest in the outer layer and it decreased towards the inner layers.

Fig. 8. Digital images of A. assamensis cocoons during reeling and towards the end of reeling in a commercial reeling.


243

Table 1 Maximum stress, strain and toughness values for raw, demineralised A. assamensis fibre with chemical method and physical method, degummed silk fibre; data is presented as mean ± standard deviations, n = 50. Serial no.

Sample

Maximum stress (MPa)

Strain (%) at max stress

Toughness (kJ/kg)

1. 2. 3. 4. 5.

Raw A. assamensis Demineralised A. assamensis with chemical method, pH 6–7, 1 h Demineralised using ultrasonication, 1 h Degummedusing base, 98 °C, 1 h Base degummed fibres ultrasonicated in base, 1 h

600 ± 90 565 ± 100 490 ± 100 480 ± 100 380 ± 80

41 ± 6 37 ± 7 50 ± 8 35 ± 4 28 ± 8

175 ± 30 165 ± 30 165 ± 65 110 ± 40 90 ± 35

We have also observed during peeling the cocoons shells that some crystal dust comes out even from the middle layers which further suggest the presence of crystals in the middle layers. This is consistent with the results of SEM and FTIR studies, indicating decrease of crystals from outer to inner layers. Two minor endotherms at 230 °C and 299 °C (shoulder form) represent the change in molecular motions of the silkfibroin chains in the amorphous and crystalline regions respectively, and the prominent endotherm at 380 °C represents thermal decomposition of silk protein [30]. 3.5. Reeling of A. assamensis and mechanical strength of silk fibres demineralised with chemical and physical method Reeling of filaments from cocoons of Antheraea species normally requires boiling the cocoons in sodium carbonate and sodium silicates [39]. In this process, silk may be partially hydrolyzed by alkali at high temperature [20,40]. Such a treatment not only changes the natural colour of silk thread but may influence their mechanical properties [33,40]. Demineralisation improves the reeling performance of silk of Antheraea species without the need for boiling using strong alkali and therefore an effective demineralisation is ideal for processing such expensive and luxury silk materials [39,41,42]. It has been shown earlier by Gheysens et al. [5] that demineralisation helped in easy reeling of the cocoons. Therefore reeling was tried on the commercial silk variety, A. assamensis silk cocoon after the optimised chemical demineralisation process. We believe that for reeling, certain amount of degumming is required and reeling in the work by Gheysens et al. [5] must have been assisted by partial degumming due to cocoon treatment at pH 10. Therefore we compared demineralisation of A. assamensis using Gheysens et al. [5] method and our optimised method. Fig. 7 shows SEM images of raw A. assamensis silk filament, A. assamensis demineralised at pH 10 using Gheysens et al. [5] method and A. assamensis demineralised at pH 6–7. We found that using pH 10 not only removed the crystals but also partially removed

sericin that perhaps assisted in reeling of the silk cocoons (Fig. 7). We performed reeling of A. assamensis cocoons using commercial reeling machines at ~12 m/min after demineralising by our optimised method (pH 6–8). It was seen that though the crystals were removed at 60 °C for 1 h at pH 6–8, however to facilitate reeling of demineralised cocoons using commercial reeling machines, a higher temperature (90 °C) for demineralisation was required to soften the gum. Therefore reeling was achieved by using the demineralisation solution for 45–60 min at pH 8, 90 °C (Fig. 8). It is clear from the digital images (Fig. 8) that the cocoons could be reeled effectively and completely. Further studies are warranted to test whether demineralised cocoons by ultra-sonication either alone or in combination with chemical methods can be used to facilitate reeling. Removal of crystals without reducing the mechanical properties of the fibre is also important. In this study, it was demonstrated that demineralisation was possible using a very low concentration (0.06 M) of EDTA solutions or without any chemicals at all. The influence of chemical demineralisation and physical demineralisation on the mechanical properties of silk fibres was investigated to determine if the process anyway reduced fibre mechanical properties. Table 1 shows the values for breaking stress, strain and toughness for raw and demineralised A. assamensis fibres with an optimised chemical and mild physical method compared to conventionally degummed fibre and fibre ultrasonicated in the presence of base. Fig. 9 shows representative stress–strain curves for raw, demineralised (chemically) and degummed A. assamensis. There was no significant difference in mechanical properties after chemical demineralisation (p N 0.05, Student's t-test). Thus, at the concentration levels used in this study, chemical demineralisation did not cause significant deterioration in the mechanical properties of silk fibres. While the physical demineralisation reduced the silk fibre strength significantly (p value b 0.0001), but did not cause any splitting of the fibres (Fig. 3). The strength of the ultrasonicated fibres was found to be slightly higher than the completely degummed fibres using base (2%) at 98 °C, but the difference was not statistically significant (p value N 0.05).When the fibres degummed using base (2%) were ultrasonicated the mechanical strength further reduced significantly from 490 MPa to 380 MPa, p value b 0.0001 (Table 1).

4. Conclusion

Fig. 9. Representative stress–strain curves for raw, demineralised (physically and chemically) and degummed A. assamensis.

This study investigated the crystals in A. assamensis silk cocoons, in particular, their distribution patterns in cocoons and effective methods of their removal from cocoons. Chemical and physical methods were used for crystal removal. It was found that both these methods have their own advantages. Ultrasonication can effectively remove crystals without changing the other components of cocoons and hence can be used to study their protective roles of different components. The shape of the crystals, size and chemical composition are retained after physical method. The work also demonstrated that damage to fibres from ultrasonication on the silk variety and A. assamensis cocoons and fibres can be processed without significant damage while the same conditions can damage fibres of B. mori. A chemical demineralisation method enables easy reeling without causing significant loss of strength and natural colour to the silk fibre due to partial degumming associated with the

244


process. We demonstrated that milder concentration and pH can be used for chemical demineralisation than reported earlier. Acknowledgements We acknowledge the support from the Australian Research Council (discovery project Grant DP120100139) for this work. We would also like to thank Mr. Patrick Brislane for his help with the ultrasonication work. Appendix A. Supplementary data Supplementary information explains the effect of demineralisation on the colour bleeding of the cocoons and excessively harsh treatment by ultrasonication on fibre properties and morphology. This material is available at free of charge via the Internet at http://pubs.acs.org. Supplementary data to this article can be found online at http://dx.doi.org/ 10.1016/j.matdes.2015.08.148. References [1] R. Rajkhowa, B. Levin, S.L. Redmond, L.H. Li, L. Wang, J.R. Kanwar, M.D. Atlas, X. Wang, Structure and properties of biomedical films prepared from aqueous and acidic silk fibroin solutions, J. Biomed. Mater. Res. A 97 (2011) 37–45, http://dx. doi.org/10.1002/jbm.a.33021. [2] A. Teshome, F. Vollrath, S.K. Raina, J.M. Kabaru, J. Onyari, Study on the microstructure of African wild silk cocoon shells and fibres, Int. J. Biol. Macromol. 50 (2012) 63–68 (doi:). [3] L. Varadarajan, Silk in northeastern and eastern India: the indigenous tradition, Mod. Asian Stud. 22 (3) (1988) 561–570. [4] F. Chen, T. Hesselberg, D. Porter, F. Vollrath, The impact behaviour of silk cocoons, J. Exp. Biol. 216 (14) (2013) 2648–2657. [5] T. Gheysens, A. Collins, S.K. Raina, F. Vollrath, D.P. Knight, Demineralisation enables reeling of wild silkmoth cocoons, Biomacromolecules 12 (6) (2011) 2257–2266. [6] D. Devi, N. Sen Sarma, B. Talukdar, P. Chetri, K.C. Baruah, N.N. Dass, Study of the structure of degummed Antheraea assamensis (muga) silk fibre, J. Text. Inst. 102 (6) (2011) 527–533. [7] J. Zhang, J. Kaur, R. Rajkhowa, J. Li, X. Liu, X. Wang, Mechanical properties and structure of silkworm cocoons: a comparative study of Bombyx mori, Antheraea assamensis, Antheraea pernyi and Antheraea mylitta silkworm cocoons, Mater. Sci. Eng. C 33 (6) (2013) 3206–3213. [8] L.P. Lounibos, The cocoon spinning behaviour of the Chinese oak silkworm, Antheraea pernyi, Anim. Behav. 23 (4) (1975) 843–853. [9] S.Y. Takahashi, G. Suzuki, E. Ohnishi, Origin of oxalic acid in Ca oxalate crystals in the malpighian tubes of the tent caterpillar, Malacosoma neustria testacea, J. Insect Physiol. 15 (3) (1969) 403–407. [10] A. Nisal, K. Trivedy, H. Mohammad, S. Panneri, S.S. Gupta, A. Lele, R. Manchala, N.S. Kumar, M. Gadgil, H. Khandewal, S. More, R.S. Laxman, Uptake of azo dyes into silk glands for production of colored silk cocoons using green feeding approach, ACS Sustain. Chem. Eng. 2 (2) (2013) 312–317. [11] N. Al-Salim, E. Barraclough, E. Burgess, B. Clothier, M. Deurer, S. Green, L. Malone, G. Weir, Quantum dot transport in soils, plants and insects, Sci. Total Environ. 409 (17) (2011) 3237–3248. [12] F. Chen, D. Porter, F. Vollrath, Structure and physical properties of silkworm cocoons, J. R. Soc. Interface 9 (74) (2012) 2299–2308. [13] F. Chen, D. Porter, F. Vollrath, Morphology and structure of silkworm cocoons, Mater. Sci. Eng. C 32 (4) (2012) 772–778. [14] Y. Zhang, H. Yang, H. Shao, X. Hu, Antheraea pernyisilk fibre: apotential resource for artificially biospinning spider dragline silk, J. Biomed. Biotechnol. (2010) 1–9, Article ID 683962.

[15] H.P. Zhao, X.Q. Feng, S.W. Yu, W.Z. Cui, F.Z. Zou, Mechanical properties of silkworm cocoons, Polymer 46 (21) (2005) 9192–9201. [16] J. Zhang, R. Rajkhowa, J. Li, X. Liu, X. Wang, Silkworm cocoon as natural material and structure for thermal insulation, Mater. Des. 49 (2013) 842–849. [17] M. Roy, S.K. Meena, T.S. Kusurkar, S.K. Singh, N.K. Sethy, K. Bhargava, S. Sarkar, M. Das, Carbon dioxide gating in silk cocoon, Biointerphases 7 (45) (2012) 1–4. [18] B. Blossman-Myer, W.W. Burggren, The silk cocoon of the silkworm, Bombyx mori: macro structure and its influence on transmural diffusion of oxygen and water vapor, Comp. Biochem. Physiol., Part A: Mol. Integr. Physiol. 155 (2) (2010) 259–263. [19] H.V. Danks, The roles of insect cocoons in cold conditions, Eur. J. Entomol. 101 (3) (2004) 433–438. [20] R.H. Walters, O.A. Hougen, Silk degumming: I. Degradation of silk sericin by alkalies, Text. Res. J. 5 (2) (1934) 92–104. [21] P. Aramwit, A. Sangcakul, The effects of sericin cream on wound healing in rats, Biosci. Biotechnol. Biochem. 71 (10) (2007) 2473–2477. [22] J. Kaur, R. Rajkhowa, T. Tsuzuki, K. Millington, J. Zhang, X.-G. Wang, Photo-protection by silk cocoons, Biomacromolecules 14 (10) (2013) 3660–3667. [23] A. Manosroi, K. Boonpisuttinant, S. Winitchai, W. Manosroi, J. Manosroi, Free radical scavenging and tyrosinase inhibition activity of oils and sericin extracted from Thai native silkworms (Bombyx mori), J. Pharm. Biol. 48 (8) (2010) 855–860. [24] S. Sarovart, B. Sudatis, P. Meesilpa, B.P. Grady, R. Magaraphan, The use of sericin as an antioxidant and antimicrobial for polluted air treatment, Rev. Adv. Mater. Sci. 5 (2003) 193–198. [25] R. Rajendran, C. Balakumar, R. Sivakumar, T. Amruta, N. Devaki, Extraction and application of natural silk protein sericin from Bombyx mori as antimicrobial finish for cotton fabrics, J. Text. Inst. 103 (4) (2012) 458–462. [26] L. Hazarika, C. Saikia, A. Kataky, S. Bordoloi, J. Hazarika, Evaluation of physico chemical characteristics of silk fibres of Antheraea assama reared on different host plants, Bioresour. Technol. 64 (1) (1998) 67–70. [27] D. Raju Phukan, Muga silk industry of Assam in historical perspectives, Glob. J. Hum. Soc. Sci. Res. 12 (9-D) (2012). [28] K. Haggag, H. El-Sayed, O.G. Allam, Degumming of silk using microwave-assisted treatments, J. Nat. Fibres 4 (3) (2007) 1–22. [29] N. Reddy, Y. Yang, Structure and properties of ultrafine silk fibres produced by Theriodopteryx ephemeraeformis, J. Mater. Sci. 45 (24) (2010) 6617–6622. [30] G. Freddi, Y. Gotoh, T. Mori, I. Tsutsui, M. Tsukada, Chemical structure and physical properties of Antheraea assama silk, J. Appl. Polym. Sci. 52 (6) (1994) 775–781. [31] H.P. Zhao, X.Q. Feng, H. Gao, Ultrasonic technique for extracting nanofibres from nature materials, Appl. Phys. Lett. 90 (7) (2007) 073112. [32] N.M. Mahmoodi, M. Arami, F. Mazaheri, S. Rahimi, Degradation of sericin (degumming) of Persian silk by ultrasound and enzymes as a cleaner and environmentally friendly process, J. Clean. Prod. 18 (2010) 146–151. [33] K. Sen, K.M. Babu, Studies on Indian silk. I. Macrocharacterization and analysis of amino acid composition, J. Appl. Polym. Sci. 92 (2) (2004) 1080–1097. [34] F. Chen, D. Porter, F. Vollrath, Silk cocoon: multilayer structure and mechanical properties, Acta Biomater. 8 (7) (2012) 2620–2627. [35] J.M. Concon, Food Toxicology. Part A: Principles and Concepts; Part B: Contaminants and Additives, Marcel Dekker Inc., 1988 [36] V.R. Franceschi, P.A. Nakata, Calcium oxalate in plants: formation and function, Annu. Rev. Plant Biol. 56 (2005) 41–71. [37] T. Shippey, Vibrational studies of calcium oxalate monohydrate (whewellite) and an anhydrous phase of calcium oxalate, J. Mol. Struct. 63 (2) (1980) 157–166. [38] K.J. Kociba, P.K. Gallagher, A study of calcium oxalate monohydrate using dynamic differential scanning calorimetry and other thermoanalytical techniques, Thermochim. Acta 282–283 (1996) 277–296. [39] R. Rajkhowa, V. Gupta, V. Kothari, Tensile stress–strain and recovery behavior of Indian silk fibres and their structural dependence, J. Appl. Polym. Sci. 77 (11) (2000) 2418–2429. [40] M.M.R. Khan, M. Tsukada, Y. Gotoh, H. Morikawa, G. Freddi, H. Shiozaki, Physical properties and dyeability of silk fibres degummed with citric acid, Bioresour. Technol. 101 (21) (2010) 8439–8445. [41] S. Das, The preparation and processing of tussah silk, J. Soc. Dye. Colour. 108 (11) (1992) 481–486. [42] S. Das, S. Chowdhury, N. Das, An improved method of tussah silk reeling, J. Text. Inst. 83 (1992) 279–281.