- Email: [email protected]
Comparing the properties of Bombyx mori silk cocoons against sericin-fibroin regummed biocomposite sheets
Materials Science and Engineering C 65 (2016) 215–220
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Comparing the properties of Bombyx mori silk cocoons against sericin-fibroin regummed biocomposite sheets Alexander Morin, Parvez Alam ⁎ Laboratory of Paper Coating and Converting, Centre for Functional Materials, Abo Akademi University, Porthaninkatu 3, 20500 Turku, Finland
a r t i c l e
i n f o
Article history: Received 22 February 2016 Received in revised form 29 March 2016 Accepted 6 April 2016 Available online 14 April 2016 Keywords: Bombyx mori Degumming Silk Biocomposite Biowaste
a b s t r a c t This paper considers the utility of sericin, a degumming waste product, in the regumming of Bombyx mori silk fibroin fibres to form sericin-fibroin biocomposites. Regummed biocomposites have a chemical character that is somewhat closer to fibroin than sericin, though sericin presence is confirmed through FT-IR spectroscopy. Using direct measurements we further find the weight fractions of sericin in the regummed biocomposites and the native cocoons differ by only 5%. Mechanically, B. mori cocoons exhibit brittle stress-strain characteristics, failing at strengths of X ¼ 16.6 MPa and at strains of X ¼ 13%. Contrarily, aligning fibroin fibres to a unidirectional axis in the regummed biocomposites causes them to exhibit characteristics of strain hardening, which is itself a typical characteristic of silk fibre pulled in tension. Though they are half as strong (X ¼ 7.2 MPa), regummed biocomposites are able to absorb five times more mechanical energy (X ¼ 5.6 MJm−3) than the B. mori cocoons (X ¼ 1.1 MJm−3) and are furthermore able to elongate to more than ten times (X ¼ 180%) that of the native cocoons prior to failure. Our research shows that degummed B. mori cocoons can be regummed into sheets that have potential for use as load bearing engineering biocomposites. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Bombyx mori caterpillars are responsible for a major share in the global production of silks [1]. Fibroin, the core fibrous protein component of B. mori cocoons, has been a ubiquitous biopolymer used primarily in the textiles industries. Nevertheless, there is growing impetus towards the application of silk in composites [2,3], sometimes as the matrix material and other times as the reinforcement. A number of researchers have considered fibroin as a polymer matrix for nanotube reinforcements [4–6]. Single walled nanotube (SWNT) reinforcements [5] are able to increase stiffness but due to complexities associated with the alignment of SWNT along fibre lengths, other properties such as strength are compromised. Contrarily, multi-walled nanotubes (MWNTs) are able to align more readily, which improves the strength and stiffness of the fibres, but reduces elongation [6]. This results from the development of significant hydrogen bond networks between the fibroin fibres and the MWNT reinforcements, which concurrently improves resistance to washing and sonication techniques [7]. Jin and co-workers [8] applied fibroin fibres within a polyethylene matrix to yield high strength composites with respectable elongation. To ensure that the fibroin fibres would carry sufficient levels of strength-inducing β −sheets [9], Jin and coworkers manufactured the composites in a humidified environment. β− sheets control essentially, the mechanical character of silks. They
⁎ Corresponding author. E-mail address: parvez.alam@abo.fi (P. Alam).
http://dx.doi.org/10.1016/j.msec.2016.04.026 0928-4931/© 2016 Elsevier B.V. All rights reserved.
improve strength by creating nanoconfined networks of hydrogen bonds [10] and influence the development of semi-crystalline matter at their interfaces, which in turn, moderates stress transfer between the crystalline and amorphous regions of silk [11]. A fundamental benefit of silk is that it is a renewable, sustainable and biodegradable (RSB) material. As such there is growing interest in combining fibroin fibres to other biodegradable polymers to form composites. In combination with cellulose, silk has been shown to develop considerable hydrogen bonding [12], which if sufficiently close-range may also instigate molecular anchoring [13], further raising bond strength. When compared against pure fibroin fibres, cellulose-fibroin composites exhibit superior mechanical performance, as do fibroinpolylactic acid composites [14]. The contrary is true for fibroin-collagen films, which have reduced mechanical performance as compared to pure fibroin fibres [15] even though they also develop vast networks of hydrogen bonds. Unlike fibroin, sericin is a proteinous waste product from the degumming of silk cocoons, which is a necessary means of isolating and purifying silk fibroin. Since sericin makes up 20–30% by weight of each cocoon, there is a considerable volume of waste that arises through the process of degumming [16]. Several methods of degumming are utilised industrially including; heating in water, heating in soap water, enzymatic degumming and acid treatments [17]. Nevertheless, degumming has been reported to detrimentally alter the properties of silk proteins, the extent of which is a function of the duration and type of degumming treatment [18]. Water and soap treatments, enzyme treatments and acid treatments have all been shown to cause undesirable
216
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
effects [19] such as loss in tensile strength, damage of the protein molecules and bond breakage. All treatments (excluding enzymatic treatment) increase the elongation properties of fibroin fibres, whilst concurrently reducing its elastic modulus [20]. The utilisation of sericin waste in biomaterials for mechanical benefit has been considered, [21], though for the most part it has been blended into polymer melts to improve the properties of the melt. Some researchers hypothecate the use of sericin in pharmaceuticals [22]. Nevertheless, there is still a significant quantity of sericin that is left unused, much of which is left as unrecovered waste [16]. The successful recovery and utility of waste sericin holds potential for both environmental and economic benefit. Sericin is the natural bioglue for fibroin, yet regumming of fibroin fibres by sericin has never been considered or undertaken as a process by which means native cocoons can be converted into engineering biocomposites. Its utilisation as a bioglue seems a natural move towards cleaner biocomposites production and waste management. The objective of the research herein, is to manufacture biocomposite sheets of fibroin fibres and waste sericin via degumming and subsequent regumming processes. We subsequently aim to compare the mechanical properties of regummed biocomposite sheets against those of natural silk cocoons.
2. Materials and Methods 2.1. Degumming process B. mori cocoons were purchased from the Kabondo Silk Factory and Marketplace, Kisumu, Kenya. The cocoons were cut in half to remove dried larvae after which they were heated in batches of 30 g in 1500 ml deionised water at 95 °C for 4 h. The temperature was checked every 30 min using a thermometer. This procedure results in the separation of sericin waste from fibroin fibres, the sericin remaining in solution. The solution of sericin was then separated from the fibroin fibres by filtration through 30 nm pore diameter Whatman paper after which the waste sericin solution was stored in a refrigerator at approximately 7 °C prior to biocomposite fabrication. To ensure that all the sericin was removed from the B. mori fibroin fibres [23,24], the cocoons were then heated again in deionised water at 95 °C for one more hour, however this time in the presence of a dishwasher tablet (Kiilto Green-Easy Tabs) containing polycarboxylates, non-ionic tensides, enzymes and oxygen based whiteners. Following this, the fully degummed fibroin fibres were washed with deionised water several times to remove all traces of these chemicals.
2.2. Manufacture of sericin-fibroin biocomposite sheets Degummed silk cocoons were stretched out while still wet to straighten and separate them using a wooden board. These were left to dry at room temperature overnight. Fibroin fibres were then cut into equal lengths and aligned as unidirectional fibres in a containment unit after which sericin was poured over the fibres to essentially, regum the fibres. The mixture was dried at 25 °C and at a relative humidity (RH) of 55% for several days until solidified. The dried manufactured biocomposite sheets were then used to determine the fractions of sericin and fibroin in the biocomposite sheet since sericin can migrate away from the fibroin fibres during the process of drying. This migration occurs at the edges and free surfaces of the material where the liquid surface area to volume ratio is high, exacerbating the rate of dynamic motion and pulling more of the free floating sericin molecules to the free surfaces. Here, the sericin shows high affinity to the containment unit and we notice this results in sericin migration. To determine the weight fractions of each, offcuts from the biocomposite sheets were dried, weighed, heated in deionised water for at 95 °C for 4 h for sericin removal, dried again and then re-weighed as fibroin only.
2.3. FT-IR spectroscopy Fourier Transform Infrared (FT-IR) spectroscopy was performed in ATR mode to characterise the chemistry and secondary structures of pure B. mori silk cocoon, pure sericin, pure fibroin and the regummed biocomposite sheets. This method was also used to qualitatively differentiate the biostructures and chemistry of the biocomposite sheets against the native B. mori cocoons. 2.4. Mechanical Testing Biocomposite sheets were cut into dog bone shapes with the fibre axis extending parallel to the gauge length of the dog bone. The samples were 4 mm wide and had a 40 mm gauge length. These were conditioned in a humidity chamber overnight at 25 °C/55% RH prior to testing. An Instron 8872 was used to perform tensile tests in the fibre axis direction at a testing speed of 10 mm/min. Native, unprocessed B. mori cocoons were cut into rectangular sections, and tested in a direction parallel to the equatorial axis of the cocoon at the same testing speed. Rectangular sections were used because it was difficult to accurately cut the preferential dog bone shapes from the cocoons. The samples were 10 mm wide and 40 mm long. In total, 22 tests were conducted for each of the biocomposites and 10 tests were conducted for the native cocoons. Though samples were of similar gauge length and width, thickness measurements were taken and averaged for each sample prior to testing for the accurate conversion of load to stress. 2.5. Scanning electron microscopy (SEM) To elucidate the microstructural arrangements in both native B. mori cocoons and in the regummed biocomposites, electron microscopy was conducted using a Jeol-JSM-6335-F Field Emission Scanning Electron Microscope. SEM was also used to characterise the fracture profiles of native cocoons and regummed biocomposites. Samples were sputter coated with 30–40 nm of platinum prior to SEM. 2.6. Molecular dynamics modelling Molecular dynamics simulations were conducted in Ascalaph Designer to determine characteristics of adhesion and binding sites of sericin to fibroin fibre surfaces under the conditions of high humidity. The sericin molecule was built using the following amino acid sequence reported in [25]: SSTGSSSNTDSNSNSVGSSTSGGSSTYGYSSNSRDGSV. This was then folded to its energetic steady state in a vacuum using an AMBER94 force field, implicit water (Sheffield) conditions defining solvation energy electrostatics [26] at a time step of 2.5 fs. The AMBER94 force field was chosen since it is commonly used for protein simulations with a focus on inter and intramolecular interactions [27]. This force field includes Coulombs law to define electrostatic values, Fourier series to estimate torsion terms and potentials to calculate angle and VDW terms [28]. The sericin after degumming is in solution and will naturally fold in aqueous conditions, prior to its use as a regumming agent for fibroin fibres. To develop a fibroin fibre surface upon which the sericin molecule could attach, individual fibroin molecules were packed into a periodic box and simulated to steady state. The fibroin molecules were built based on the most frequent motifs that are reported to recur in fibroin [29]. The following four chain sequences of fibroin that were modelled follow. 1. SAGSGAGAGYGAGVGAGYGAGYGAGAGSGAGAGSGAGAGSGAGAG SGAGAGSGAGAGAG 2. SAGSGAGAGYGAGAGAGYGAGYGAGVGAGYGAGAGSGAGAGSGA GAGSGAGAGSGAGAGSGAGAG 3. SAGSGAGAGYGAGAGSGAGAGSGAGAGSGAGSGAGAGSGAGAGS GAGAGSGAGAGAGSGAGA
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
217
4. SGTGSGAGAGYGAGAGAGYGAGYGAGVGAGYGAGAGSGAGAGAG AGAGSGAGAGSGAGAG 24 such fibroin molecules were arranged within the periodic box in the following order: (top row) 12341234 (2nd row) 24132413 (3rd row) 31423142 (bottom row) 43214321. These were geometrically optimised using the Monte Carlo method, after which molecular dynamics simulations were undertaken to pack together the fibroin structure. The same parameters as described above were used for the simulation and the resultant structure comprised an array of fibroin molecules that were aligned approximately unidirectionally at steady state. Finally, to determine the adhesive energies between the folded sericin molecule and the fibroin fibre surface, molecular dynamics simulations were conducted between the two, the molecular fibroin array remaining fixed in space. Subsequent postprocessing was performed using VMD developed by the University of Illinois [30]. 3. Results and discussion 3.1. General observations When peeled from the containment unit in which the biocomposite sheets were manufactured, we found there were no cracks or holes in the sheet. Fibres were for the most part aligned in parallel, Fig. 1(a), though some deviation was found to exist especially at the interface between containment unit and sheet, and at the free surface where fibres have greater freedom of mobility, Fig. 1(b). The free surface of the native B. mori cocoon actually exhibited considerably greater randomness in fibre orientation than the regummed biocomposite sheet at its free surface, Fig. 1(c). The weight fraction of fibroin in the native cocoons was found to be on average 74% (±1.73% SD), whereas in the regummed biocomposites the weight fraction of fibroin was measured to be on average 79% (±0.58% SD). Thus after manufacturing, there is a slightly higher percentage of fibroin in the composite sheet, but when observing the standard deviations we note that this appears to be distributed more homogenously throughout the material than in the native cocoon. Nevertheless, the percentage of fibroin is not distinctly different (ca. 5%) between native cocoon and biocomposite sheet. 3.2. FT-IR characterisation Using tables from [31] we find that both native cocoons and regummed biocomposite sheets are made up of proteins with similar secondary structures, Fig. 2. Amide I 1616 cm−1 peaks correspond to β-sheet secondary structures, amine II peaks at 1516 cm− 1 reflect random coils and the peak at 1235 cm− 1 is an amide III peak that is associated with random coils and α-helices. These and other relevant shoulder-peaks that relate to protein secondary structures are summarised in Table 1. Degumming and regumming cocoons into biocomposite sheets does not therefore have any detrimental effect on the secondary structures of the proteins. Nevertheless, what is noticeable is that whereas the regummed biocomposite shows greater similarity to pure fibroin, the cocoon has a more evident mix of both sericin and fibroin. These results are presumably a reflection of (a) the higher weight fractions of fibroin measured in the regummed biocomposites as compared to the native cocoons and (b) sericin migration from the regummed biocomposite free surfaces.
Fig. 1. SEM micrographs showing (a) a sericin-fibroin biocomposite sheet cross section, scale bar = 200 μm (b) a plan view of the free surface of a sericin-fibroin biocomposite sheet, scale bar = 50 μm and (c) a plan view of the free surface of a native B. mori cocoon, scale bar = 200 μm.
3.3. Mechanical properties and failure analyses Table 2 provides details on the mechanical properties of both native B. mori cocoons and regummed biocomposite sheets. The native cocoons are of higher strength and stiffness than the regummed biocomposites, but they are concurrently extremely brittle and exhibit comparatively low characteristics of elongation. The mechanical property measurements for native cocoons are numerically comparable to those reported in [34].
SEM observation of the native cocoons (refer to Fig. 1(c) and 3(a)) revealed that the fibre orientation of the cocoons is somewhat random. As such, the onset of fracture in such a construction would give rise to more rapid crack propagation than in unidirectional composites, since there is little continuity in the lateral resistance to crack propagation. This can be observed in Fig. 3(b) where fibres are pulled out in the loading direction, but extend from several different fibre directions. In this figure,
218
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220 Table 2 Mean mechanical properties of native B. mori cocoons (sample number = 10) and of regummed biocomposites (sample number = 22). Standard errors are provided in parentheses below each mean value.
Tensile Modulus [MPa] Strain Hardening Slope [MPa] Tensile Strength [MPa] Elongation [%] Toughness [MJm−3] Yield Strength [MPa]
Native cocoon
Regummed biocomposite
300 (±20) NA (± NA) 16.6 (±0.5) 13 (±0.25) 1.1 (±0.01) 10.3 (±0.05)
98 (±30) 100 (±30) 7.2 (±0.3) 180 (±3.2) 5.6 (±0.2) 1.2 (±0.03)
Fig. 2. FT-IR spectra for fibroin, regummed biocomposites, native B. mori cocoons and sericin.
it is also clear to see that fracture propagation is relatively linear and does not meander frequently, indicating that fracture energy is not efficiently redirected in the native cocoons during load-induced failure. The regummed biocomposites function dissimilarly to the cocoons, since they have fibres oriented to the loading axis, crack propagation perpendicular to the loading axis is frequently blocked by adjacent fibres. As such, the regummed biocomposites are able to hold load over far greater lengths of deformation and are consequently able to absorb considerably more mechanical energy, Fig. 4. High elongation in fibroin/sericin films has also been achieved by cross-linking with genipin [35]. Importantly, the regummed biocomposites exhibit fivetimes higher toughness values than the native cocoons. Unlike the cocoons, the regummed biocomposites are able to strain harden (refer to Fig. 4). We suggest this characteristic of strain hardening comes from that the fibres are oriented in one direction and as such are pulled in tension along their individual fibre axes (unlike for the cocoons). Since individual silk fibres typically strain harden when loaded in tension along their fibre axes [36,37], this characteristic logically manifests itself in the regummed biocomposite. The toughness values of the biocomposites are thus a coupled consequence of strain hardening and elongation. It should be noted that the high elongation observed here may also be a function of chemically cleaning the fibres after degumming [20], which was undertaken to remove all traces of sericin. Using the height method to determine crystallinity from the FTIR spectra [38] we notice that the native cocoons are 37% crystalline, whereas the biocomposite sheets are 34% crystalline. Several spectra were analysed to ensuring statistical repeatability. The lower crystallinity through postprocessing might contribute to lower strength and stiffness values of the regummed biocomposites as compared to the native cocoons. Postprocessing is generally understood to be detrimental to the strength and stiffness of fibroin fibres [18,19], which in turn may relate in part, to a reduction of crystallinity as is reported herein.
It could be expected that the unidirectional fibre orientation of the regummed biocomposite should also give it superior strength and stiffness. Yet, the contrary is observed. We suggest the reason for this is due to a reduced strength of interfacial adhesion between sericin and fibroin fibres during the processing and manufacturing of degummed and regummed biocomposites. A high interfacial strength between sericin and fibroin fibre in the native cocoons presumably allows them to bear considerable load prior to catastrophic fast fracture, as is observed in Fig. 4. Contrarily, degumming may give rise to a reduction in the molecular weight of sericin [18], and it would follow that a reduction in the molecular weight of sericin bioglue reduces the number of potential secondary force interaction binding sites between the sericin and fibroin molecules. Additionally, fibres connected by sericin were observed to separate during testing, pinpointing sericin bioglues as the weakest links of the biocomposites that are prone to interfacial failure in shear.
Table 1 Summary of relevant FTIR spectral peaks/shoulders and their assignments. Assignments
Wavenumber [cm−1]
Amide I, C_O (stretch), CN (stretch), β-turn (shoulder) Amide I, C_O (stretch), CN (stretch), random coil and/or α-helical (shoulder) Amide I, C_O (stretch), CN (stretch), β-sheet (peak) Amide II, NH (bend), CN (stretch), β-sheet (shoulder) Amide II, NH (bend), CN (stretch), random coil and/or α-helical (peak) Amide III, random coil and/or α-helical (peak)
1700 1660 1616 1530 1516 1235
Fig. 3. Example SEM images of B. mori cocoon mechanical test samples (a) unfractured and untested, and (b) fractured after mechanical testing. Scale bars = 1 mm.
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
219
cocoon. This protein acts as a sticky glue, binding together fibroin fibres within the native cocoon. In this paper, we consider the re-use of sericin waste protein as a bioglue for regumming fibroin fibres into biocomposite sheets. When comparing against the native cocoons we find that the strength and stiffness of the native cocoons are considerably higher than for the regummed biocomposites, but that the native cocoons fracture under lower elongation. The elongation and toughness values of regummed biocomposites nevertheless, far exceed those of the native cocoons and as such, regummed biocomposites show promise in specific applications where high stretch and high energy absorption are desired properties. Unlike the native cocoons, the regummed biocomposites show clear characteristics of strain hardening. This indicates that the silk fibroin fibres play a predominating role in bearing imposed tensile loads. We find that the protein secondary structures of both fibroin and sericin through regumming are preserved, though a potential reduction in the molecular weight of sericin may give rise to decreased interfacial strength of interactions (predominantly serine-based), which may be a reason for the reduced strength and stiffness of the regummed biocomposite sheets as compared to native cocoons. Fig. 4. Example stress strain curves for a native B. mori cocoon and for a regummed biocomposite. The insert shows the curve for the cocoon enlarged for visual clarity. The green box indicates the region from which the strain hardening slope (see Table 2) is calculated.
3.4. Molecular modelling Molecular models are a useful tool in developing a more holistic understanding of sericin-fibroin interactions at the molecular level. Table 3 shows the short range (b 3Å) and mid-range (3− 4Å) hydrogen bonds that develop between sericin molecules and the fibroin surface. The primary amino acids that form H-bonds as observed in our models (and percentages in parentheses) are Ser (48%), Ala (22%), Asn (13%), Gly (8.5%) and Arg (8.5%). It is immediately noticeable that serine is the most important amino acid for binding sericin to fibroin in view of H-bonding. This further evidences aforementioned hypotheses by Gray et al. [32] and Kundu et al. [33] who postulate upon the contribution of serine to stickiness between molecules. Their hypotheses single out serine as a predominant sticky amino acid since it is present in all regions of the sericin protein and constitutes 50% of the total amino acid sequence. Moreover, serine together with the other amino acids in Table 3 possesses specific properties which encourage the formation of H-bonds. These include: 1. A lack of stearic limitations (due to its simple structure). 2. An abundancy of hydrogen, oxygen and nitrogen atoms (since they participate in H-bonding).
4. Conclusions Sericin is a waste product from Bombyx mori silk cocoon purification, resulting in weight for weight between 20 and 30% of the original Table 3 Short and mid range H-bond connections in our model between amino acids of sericin and fibroin.
b3Å 3 − 4Å
Sericin
Fibroin
Ser Arg Ser Ser Ser Asn Asn Asn Asp Arg
Gly Ala Ser Ser Ala Ser Ser Ser Ala Gly
References [1] K. Yukuhiro, H. Sezutsu, N. Yonemura, Evolutionary Divergence of Lepidopteran and Trichopteran Fibroins, in: T. Asakura, T. Miller (Eds.), Biotechnology of Silk, Springer Science and Business Media, Dordrecht, 2014 (2014). [2] D. Shah, Natural fibre composites: Comprehensive Ashby-type materials selection charts, Mater. Des. 62 (2014) 21–31. [3] D. Shah, D. Porter, F. Vollrath, Opportunities for silk textiles in reinforced biocomposites: Studying through-thickness compaction behaviour, Compos. Part A-Appl. Sci. 62 (2014) 1–10. [4] R. Khare, S. Bose, Carbon nanotube based composite, J. Miner. Miner. Charact. Eng. 4 (1) (2005) 31–46. [5] J. Ayutsede, M. Gandhi, S. Sukigara, H. Ye, C. Hsu, Y. Gogotsi, F. Ko, Carbon Nanotube Reinforced Bombyx mori Silk Nanofibers by the Electrospinning Process, Biomacromolecules 7 (1) (2006) 208–214. [6] H. Kim, W. Park, Y. Kim, H. Jin, Silk fibroin films crystallized by multiwalled carbon nanotubes, Int. J. Mod. Phys. B 22 (2008) 1807–1812. [7] M. Kang, H. Jin, Electrically conducting electrospun silk membranes fabricated by adsorption of carbon nanotubes, Colloid Polym. Sci. 285 (10) (2007) 1163–1167. [8] H. Jin, J. Park, R. Valluzzi, P. Cebe, D. Kaplan, Biomaterial Films of Bombyx Mori Silk Fibroin with Poly(ethylene oxide), Biomacromolecules 5 (3) (2004) 711–717. [9] P. Alam, Protein unfolding versus beta-sheet separation in spider-silk nanocrystals, Adv. Nat. Sci.: Nanosci. Nanotechnol. 5 (1) (2014) 015015(6). [10] M. Pahlevan, M. Toivakka, P. Alam, Electrostatic charges instigate ‘concertina-like’ mechanisms of molecular toughening in MaSp1 (spider silk) proteins, Mater. Sci. Eng. C 41 (2014) 329–334. [11] E. Sintya, P. Alam, Self-assembled semi-crystallinity at parallel beta-sheet nanocrystal interfaces in clustered MaSp1 (spider silk) proteins, Mater. Sci. Eng. C 58 (2016) 366–371. [12] G. Freddi, M. Romano, M. Massafra, M. Tsukada, Silk fibroin/cellulose blend films: Preparation, structure, and physical properties, J. Appl. Polym. Sci. 56 (12) (1995) 1537–1545. [13] P. Alam, Fagerlund Pi, P. Hagerstrand, J. Toyryla, S. Amini, M. Tadayon, A. Miserez, V. Kumar, M. Pahlevan, M. Toivakka, L-lysine templated CaCO3 precipitated to flax develops flowery crystal structures that improve the mechanical properties of natural fibre reinforced composites, Compos. Part A-Appl. Sci. 75 (2015) 84–88. [14] X. Hu, D. Kaplan, P. Cebe, Effect of water on the thermal properties of silk fibroin, Thermochim. Acta 461 (1–2) (2007) 137–144. [15] K. Hu, Biocompatible fibroin blended films with recombinant human-like collagen for hepatic tissue engineering, J. Bioact. Compat. Polym. 21 (1) (2006) 23–37. [16] M.H. Wu, J.X. Yue, Y.Q. Zhang, Ultrafiltration recovery of sericin from the alkaline waste of silk floss processing and controlled enzymatic hydrolysis, J. Clean. Prod. 76 (2014) 154–160. [17] M. Ho, H. Wang, K. Lau, J. Lee, D. Hui, Interfacial bonding and degumming effects on silk fibre/polymer biocomposites, Compos. Part B 43 (7) (2012) 2801–2812. [18] G. Altman, F. Diaz, C. Jakuba, T. Calabro, R. Horan, J. Chen, et al., Silk-based biomaterials, Biomaterials 24 (3) (2003) 401–416. [19] A. MacIntosh, V. Kearns, A. Crawford, P. Hatton, Silk-based biomaterials for tissue engineering, Top. Tiss. Eng. 4 (2008). [20] M. Khan, M. Tsukada, Y. Gotoh, H. Morikawa, G. Freddi, H. Shiozaki, Physical properties and dyeability of silk fibers degummed with citric acid, Bioresour. Technol. 101 (21) (2010) 8439–8445. [21] Y.Q. Zhang, Applications of natural silk protein sericin in biomaterials, Biotechnol. Adv. 20 (2) (2002) 91–100. [22] P. Aramwait, T. Siritientong, T. Srichana, Potential applications of silk sericin, a natural protein from textile industry by-products, Waste Manag. Res. 30 (3) (2012) 217–224. [23] P. Jiang, H. Liu, C. Wang, L. Wu, J. Huang, C. Guo, Tensile behavior and morphology of differently degummed silkworm (Bombyx mori) cocoon silk fibres, Mater. Lett. 60 (7) (2006) 919–925.
220
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
[24] J. Hardy, T. Scheibel, Silk-inspired polymers and proteins, Biochem. Soc. Trans. 37 (4) (2009) 677–681. [25] J. Huang, R. Valluzzi, E. Bini, B. Vernaglia, D. Kaplan, Cloning, expression, and assembly of sericin-like protein, J. Biol. Chem. 278 (46) (2003) 46117–46123. [26] J. Grant, B. Pickup, M. Sykes, C. Kitchen, A. Nicholls, A simple formula for dielectric polarisation energies: the Sheffield Solvation Model, Chem. Phys. Lett. 441 (1–3) (2007) 163–166. [27] D. Case, T. Cheatham, T. Darden, H. Gohlke, R. Luo, K. Merz, et al., The Amber biomolecular simulation programs, J. Comput. Chem. 26 (16) (2005) 1668–1688. [28] L. Yang, C. Tan, M. Hsieh, J. Wang, Y. Duan, P. Cieplak, et al., New-generation amber united-atom force field, J. Phys. Chem. B. 110 (26) (2006) 13166–13176. [29] C. Zhou, F. Confalonieri, M. Jacquet, R. Perasso, Z. Li, J. Janin, Silk fibroin: structural implications of a remarkable amino acid sequence, Protein 44 (2) (2001) 19–122 (Grant2007). [30] W. Humphrey, A. Dalke, K. Schulten, VMD - visual molecular dynamics, J. Mol. Graph. 14 (1996) 33–38. [31] S. Ling, Z. Qi, B. Watts, Z. Shao, X. Chen, Structural determination of protein-based polymer blends with a promising tool: combination of FTIR and STXM spectroscopic imaging, Phys. Chem. Chem. Phys. 16 (17) (2014) 7741.
[32] T. Gray, B. Matthews, Intrahelical hydrogen bonding of serine, threonine and cysteine residues within α-helices and its relevance to membrane-bound proteins, J. Mol. Biol. 175 (1) (1984) 75–81. [33] S. Kundu, B. Dash, R. Dash, D. Kaplan, Natural protective glue protein, sericin bioengineered by silkworms: potential for biomedical and biotechnological applications, Prog. Polym. Sci. 33 (10) (2008) 998–1012. [34] H.P. Zhao, X.Q. Feng, S.W. Yu, W.Z. Cui, F.Z. Zou, Mechanical properties of silkworm cocoons, Polymer 46 (2005) 9192–9201. [35] A. Motta, B. Barbato, C. Foss, P. Torricelli, C. Migliaresi, Stabilization of Bombyx mori silk fibroin/sericin films by crosslinking with PEG-DE 600 and genipin, J. Bioact. Compat. Polym. 26 (2) (2011) 130–143. [36] P. Poza, J. Perez-Rigueiro, M. Elices, J. Llorca, Fractographic analysis of silkworm and spider silk, Eng. Fract. Mech. 69 (2002) 1035–1048. [37] J. Perez-Rigueiro, M. Elices, J. Llorca, C. Viney, Effect of degumming on the tensile properties of silkworm (Bombyx mori) Silk Fiber, J. Appl. Polym. Sci. 84 (7) (2002) 1431–1437. [38] J. Reyes-Gasga, E.L. Martinex-Pineiro, G. Rodriguez-Alvarez, G.E. Tiznado-Orozco, R. Garcia-Garcia, E.F. Bres, XRD and FTIR crystallinity indices in sound hjuman tooth enamel and synthetic hydroxyapatite, Mater. Sci. Eng. C 33 (2013) 4568–4574.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Comparing the properties of Bombyx mori silk cocoons against sericin-fibroin regummed biocomposite sheets Alexander Morin, Parvez Alam ⁎ Laboratory of Paper Coating and Converting, Centre for Functional Materials, Abo Akademi University, Porthaninkatu 3, 20500 Turku, Finland
a r t i c l e
i n f o
Article history: Received 22 February 2016 Received in revised form 29 March 2016 Accepted 6 April 2016 Available online 14 April 2016 Keywords: Bombyx mori Degumming Silk Biocomposite Biowaste
a b s t r a c t This paper considers the utility of sericin, a degumming waste product, in the regumming of Bombyx mori silk fibroin fibres to form sericin-fibroin biocomposites. Regummed biocomposites have a chemical character that is somewhat closer to fibroin than sericin, though sericin presence is confirmed through FT-IR spectroscopy. Using direct measurements we further find the weight fractions of sericin in the regummed biocomposites and the native cocoons differ by only 5%. Mechanically, B. mori cocoons exhibit brittle stress-strain characteristics, failing at strengths of X ¼ 16.6 MPa and at strains of X ¼ 13%. Contrarily, aligning fibroin fibres to a unidirectional axis in the regummed biocomposites causes them to exhibit characteristics of strain hardening, which is itself a typical characteristic of silk fibre pulled in tension. Though they are half as strong (X ¼ 7.2 MPa), regummed biocomposites are able to absorb five times more mechanical energy (X ¼ 5.6 MJm−3) than the B. mori cocoons (X ¼ 1.1 MJm−3) and are furthermore able to elongate to more than ten times (X ¼ 180%) that of the native cocoons prior to failure. Our research shows that degummed B. mori cocoons can be regummed into sheets that have potential for use as load bearing engineering biocomposites. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Bombyx mori caterpillars are responsible for a major share in the global production of silks [1]. Fibroin, the core fibrous protein component of B. mori cocoons, has been a ubiquitous biopolymer used primarily in the textiles industries. Nevertheless, there is growing impetus towards the application of silk in composites [2,3], sometimes as the matrix material and other times as the reinforcement. A number of researchers have considered fibroin as a polymer matrix for nanotube reinforcements [4–6]. Single walled nanotube (SWNT) reinforcements [5] are able to increase stiffness but due to complexities associated with the alignment of SWNT along fibre lengths, other properties such as strength are compromised. Contrarily, multi-walled nanotubes (MWNTs) are able to align more readily, which improves the strength and stiffness of the fibres, but reduces elongation [6]. This results from the development of significant hydrogen bond networks between the fibroin fibres and the MWNT reinforcements, which concurrently improves resistance to washing and sonication techniques [7]. Jin and co-workers [8] applied fibroin fibres within a polyethylene matrix to yield high strength composites with respectable elongation. To ensure that the fibroin fibres would carry sufficient levels of strength-inducing β −sheets [9], Jin and coworkers manufactured the composites in a humidified environment. β− sheets control essentially, the mechanical character of silks. They
⁎ Corresponding author. E-mail address: parvez.alam@abo.fi (P. Alam).
http://dx.doi.org/10.1016/j.msec.2016.04.026 0928-4931/© 2016 Elsevier B.V. All rights reserved.
improve strength by creating nanoconfined networks of hydrogen bonds [10] and influence the development of semi-crystalline matter at their interfaces, which in turn, moderates stress transfer between the crystalline and amorphous regions of silk [11]. A fundamental benefit of silk is that it is a renewable, sustainable and biodegradable (RSB) material. As such there is growing interest in combining fibroin fibres to other biodegradable polymers to form composites. In combination with cellulose, silk has been shown to develop considerable hydrogen bonding [12], which if sufficiently close-range may also instigate molecular anchoring [13], further raising bond strength. When compared against pure fibroin fibres, cellulose-fibroin composites exhibit superior mechanical performance, as do fibroinpolylactic acid composites [14]. The contrary is true for fibroin-collagen films, which have reduced mechanical performance as compared to pure fibroin fibres [15] even though they also develop vast networks of hydrogen bonds. Unlike fibroin, sericin is a proteinous waste product from the degumming of silk cocoons, which is a necessary means of isolating and purifying silk fibroin. Since sericin makes up 20–30% by weight of each cocoon, there is a considerable volume of waste that arises through the process of degumming [16]. Several methods of degumming are utilised industrially including; heating in water, heating in soap water, enzymatic degumming and acid treatments [17]. Nevertheless, degumming has been reported to detrimentally alter the properties of silk proteins, the extent of which is a function of the duration and type of degumming treatment [18]. Water and soap treatments, enzyme treatments and acid treatments have all been shown to cause undesirable
216
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
effects [19] such as loss in tensile strength, damage of the protein molecules and bond breakage. All treatments (excluding enzymatic treatment) increase the elongation properties of fibroin fibres, whilst concurrently reducing its elastic modulus [20]. The utilisation of sericin waste in biomaterials for mechanical benefit has been considered, [21], though for the most part it has been blended into polymer melts to improve the properties of the melt. Some researchers hypothecate the use of sericin in pharmaceuticals [22]. Nevertheless, there is still a significant quantity of sericin that is left unused, much of which is left as unrecovered waste [16]. The successful recovery and utility of waste sericin holds potential for both environmental and economic benefit. Sericin is the natural bioglue for fibroin, yet regumming of fibroin fibres by sericin has never been considered or undertaken as a process by which means native cocoons can be converted into engineering biocomposites. Its utilisation as a bioglue seems a natural move towards cleaner biocomposites production and waste management. The objective of the research herein, is to manufacture biocomposite sheets of fibroin fibres and waste sericin via degumming and subsequent regumming processes. We subsequently aim to compare the mechanical properties of regummed biocomposite sheets against those of natural silk cocoons.
2. Materials and Methods 2.1. Degumming process B. mori cocoons were purchased from the Kabondo Silk Factory and Marketplace, Kisumu, Kenya. The cocoons were cut in half to remove dried larvae after which they were heated in batches of 30 g in 1500 ml deionised water at 95 °C for 4 h. The temperature was checked every 30 min using a thermometer. This procedure results in the separation of sericin waste from fibroin fibres, the sericin remaining in solution. The solution of sericin was then separated from the fibroin fibres by filtration through 30 nm pore diameter Whatman paper after which the waste sericin solution was stored in a refrigerator at approximately 7 °C prior to biocomposite fabrication. To ensure that all the sericin was removed from the B. mori fibroin fibres [23,24], the cocoons were then heated again in deionised water at 95 °C for one more hour, however this time in the presence of a dishwasher tablet (Kiilto Green-Easy Tabs) containing polycarboxylates, non-ionic tensides, enzymes and oxygen based whiteners. Following this, the fully degummed fibroin fibres were washed with deionised water several times to remove all traces of these chemicals.
2.2. Manufacture of sericin-fibroin biocomposite sheets Degummed silk cocoons were stretched out while still wet to straighten and separate them using a wooden board. These were left to dry at room temperature overnight. Fibroin fibres were then cut into equal lengths and aligned as unidirectional fibres in a containment unit after which sericin was poured over the fibres to essentially, regum the fibres. The mixture was dried at 25 °C and at a relative humidity (RH) of 55% for several days until solidified. The dried manufactured biocomposite sheets were then used to determine the fractions of sericin and fibroin in the biocomposite sheet since sericin can migrate away from the fibroin fibres during the process of drying. This migration occurs at the edges and free surfaces of the material where the liquid surface area to volume ratio is high, exacerbating the rate of dynamic motion and pulling more of the free floating sericin molecules to the free surfaces. Here, the sericin shows high affinity to the containment unit and we notice this results in sericin migration. To determine the weight fractions of each, offcuts from the biocomposite sheets were dried, weighed, heated in deionised water for at 95 °C for 4 h for sericin removal, dried again and then re-weighed as fibroin only.
2.3. FT-IR spectroscopy Fourier Transform Infrared (FT-IR) spectroscopy was performed in ATR mode to characterise the chemistry and secondary structures of pure B. mori silk cocoon, pure sericin, pure fibroin and the regummed biocomposite sheets. This method was also used to qualitatively differentiate the biostructures and chemistry of the biocomposite sheets against the native B. mori cocoons. 2.4. Mechanical Testing Biocomposite sheets were cut into dog bone shapes with the fibre axis extending parallel to the gauge length of the dog bone. The samples were 4 mm wide and had a 40 mm gauge length. These were conditioned in a humidity chamber overnight at 25 °C/55% RH prior to testing. An Instron 8872 was used to perform tensile tests in the fibre axis direction at a testing speed of 10 mm/min. Native, unprocessed B. mori cocoons were cut into rectangular sections, and tested in a direction parallel to the equatorial axis of the cocoon at the same testing speed. Rectangular sections were used because it was difficult to accurately cut the preferential dog bone shapes from the cocoons. The samples were 10 mm wide and 40 mm long. In total, 22 tests were conducted for each of the biocomposites and 10 tests were conducted for the native cocoons. Though samples were of similar gauge length and width, thickness measurements were taken and averaged for each sample prior to testing for the accurate conversion of load to stress. 2.5. Scanning electron microscopy (SEM) To elucidate the microstructural arrangements in both native B. mori cocoons and in the regummed biocomposites, electron microscopy was conducted using a Jeol-JSM-6335-F Field Emission Scanning Electron Microscope. SEM was also used to characterise the fracture profiles of native cocoons and regummed biocomposites. Samples were sputter coated with 30–40 nm of platinum prior to SEM. 2.6. Molecular dynamics modelling Molecular dynamics simulations were conducted in Ascalaph Designer to determine characteristics of adhesion and binding sites of sericin to fibroin fibre surfaces under the conditions of high humidity. The sericin molecule was built using the following amino acid sequence reported in [25]: SSTGSSSNTDSNSNSVGSSTSGGSSTYGYSSNSRDGSV. This was then folded to its energetic steady state in a vacuum using an AMBER94 force field, implicit water (Sheffield) conditions defining solvation energy electrostatics [26] at a time step of 2.5 fs. The AMBER94 force field was chosen since it is commonly used for protein simulations with a focus on inter and intramolecular interactions [27]. This force field includes Coulombs law to define electrostatic values, Fourier series to estimate torsion terms and potentials to calculate angle and VDW terms [28]. The sericin after degumming is in solution and will naturally fold in aqueous conditions, prior to its use as a regumming agent for fibroin fibres. To develop a fibroin fibre surface upon which the sericin molecule could attach, individual fibroin molecules were packed into a periodic box and simulated to steady state. The fibroin molecules were built based on the most frequent motifs that are reported to recur in fibroin [29]. The following four chain sequences of fibroin that were modelled follow. 1. SAGSGAGAGYGAGVGAGYGAGYGAGAGSGAGAGSGAGAGSGAGAG SGAGAGSGAGAGAG 2. SAGSGAGAGYGAGAGAGYGAGYGAGVGAGYGAGAGSGAGAGSGA GAGSGAGAGSGAGAGSGAGAG 3. SAGSGAGAGYGAGAGSGAGAGSGAGAGSGAGSGAGAGSGAGAGS GAGAGSGAGAGAGSGAGA
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
217
4. SGTGSGAGAGYGAGAGAGYGAGYGAGVGAGYGAGAGSGAGAGAG AGAGSGAGAGSGAGAG 24 such fibroin molecules were arranged within the periodic box in the following order: (top row) 12341234 (2nd row) 24132413 (3rd row) 31423142 (bottom row) 43214321. These were geometrically optimised using the Monte Carlo method, after which molecular dynamics simulations were undertaken to pack together the fibroin structure. The same parameters as described above were used for the simulation and the resultant structure comprised an array of fibroin molecules that were aligned approximately unidirectionally at steady state. Finally, to determine the adhesive energies between the folded sericin molecule and the fibroin fibre surface, molecular dynamics simulations were conducted between the two, the molecular fibroin array remaining fixed in space. Subsequent postprocessing was performed using VMD developed by the University of Illinois [30]. 3. Results and discussion 3.1. General observations When peeled from the containment unit in which the biocomposite sheets were manufactured, we found there were no cracks or holes in the sheet. Fibres were for the most part aligned in parallel, Fig. 1(a), though some deviation was found to exist especially at the interface between containment unit and sheet, and at the free surface where fibres have greater freedom of mobility, Fig. 1(b). The free surface of the native B. mori cocoon actually exhibited considerably greater randomness in fibre orientation than the regummed biocomposite sheet at its free surface, Fig. 1(c). The weight fraction of fibroin in the native cocoons was found to be on average 74% (±1.73% SD), whereas in the regummed biocomposites the weight fraction of fibroin was measured to be on average 79% (±0.58% SD). Thus after manufacturing, there is a slightly higher percentage of fibroin in the composite sheet, but when observing the standard deviations we note that this appears to be distributed more homogenously throughout the material than in the native cocoon. Nevertheless, the percentage of fibroin is not distinctly different (ca. 5%) between native cocoon and biocomposite sheet. 3.2. FT-IR characterisation Using tables from [31] we find that both native cocoons and regummed biocomposite sheets are made up of proteins with similar secondary structures, Fig. 2. Amide I 1616 cm−1 peaks correspond to β-sheet secondary structures, amine II peaks at 1516 cm− 1 reflect random coils and the peak at 1235 cm− 1 is an amide III peak that is associated with random coils and α-helices. These and other relevant shoulder-peaks that relate to protein secondary structures are summarised in Table 1. Degumming and regumming cocoons into biocomposite sheets does not therefore have any detrimental effect on the secondary structures of the proteins. Nevertheless, what is noticeable is that whereas the regummed biocomposite shows greater similarity to pure fibroin, the cocoon has a more evident mix of both sericin and fibroin. These results are presumably a reflection of (a) the higher weight fractions of fibroin measured in the regummed biocomposites as compared to the native cocoons and (b) sericin migration from the regummed biocomposite free surfaces.
Fig. 1. SEM micrographs showing (a) a sericin-fibroin biocomposite sheet cross section, scale bar = 200 μm (b) a plan view of the free surface of a sericin-fibroin biocomposite sheet, scale bar = 50 μm and (c) a plan view of the free surface of a native B. mori cocoon, scale bar = 200 μm.
3.3. Mechanical properties and failure analyses Table 2 provides details on the mechanical properties of both native B. mori cocoons and regummed biocomposite sheets. The native cocoons are of higher strength and stiffness than the regummed biocomposites, but they are concurrently extremely brittle and exhibit comparatively low characteristics of elongation. The mechanical property measurements for native cocoons are numerically comparable to those reported in [34].
SEM observation of the native cocoons (refer to Fig. 1(c) and 3(a)) revealed that the fibre orientation of the cocoons is somewhat random. As such, the onset of fracture in such a construction would give rise to more rapid crack propagation than in unidirectional composites, since there is little continuity in the lateral resistance to crack propagation. This can be observed in Fig. 3(b) where fibres are pulled out in the loading direction, but extend from several different fibre directions. In this figure,
218
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220 Table 2 Mean mechanical properties of native B. mori cocoons (sample number = 10) and of regummed biocomposites (sample number = 22). Standard errors are provided in parentheses below each mean value.
Tensile Modulus [MPa] Strain Hardening Slope [MPa] Tensile Strength [MPa] Elongation [%] Toughness [MJm−3] Yield Strength [MPa]
Native cocoon
Regummed biocomposite
300 (±20) NA (± NA) 16.6 (±0.5) 13 (±0.25) 1.1 (±0.01) 10.3 (±0.05)
98 (±30) 100 (±30) 7.2 (±0.3) 180 (±3.2) 5.6 (±0.2) 1.2 (±0.03)
Fig. 2. FT-IR spectra for fibroin, regummed biocomposites, native B. mori cocoons and sericin.
it is also clear to see that fracture propagation is relatively linear and does not meander frequently, indicating that fracture energy is not efficiently redirected in the native cocoons during load-induced failure. The regummed biocomposites function dissimilarly to the cocoons, since they have fibres oriented to the loading axis, crack propagation perpendicular to the loading axis is frequently blocked by adjacent fibres. As such, the regummed biocomposites are able to hold load over far greater lengths of deformation and are consequently able to absorb considerably more mechanical energy, Fig. 4. High elongation in fibroin/sericin films has also been achieved by cross-linking with genipin [35]. Importantly, the regummed biocomposites exhibit fivetimes higher toughness values than the native cocoons. Unlike the cocoons, the regummed biocomposites are able to strain harden (refer to Fig. 4). We suggest this characteristic of strain hardening comes from that the fibres are oriented in one direction and as such are pulled in tension along their individual fibre axes (unlike for the cocoons). Since individual silk fibres typically strain harden when loaded in tension along their fibre axes [36,37], this characteristic logically manifests itself in the regummed biocomposite. The toughness values of the biocomposites are thus a coupled consequence of strain hardening and elongation. It should be noted that the high elongation observed here may also be a function of chemically cleaning the fibres after degumming [20], which was undertaken to remove all traces of sericin. Using the height method to determine crystallinity from the FTIR spectra [38] we notice that the native cocoons are 37% crystalline, whereas the biocomposite sheets are 34% crystalline. Several spectra were analysed to ensuring statistical repeatability. The lower crystallinity through postprocessing might contribute to lower strength and stiffness values of the regummed biocomposites as compared to the native cocoons. Postprocessing is generally understood to be detrimental to the strength and stiffness of fibroin fibres [18,19], which in turn may relate in part, to a reduction of crystallinity as is reported herein.
It could be expected that the unidirectional fibre orientation of the regummed biocomposite should also give it superior strength and stiffness. Yet, the contrary is observed. We suggest the reason for this is due to a reduced strength of interfacial adhesion between sericin and fibroin fibres during the processing and manufacturing of degummed and regummed biocomposites. A high interfacial strength between sericin and fibroin fibre in the native cocoons presumably allows them to bear considerable load prior to catastrophic fast fracture, as is observed in Fig. 4. Contrarily, degumming may give rise to a reduction in the molecular weight of sericin [18], and it would follow that a reduction in the molecular weight of sericin bioglue reduces the number of potential secondary force interaction binding sites between the sericin and fibroin molecules. Additionally, fibres connected by sericin were observed to separate during testing, pinpointing sericin bioglues as the weakest links of the biocomposites that are prone to interfacial failure in shear.
Table 1 Summary of relevant FTIR spectral peaks/shoulders and their assignments. Assignments
Wavenumber [cm−1]
Amide I, C_O (stretch), CN (stretch), β-turn (shoulder) Amide I, C_O (stretch), CN (stretch), random coil and/or α-helical (shoulder) Amide I, C_O (stretch), CN (stretch), β-sheet (peak) Amide II, NH (bend), CN (stretch), β-sheet (shoulder) Amide II, NH (bend), CN (stretch), random coil and/or α-helical (peak) Amide III, random coil and/or α-helical (peak)
1700 1660 1616 1530 1516 1235
Fig. 3. Example SEM images of B. mori cocoon mechanical test samples (a) unfractured and untested, and (b) fractured after mechanical testing. Scale bars = 1 mm.
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
219
cocoon. This protein acts as a sticky glue, binding together fibroin fibres within the native cocoon. In this paper, we consider the re-use of sericin waste protein as a bioglue for regumming fibroin fibres into biocomposite sheets. When comparing against the native cocoons we find that the strength and stiffness of the native cocoons are considerably higher than for the regummed biocomposites, but that the native cocoons fracture under lower elongation. The elongation and toughness values of regummed biocomposites nevertheless, far exceed those of the native cocoons and as such, regummed biocomposites show promise in specific applications where high stretch and high energy absorption are desired properties. Unlike the native cocoons, the regummed biocomposites show clear characteristics of strain hardening. This indicates that the silk fibroin fibres play a predominating role in bearing imposed tensile loads. We find that the protein secondary structures of both fibroin and sericin through regumming are preserved, though a potential reduction in the molecular weight of sericin may give rise to decreased interfacial strength of interactions (predominantly serine-based), which may be a reason for the reduced strength and stiffness of the regummed biocomposite sheets as compared to native cocoons. Fig. 4. Example stress strain curves for a native B. mori cocoon and for a regummed biocomposite. The insert shows the curve for the cocoon enlarged for visual clarity. The green box indicates the region from which the strain hardening slope (see Table 2) is calculated.
3.4. Molecular modelling Molecular models are a useful tool in developing a more holistic understanding of sericin-fibroin interactions at the molecular level. Table 3 shows the short range (b 3Å) and mid-range (3− 4Å) hydrogen bonds that develop between sericin molecules and the fibroin surface. The primary amino acids that form H-bonds as observed in our models (and percentages in parentheses) are Ser (48%), Ala (22%), Asn (13%), Gly (8.5%) and Arg (8.5%). It is immediately noticeable that serine is the most important amino acid for binding sericin to fibroin in view of H-bonding. This further evidences aforementioned hypotheses by Gray et al. [32] and Kundu et al. [33] who postulate upon the contribution of serine to stickiness between molecules. Their hypotheses single out serine as a predominant sticky amino acid since it is present in all regions of the sericin protein and constitutes 50% of the total amino acid sequence. Moreover, serine together with the other amino acids in Table 3 possesses specific properties which encourage the formation of H-bonds. These include: 1. A lack of stearic limitations (due to its simple structure). 2. An abundancy of hydrogen, oxygen and nitrogen atoms (since they participate in H-bonding).
4. Conclusions Sericin is a waste product from Bombyx mori silk cocoon purification, resulting in weight for weight between 20 and 30% of the original Table 3 Short and mid range H-bond connections in our model between amino acids of sericin and fibroin.
b3Å 3 − 4Å
Sericin
Fibroin
Ser Arg Ser Ser Ser Asn Asn Asn Asp Arg
Gly Ala Ser Ser Ala Ser Ser Ser Ala Gly
References [1] K. Yukuhiro, H. Sezutsu, N. Yonemura, Evolutionary Divergence of Lepidopteran and Trichopteran Fibroins, in: T. Asakura, T. Miller (Eds.), Biotechnology of Silk, Springer Science and Business Media, Dordrecht, 2014 (2014). [2] D. Shah, Natural fibre composites: Comprehensive Ashby-type materials selection charts, Mater. Des. 62 (2014) 21–31. [3] D. Shah, D. Porter, F. Vollrath, Opportunities for silk textiles in reinforced biocomposites: Studying through-thickness compaction behaviour, Compos. Part A-Appl. Sci. 62 (2014) 1–10. [4] R. Khare, S. Bose, Carbon nanotube based composite, J. Miner. Miner. Charact. Eng. 4 (1) (2005) 31–46. [5] J. Ayutsede, M. Gandhi, S. Sukigara, H. Ye, C. Hsu, Y. Gogotsi, F. Ko, Carbon Nanotube Reinforced Bombyx mori Silk Nanofibers by the Electrospinning Process, Biomacromolecules 7 (1) (2006) 208–214. [6] H. Kim, W. Park, Y. Kim, H. Jin, Silk fibroin films crystallized by multiwalled carbon nanotubes, Int. J. Mod. Phys. B 22 (2008) 1807–1812. [7] M. Kang, H. Jin, Electrically conducting electrospun silk membranes fabricated by adsorption of carbon nanotubes, Colloid Polym. Sci. 285 (10) (2007) 1163–1167. [8] H. Jin, J. Park, R. Valluzzi, P. Cebe, D. Kaplan, Biomaterial Films of Bombyx Mori Silk Fibroin with Poly(ethylene oxide), Biomacromolecules 5 (3) (2004) 711–717. [9] P. Alam, Protein unfolding versus beta-sheet separation in spider-silk nanocrystals, Adv. Nat. Sci.: Nanosci. Nanotechnol. 5 (1) (2014) 015015(6). [10] M. Pahlevan, M. Toivakka, P. Alam, Electrostatic charges instigate ‘concertina-like’ mechanisms of molecular toughening in MaSp1 (spider silk) proteins, Mater. Sci. Eng. C 41 (2014) 329–334. [11] E. Sintya, P. Alam, Self-assembled semi-crystallinity at parallel beta-sheet nanocrystal interfaces in clustered MaSp1 (spider silk) proteins, Mater. Sci. Eng. C 58 (2016) 366–371. [12] G. Freddi, M. Romano, M. Massafra, M. Tsukada, Silk fibroin/cellulose blend films: Preparation, structure, and physical properties, J. Appl. Polym. Sci. 56 (12) (1995) 1537–1545. [13] P. Alam, Fagerlund Pi, P. Hagerstrand, J. Toyryla, S. Amini, M. Tadayon, A. Miserez, V. Kumar, M. Pahlevan, M. Toivakka, L-lysine templated CaCO3 precipitated to flax develops flowery crystal structures that improve the mechanical properties of natural fibre reinforced composites, Compos. Part A-Appl. Sci. 75 (2015) 84–88. [14] X. Hu, D. Kaplan, P. Cebe, Effect of water on the thermal properties of silk fibroin, Thermochim. Acta 461 (1–2) (2007) 137–144. [15] K. Hu, Biocompatible fibroin blended films with recombinant human-like collagen for hepatic tissue engineering, J. Bioact. Compat. Polym. 21 (1) (2006) 23–37. [16] M.H. Wu, J.X. Yue, Y.Q. Zhang, Ultrafiltration recovery of sericin from the alkaline waste of silk floss processing and controlled enzymatic hydrolysis, J. Clean. Prod. 76 (2014) 154–160. [17] M. Ho, H. Wang, K. Lau, J. Lee, D. Hui, Interfacial bonding and degumming effects on silk fibre/polymer biocomposites, Compos. Part B 43 (7) (2012) 2801–2812. [18] G. Altman, F. Diaz, C. Jakuba, T. Calabro, R. Horan, J. Chen, et al., Silk-based biomaterials, Biomaterials 24 (3) (2003) 401–416. [19] A. MacIntosh, V. Kearns, A. Crawford, P. Hatton, Silk-based biomaterials for tissue engineering, Top. Tiss. Eng. 4 (2008). [20] M. Khan, M. Tsukada, Y. Gotoh, H. Morikawa, G. Freddi, H. Shiozaki, Physical properties and dyeability of silk fibers degummed with citric acid, Bioresour. Technol. 101 (21) (2010) 8439–8445. [21] Y.Q. Zhang, Applications of natural silk protein sericin in biomaterials, Biotechnol. Adv. 20 (2) (2002) 91–100. [22] P. Aramwait, T. Siritientong, T. Srichana, Potential applications of silk sericin, a natural protein from textile industry by-products, Waste Manag. Res. 30 (3) (2012) 217–224. [23] P. Jiang, H. Liu, C. Wang, L. Wu, J. Huang, C. Guo, Tensile behavior and morphology of differently degummed silkworm (Bombyx mori) cocoon silk fibres, Mater. Lett. 60 (7) (2006) 919–925.
220
A. Morin, P. Alam / Materials Science and Engineering C 65 (2016) 215–220
[24] J. Hardy, T. Scheibel, Silk-inspired polymers and proteins, Biochem. Soc. Trans. 37 (4) (2009) 677–681. [25] J. Huang, R. Valluzzi, E. Bini, B. Vernaglia, D. Kaplan, Cloning, expression, and assembly of sericin-like protein, J. Biol. Chem. 278 (46) (2003) 46117–46123. [26] J. Grant, B. Pickup, M. Sykes, C. Kitchen, A. Nicholls, A simple formula for dielectric polarisation energies: the Sheffield Solvation Model, Chem. Phys. Lett. 441 (1–3) (2007) 163–166. [27] D. Case, T. Cheatham, T. Darden, H. Gohlke, R. Luo, K. Merz, et al., The Amber biomolecular simulation programs, J. Comput. Chem. 26 (16) (2005) 1668–1688. [28] L. Yang, C. Tan, M. Hsieh, J. Wang, Y. Duan, P. Cieplak, et al., New-generation amber united-atom force field, J. Phys. Chem. B. 110 (26) (2006) 13166–13176. [29] C. Zhou, F. Confalonieri, M. Jacquet, R. Perasso, Z. Li, J. Janin, Silk fibroin: structural implications of a remarkable amino acid sequence, Protein 44 (2) (2001) 19–122 (Grant2007). [30] W. Humphrey, A. Dalke, K. Schulten, VMD - visual molecular dynamics, J. Mol. Graph. 14 (1996) 33–38. [31] S. Ling, Z. Qi, B. Watts, Z. Shao, X. Chen, Structural determination of protein-based polymer blends with a promising tool: combination of FTIR and STXM spectroscopic imaging, Phys. Chem. Chem. Phys. 16 (17) (2014) 7741.
[32] T. Gray, B. Matthews, Intrahelical hydrogen bonding of serine, threonine and cysteine residues within α-helices and its relevance to membrane-bound proteins, J. Mol. Biol. 175 (1) (1984) 75–81. [33] S. Kundu, B. Dash, R. Dash, D. Kaplan, Natural protective glue protein, sericin bioengineered by silkworms: potential for biomedical and biotechnological applications, Prog. Polym. Sci. 33 (10) (2008) 998–1012. [34] H.P. Zhao, X.Q. Feng, S.W. Yu, W.Z. Cui, F.Z. Zou, Mechanical properties of silkworm cocoons, Polymer 46 (2005) 9192–9201. [35] A. Motta, B. Barbato, C. Foss, P. Torricelli, C. Migliaresi, Stabilization of Bombyx mori silk fibroin/sericin films by crosslinking with PEG-DE 600 and genipin, J. Bioact. Compat. Polym. 26 (2) (2011) 130–143. [36] P. Poza, J. Perez-Rigueiro, M. Elices, J. Llorca, Fractographic analysis of silkworm and spider silk, Eng. Fract. Mech. 69 (2002) 1035–1048. [37] J. Perez-Rigueiro, M. Elices, J. Llorca, C. Viney, Effect of degumming on the tensile properties of silkworm (Bombyx mori) Silk Fiber, J. Appl. Polym. Sci. 84 (7) (2002) 1431–1437. [38] J. Reyes-Gasga, E.L. Martinex-Pineiro, G. Rodriguez-Alvarez, G.E. Tiznado-Orozco, R. Garcia-Garcia, E.F. Bres, XRD and FTIR crystallinity indices in sound hjuman tooth enamel and synthetic hydroxyapatite, Mater. Sci. Eng. C 33 (2013) 4568–4574.